DRUG TESTING ADVISORY BOARD

Scientific Meeting on:
Drug Testing of Alternative Specimens and Technologies
Day 1

April 28, 1997

Double Tree Hotel

1750 Rockville Pike

Rockville, Maryland

Proceedings By:

CASET Associates, Ltd.

10201 Lee Highway, Suite 160

Fairfax, Virginia 22030 703-352-0091

TABLE OF CONTENTS

Welcome - Paul Schwab 2

Federal Workplace Drug Testing - Dr. Joseph Autry 6

Forensic and Scientific Requirements for a Workplace 15 Drug Testing Program - Dr. Ed Cone

Overview of Alternative Specimens/Technologies 41 - Dr. Yale Caplan

Specimen Collection/Chain of Custody Panel 72 - Moderator: Dr. Robert Willette

Initial Test Reagents and Procedures Panel 134 - Moderator: Dr. Marilyn Huestis Confirmatory Test Procedures Panel 216 - Moderator: Dr. Richard Hilderbrand

DRUG TESTING ADVISORY BOARD

JOSEPH AUTRY, MD (Chair), CSAP, Rockville, MD

DONNA BUSH, PhD, (Exec. Secretary), CSAP, Rockville, MD

ROSEMARY BAKES-MARTIN, CDC, Chamblee, Georgia

YALE CAPLAN, PhD, CORNING National Center for Forensic Science, Baltimore, Maryland

CHRISTOPHER HOLLAND, MD, HRSA, Bethesda, MD

AARON JACOBS, PhD, USAMC, Fort Sam Houston, Texas

ALAN JONES, PhD, University of Mississippi, University, MS

TAI KWONG, PhD, University of Rochester, Rochester, NY

MELANIE MALLORY, Consultant, Scottsdale, Arizona

RICHARD PINDER, PhD, Public Health and Addiction Services, Hartford, CT DIANA WILKINS, PhD, University of Utah, Salt Lake City, Utah PARTICIPANTS

RICK ANDERSON, PhD, Biosite Diagnostics, San Diego, CA

DAVID ARMBRUSTER, PhD, PharmChem Labs, Fort Worth, TX

WERNER BAUMGARTNER, PhD, Psychemedics Corp, Culver City, CA

MICHAEL BAYLOR, PhD, Research Triangle Institute, Research Triangle Park, NC

THOMAS CAIRNS, PhD, DSc, Psychemedics Corp, Culver City, CA

PAULA CHILDS, PhD, DABFT, CompuChem Lab, Research Triangle Park, NC EDWARD CONE, PhD, NIDA, Baltimore, Maryland

ANTHONY COSTANTINO, PhD, DABFT, American Medical Lab, Inc, Chantilly, Virginia

KENNETH DAVIS, Research Triangle Institute, Research Triangle Park, NC DAVID EVANS, National On-Site Testing Assn, Flemington, NJ

ROBERT FOGERSON, PharmChem Labs, Inc, Menlo Park, California NEIL FORTNER, PharmChem Labs, Inc, Menlo Park, California ANN MARIE GORDON, Psychemedics Corporation Culver City, CA RICHARD HILDERBRAND, PhD, UCLA Olympic Lab, Los Angeles, CA MARILYN HUESTIS, PhD, NIDA, Baltimore, Maryland

DANIEL ISENSCHMID, PhD, Wayne Co Med. Examiner, Detroit MI REESE JONES, MD, University of California, San Francisco, CA DON KIPPENBERGER, PhD, Psychemedics Corp, Culver City, CA DONALD I MACDONALD, MD, Employee Hlth Programs, Bethesda, MD

JOHN MITCHEL, PhD, Research Triangle Institute, Research Triangle Park,

NC

SAM NIEDBALA, PhD, STC Technologies, Inc, Bethlehem, PA

MIKE PEAT, PhD, LabOne, Inc, Overland Park, Kansas

TIMOTHY ROHRIG, PhD, DABFT, Osborn Labs, Shawnee Mission, KS

DOUGLAS ROLLINS, MD, PhD, University of Utah, Salt Lake City, UT HANS SACHS, PhD, University of Munich, Munich, GERMANY

SAL SALAMONE, PhD, Roche Diagnostic Systems, Somerville, NJ

CARL SELAVKA, PhD, D-ABC, NYSDCJS, Albany, New York

MICHAEL SMITH, PhD, DABFT, Armed Forces Medical Examiner Office, Washington, DC

TAMARA ST CLAIRE, PhD, PharmChem Labs, Inc, Menlo Park, CA KENNETH STEINER, MD, Medical Review Officer, Woodbridge, NJ

JAMES TOWT, PhD, Roche Diagnostic Systems, Somerville, NJ J MICHAEL WALSH, PhD, The Walsh Group, Bethesda, Maryland

ROBERT WILLETTE, PhD, President, DuoResearch, Denver, CO

P R O C E E D I N G S (8:32 a.m.)

DR. BUSH: Good morning. My name is Donna Bush. I am with the Drug Testing Section, Division of Workplace Programs, Substance Abuse Mental Health Services Administration.

It is my pleasure to welcome you to the Washington, D.C. area, and to this meeting of the Drug Testing Advisory Board.

I would like to introduce Mr. Paul Schwab. Mr. Schwab is presently the deputy administrator of the Substance Abuse and Mental Health Services Administration of the Department of Health and Human Services.

In 1996 Dr. Schwab served as the acting director of the Health Resources and Services Administration, HRSA, the Bureau of Health Professions, having served as the bureau's deputy director since 1988.

In earlier years he was HRSA's associate administrator for policy coordination and was the executive secretary of several national public advisory committees, including the Council on Graduate Medical Education, the Graduate Medical Education National Advisory Committee, and the National Council on Health Planning.

He has authored a number of publications and is frequently consulted on the areas of health, work force analysis, and the application of continuing quality improvement principles in public administration.

A Phi Beta Kappa graduate of George Washington University, Mr. Schwab has also earned graduate degrees in economics and public administration from Indiana University and the Kennedy School at Harvard respectively.

He has received numerous honors and awards, including the Presidential, Distinguished and Meritorious executive rank awards, the HHS Secretary's award for exceptional achievement, and the Public Health Service Superior Service Award. Mr. Schwab.

Agenda Item: Welcome.

MR. SCHWAB: Thank you, Dr. Bush, for that introduction. I promise you I will try to keep my remarks briefer than the introduction this morning, but I appreciate it. I do want to welcome you all, and good morning, here.

The Administrator of the Substance Abuse and Mental Health Services Administration, Dr. Nelba Chavez, was not able to join with us this morning, but I do send her greetings.

SAMHSA, which is the agency in which our program here is housed, is not always a household word, to say the least, in terms of government agencies.

It is very unique as far as government health agencies are concerned in being really the only part of the federal government where substance abuse and mental health come together. Our agenda, working with the public, is quite broad.

In Dr. Chavez' absence, it is indeed my pleasure to welcome you today to this meeting of the Drug Testing Advisory Board, the scientific meeting on drug testing of alternative specimens and technologies.

We have invited you here today because of our commitment to making our work places safe from drugs.

The statistics clearly tell the story. Seven out of ten drug users hold down jobs. Economists say now that work place drug abuse costs us somewhere between $60 and $100 billion a year in lost productivity.

Whether it is hospital workers or truck drivers, mail carriers or mechanics, the results are always the same. They are always tragic.

That is why our federal work place drug testing programs are critical and why this meeting is so important.

Just a brief word on the board itself. The board was first chartered in 1990. It advises the Administrator of SAMHSA and recommends new areas for emphasis or de-emphasis, new or changed directions, and mechanisms or approaches for implementing recommendations.

As part of this function, the board periodically reviews scientific areas on new areas of abuse and the methods necessary to detect their presence.

We consider this meeting, and you, all helping us to maintain high standards for our federal work place drug testing programs.

That means we must enlist you as a resource as we develop our knowledge of the scientific advances. We regard this meeting as an open and objective forum for the presentation of new technologies, and focus on the science, and be as fair as possible in reviewing that science.

In that regard, we are indeed pleased that so many distinguished scientists have joined us today. There is a great deal of interest on our part in evolving drug testing technologies and alternative specimens, so that the determination of substance abuse can remain state of the science, and so that less invasive techniques or specimens can be used fairly, accurately, and reliably.

in closing, it is going to take all of us working together to get drugs out of the work place, and to send a clear message to American workers that if they use drugs, we will not hire them.

Now I would like to present Dr. Joe Autry, the director of the Division of Work Place programs, who will tell you more about this meeting. Thank you.

DR. BUSH: Thank you, Mr. Schwab. Now I would like to introduce Dr. Joseph Autry. For those of you who don't know, he is my boss.

Dr. Autry received his bachelor's degree from Rhodes University with majors in chemistry and psychology. He received his doctorate of medicine from the University of Tennessee.

He completed an internship in internal medicine at Baptist Memorial Hospital, where he ran a coronary care unit for several months. He completed his residency training in psychiatry at the

National Institute of Mental Health, St. Elizabeth's Hospital, model residency training program.

He was board certified by the American Board of Psychiatry and Neurology in 1976.

Dr. Autry had an active career as a practicing psychiatrist prior to coming to the Washington area, joining the National Institute of Mental Health and becoming the director of extramural research, taking on responsibility for basic research, clinical research, treatment research and applied research.

He took on the challenges of directing the Office of Policy Analysis and Coordination for the National Institute of Mental Health, and then for the Alcohol, Drug Abuse and Mental Health Administration, NIMH's parent organization, for NIMH.

In 1990, Dr. Autry decided to move back into research, and became the director of the Division of Applied Research at the National Institute on Drug Abuse, with responsibility for work on drug-free work place programs, work place policies related to substance abuse and AIDS in the work place, health services research, research on HIV infection and AIDS in the community.

He had oversight for the national laboratory certification program, which certifies laboratories to conduct drug testing for federal drug free work place programs and the federally regulated industries.

Following the reorganization of ADAMHA in 1992, Dr. Autry became director of the Division of Work Place Programs for SAMHSA, the Center of Substance Abuse Prevention, with oversight responsibility for all the federal drug free work place programs and the national lab

certification program for forensic urine drug testing. Additionally, Dr. Autry continues his part time private

practice of psychiatry, which he frequently describes as his link to the non-bureaucratic world, and an opportunity to constantly infuse his federal research and services responsibility with a continuing need for new knowledge to deal with the problems of mental illness and substance abuse in clinical practice. Dr. Autry.

Agenda Item: Federal Workplace Drug Testing.

DR. AUTRY: Actually, what I have said is that private

practice is frequently my link to sanity.

Let me just start by saying that it is real pleasure to see so many friends and colleagues and other interested parties at this Drug Testing Advisory Board.

As Paul told you, the board periodically reviews emerging drug testing technologies, alternative specimens and inclusion of additional drugs and drug classes, and has provided ongoing advice to the National Laboratory Certification Program and the federal drug free work program since 1990.

It is a key component in formulating or recommending any change in policy in these important areas for the federal government. This is the first time since Mike Walsh was a pup -- for

those of you who know Mike, that has been a while -- that the board has formally reviewed the principles and criteria necessary for any workplace drug testing program to be forensically defensible.

Then, it is to evaluate all the emerging and current technologies and alternative specimens, to see how they meet these principles and criteria.

If there are areas in which they are deficient, they are to make recommendations on how they can meet these criteria.

I want to emphasize for all of you that the focus of this meeting is on the science of testing technologies and alternative specimens.

It is not a meeting to advocate for any testing technology or alternative specimen.

Let me tell you what the meeting will not do. The meeting will not result in the recommendation of any emerging technology or alternative specimen for inclusion in the federal drug free workplace program, or the mandatory guidelines.

What it will do is to formalize the principles and criteria necessary for a forensically defensible work place program, and it will gather all the extant scientific data that is available for the current or emerging technologies and alternative specimens for future evaluation and recommendation.

What we will do today is we will start with an overview of the forensic and scientific requirements for a workplace drug testing program, followed by an overview of the alternative technologies and specimens that are currently available.

We have asked Dr. Ed Cone, who most of you know, from the National Institute on Drug Abuse, Addiction Research Center, and Dr. Yale Caplan of Quest Diagnostics, to present these two overviews.

As you can see from the program, each one of the essential areas for a successful workplace program will then begin with a brief overview of the principles and criteria necessary for that area, followed by a response from representatives of current or emerging technologies or alternative specimens.

These include urine specimen, hair testing, saliva testing, sweat testing and on-site testing.

The moderator of each session will give a brief overview.

The respondents were chosen by a scientist or other representative who works in the field of the testing technology or alternative specimen.

This method was chosen to eliminate any potential for conflict of interest or bias in selecting the respondents.

The coordinators who chose these respondents are as follows:

for urine testing, Walt Vogel who works with the National Laboratory Certification Program; for hair testing Don Kippenberger from Psychemedics; for saliva testing Sam Niedbala from STC Technologies; for

sweat testing Neil Fortner from PharmChem; and for on-site testing Dave Evans who works with the National On-Site Testing Association.

We are going to limit the overview remarks in each session to five minutes and each respondent will be limited to 15 minutes.

Each presenter has been asked to finish his or her sentence upon hearing the signal indicating that time is up, and not to go on beyond that sentence.

Hopefully we will allow additional time for discussion and comments at the end of each session.

Questions and answers are going to be limited to board members and presenters. If members of the audience have questions that they wish to address to presenters, please write them on the four-by-six cards which you will be provided and pass them up front. We will make sure that the board members get them for consideration.

We have set aside Wednesday morning, April 30, for presentations and comments from the public. You must sign up in advance, and I underscore that. You must sign up in advance to present or comment on Wednesday morning.

The order of presentation or comment will be determined by the order in which you sign up. The time that each presenter has will be determined by taking the total amount of time we have available and dividing by the number of people we have signed up. That will be announced to you either late Tuesday or first thing Wednesday morning.

Following clarification and discussion of additional issues by the board, the meeting will adjourn, and there will be a press conference at the end of the meeting for any interested members of the press.

The board will reconvene sometime in mid-summer to consider all the scientific information and deliberations, and to at that time make recommendations about the status of the existing or emerging technologies or alternative specimens.

Of course, we will make sure that all of you know well in advance when that meeting will be scheduled and where it will be scheduled.

If this approach is as successful as I hope it will be, we will schedule periodic updates and public meetings of the board to do this same thing again, so that we can constantly look at emerging technologies and alternative specimens and their applicability in the work force for the federal drug free workplace program.

With that, I am going to be quiet and let us turn to the focus of this meeting, and that is the science. Thank you.

DR. BUSH: Thank you, Dr. Autry. We now have two major presentations. They are going to be full of great science and great overviews for us.

I have a biosketch for each of these scientists, Dr. Ed.

Cone and Dr. Yale Caplan.

We try to make these bios brief. But you know, when people have a long and illustrious career in science, they are not brief, they have done so much.

I want to read Ed's bio to you. It is a bit lengthy, but bear with me, and consider all the information that you hear. I think that will better seat his presentation in his mind, the depth of experience that comes from it.

Edward J. Cone received his BS in chemistry from Mobile College and earned his PhD in organic chemistry from the University of Alabama.

From 1971 to 1972, he held a post-doc appointment in the department of chemistry, University of Kentucky, under the direction of Professor Ellis Brown.

His work at the University of Kentucky was in the area of tobacco chemistry.

In the summer of 1972, he joined the staff of the Addiction Research Center, National Institute on Drug Abuse, where he currently serves as chief of the chemistry and drug metabolism section.

Dr. Cone is a commissioned officer in the United States Public Health Service and has the rank of permanent director grade, CO-06.

He was appointed as a member of the Research Officers Group of the United States Public Health Service in 1989.

From 1977 to 1985, he served as a member of the adjunct faculty in the Division of Pharmaceutics and Pharmaceutical Analysis, School of Pharmacy, University of Kentucky.

Presently he is adjunct professor in the toxicology program, University of Maryland at Baltimore.

Dr. Cone has maintained his academic association, by training four candidates for their PhD degree in forensic toxicology, by presenting lectures to medical, pharmacy, pharmacology and toxicology students on the chemistry and toxicology of drugs of abuse, and by hosting graduate students -- post-doctoral fellows and visiting foreign fellows -- in his laboratory.

In addition, Dr. Cone's laboratory has served as a training station for three World Health Organization fellows from Thailand and Malaysia.

He was a member of the editorial boards of the Journal of Drug Metabolism and Kinetics, and Employment Testing, and is presently serving on the editorial board of the Journal of Analytical Toxicology.

At present, Dr. Cone is serving as vice chairman of the Institutional Review Board for the intramural research program, NIDA, in Baltimore.

His major research interests are in the disposition and metabolisms of drugs of abuse, and drug-induced behavioral effects.

Dr. Cone's early research led to the elucidation of two new metabolic pathways for the environmental exposure to drugs of abuse, and on saliva and hair testing as means of identifying drugs exposure has been recognized internationally.

Dr. Cone has presented his findings at numerous international meetings, conferences and workshops.

Currently he is conducting clinical research on the relationship of behavior and performance to specific levels of drugs in blood, saliva, sweat, urine, hair and skin.

Dr. Cone's research has resulted in the publication of numerous book chapters, and over 200 scientific articles on the analysis of drugs in biological media.

Dr. Cone was cited by Science Watch, a journal which tracks trends and performance in basic research, as being the most highly cited author in forensic science over the period 1981 to 1993.

Dr. Cone was awarded the Irving Sunshine Award in clinical toxicology, in recognition of pioneering work in clinical toxicology in 1995, at the International Association of Therapeutic Drug Monitoring.

Recently, Dr. Cone was recognized by the Commissioned Officers Association as the CIPAC Career Scientist of the Year Award in 1996.

This award was to acknowledge an exemplary career of pioneering research on the detection of psychoactive drugs in humans, and for the broad implication of discoveries to improve drug abuse, prevention and treatment. Dr. Cone.

(Applause.)

Agenda Item: Forensic and Scientific Requirements for a workplace Drug Testing Program.

DR. CONE: It is truly a real honor to be here. As Donna overstated many of the things that I have done, I have been involved in drug analysis for many years and have met many puzzled expressions on the clinical ward when I have walked down. They have looked at me and said, you want to test what?

Thank you, anyway, for being here, and welcome to this conference. It is an honor to be invited to present this opening talk.

We are all keenly aware of the many problems that drugs impose upon our society, by the epidemic of drug abuse that currently engulfs our nation.

The scourge of drug abuse robs us of many valuable lives, intellect and property that we can sorely afford to lose.

Among the many approaches that we can take to eliminate the problem, the application of our unique scientific talents is the one that we in this room can most contribute.

It is really my belief that solutions must first begin with acknowledgement of the problem and good tools. I think that drug testing technology offers the clinician the tools to work on the problem, recognition of whether a person is using drugs, is under the influence or has been exposed to drugs. I think that is what we are here to talk about.

It is my charge to review briefly the forensic and scientific requirements for a workplace drug testing program. That is a pretty big charge.

What I am going to try to do very quickly is go through and talk briefly about the alternate matrices and, as well, many of my comments will be applicable to on-site drug testing.

I am going to spend most of the time talking about, first, how forensic standards have been established through promulgation of the guidelines.

I will go through many of the specific requirements, now, that I think we have evolved for forensic requirements in drug testing in the work place.

We are going to talk about how you go about validating a specimen, an assay, and the need for quality assurance methods and quality control materials.

Finally, I am just going to spend a very brief amount of time talking about the maturation and how technology evolves, and how this technology affects public health.

I want to start with this quote from Walter Haines in 1904 who said that there is scarcely a part of the body that may not be examined with profit.

Clearly, what he means is that you can find out a lot of information from drug testing.

This is a slide that I have used many times now in talks where I have illustrated the luxury and richness of the availability of different biological fluids in tissues for drug testing.

Many of them we are going to be focusing on today are specific ones of sweat, hair, saliva, as compared to urine and on-site testing.

There are other possible tissues that you will be hearing about in the future we are hearing about. Commercial methods are already available for marconium.

We are still working on how to drug test for sneezes, but there are a number of things that we can get from alternate matrices.

This is just a quick sampling of some of the things. There is unique information, surely, in different biological matrices that you can't get from urine or blood.

There are other advantages that one could possibly get.

Some of the matrices will involve less invasive collection techniques. Hair sampling possibly will allow you to go back and get a

specific specimen within a few days that represents almost the identical specimen, something you can't do, of course, with liquid samples.

Some of the other advantages, quite often now in urine we test for inactive metabolites. In different alternate matrices you often find the parent drug.

It is much more satisfying when you are trying to interpret what is happening to a subject to know that you are testing actual active drug rather than an inactive metabolite.

Some of the matrices also may exhibit greater stability than urine, lower disease risk, easy collection, shipment and storage.

There are a list of things we can think about as we go through these days in terms of what the advantages are there.

There may be disadvantages as well. You have to consider the difference in matrix as you collect and begin to develop tests for different biological specimens.

You have to consider, of course, the stability of the analyte, what the analyte is, the stability, how you go about getting the analyte out in a pure form, what form of assay is going to be developed.

A very important function is how are you going to develop QC materials that go hand in hand with the development of the assay that are appropriate for that specific assay.

There are major differences. Urine, of course, is mainly an aqueous matrix. Saliva offers some advantage, in that it is somewhat easier to collect and a little bit less embarrassing to collect.

It is also an aqueous base, but it has other things like mucous, that also affects things like how easy it is to pipette a specimen. So, it is a little bit more difficult to measure saliva.

Sweat is an aqueous medium, but it is mainly collected by means of adsorption onto some sort of collection device.

You have to worry about how do you get the drug off the collection device. You have to worry about absorption.

You also have to worry, if you are using a collection device that applies to the skin, about environmental contamination, both before you put the specimen on, the patch on, and during the process of taking the collection device off the skin. Environmental contamination is certainly a possible risk.

Specific advantages, we have mentioned already a little bit about hair. It is a very different matrix. It is a dry, protein complex.

So, it presents its own problems in terms of how do you get the drug out of the complex as well. Of course, there are risks from environmental contamination.

There is the issue of what effect does melanin have, or what is the effect of melanin on the type of drug and the amount of drug that is present.

So, there are major differences in the matrices that we will be talking about throughout the conference. We want to look at them objectively, fairly, and consider what they offer in terms of new information, and advantages over existing technology.

Now, in terms of what are standards, we have seen over the last decade an incredible evolution toward where we are today in terms of forensic drug testing standards.

President Reagan issued his Executive Order in 1986, 12564, where he stated essentially that he wanted to achieve a drug free federal workplace.

That really set the standard of where he wanted us to be. Over the years we have certainly established some specific standards.

The HHS was authorized to promulgate scientific and technical guidelines for drug testing programs, and those standards have been published and updated.

They have been put into effect and really become the expected standard for forensic drug testing over the years.

Not only have federal agencies adopted these guidelines, but so have DOD, DOT, states, as well as private industry have come to expect the quality of drug testing that was established through this program.

This is a quote about where we have come to in terms of society's acceptance of drug testing. It simply says:

"Over the past decade, incumbent presidents and Congress have not only authorized drug testing by public and private employees, but have required or encouraged it in some work places...

"...the policies of previous administrations, the legislature and judicial systems have all contributed to establishing drug testing as a major component of the nation's `war on drugs'."

Obviously we have these standards. We have to remember in what context we are trying to apply them, in this specific instance the workplace setting.

It is a very unique setting. It is a setting where a single test result is generally the only piece of evidence involved in hiring, forced treatment, or firing of employees.

So, it is very important that we place the highest degree of expectation of accuracy on our drug testing processes.

The quality of the tests, we know, is only as good as the sum total of the entire process of collection, the testing process, reporting and review by medical review officers.

Now to those standards. There are a host of very specific standards laid out in the guidelines. I would like to start out by saying these are the guidelines for urine drug testing.

I fully understand that the matrix that we are talking about is different and the guidelines may not apply to the matrix that we are talking about in this conference as well as on-site testing, but they certainly provide a place to start.

They are the expected standards for forensic testing as it exists today.

We are going to go through very quickly and just talk about some of these different sections of forensic workplace drug testing and how they should apply.

When we are talking about drug testing, we first have to identify specifically, exactly what happens to the specimen during the entire course of collection, handling and storage of specimens.

That has to be clearly stated in a standard operating procedure document. This document is the legal description of what actually happens to the specimen. It has to be an accurate updated version which includes all processes in the receipt, collection, handling, testing and storage of the specimen.

It should describe in detail how a chain of custody is applied to these processes. It should describe how quality control and quality assurance practices are implemented at each step along the way.

It should describe the training and expectations of personnel. It should have considerations of what is needed in the facilities and how it is organized, and an entire detailed description of the other related issues like instrumentation and processes for reporting. This has to be a complete document.

There are considerations for how the test facility should be organized. In urine drug testing, there is the requirement that screening and confirmation be performed at the same site, that security be given the highest order of magnitude for consideration, that forensic work be generally separated from non-forensic work, and there be limited access to the drug testing areas, to specimens and to records, and there be a general infrastructure of support.

Chain of custody I couldn't say much about the importance of an adequate chain of custody. This is just a definition, one definition, which says:

Procedures to account for the integrity, identification and security of specimens by tracking their handling, storage from of collection, to the final disposition of the specimen.

This is the legal document that tracks and identifies how it was collected, when it was collected, how it was shipped to the laboratory, what the laboratory did to it and the final disposition.

It is an essential document that has to exist in the context of workplace drug testing.

The requirement, of course, for workplace drug testing is that each and every test can potentially result in litigation, and every step along the way has to be defensible.

The chain of custody covers collection, as I said, transport to the laboratory, what is happening in terms of accessioning and processing and testing throughout the laboratory, all the way through to the final reporting of the result.

Security at the collection site is also an important consideration, how do you collect the specimen, how do you identify the specimen, how do you get it into some form of container that can then be either tested or shipped to the laboratory.

These are all considerations that we are very familiar with,

but they have to be adapted to the new types of specimens.

In some cases they may be a challenge, but it requires that

we have specific methodologies for assuring that the specimen belongs to the subject or the employee, that it not be tampered with, that it not be adulterated, that there be no switching.

So, these are important procedures that we have to think about when we are designing collection processes.

The security at the test facility, as I said, is absolutely of the highest regard. Access must be restricted to authorized personnel.

Security procedures must be in place, both during and after working hours. There is the highest level of security accorded to both specimens and, frequently forgotten, those important records.

Security has to be considered when you are designing your computer systems, the laboratory information system, as well as documentation of the processes that are used in security.

Many laboratories are designing their facilities so that frequent tours -- there is always somebody wanting to take a tour of a laboratory -- so that those tours are performed so that they can be performed without having the people on the tour go into the test facility.

They are actually designing laboratories -- this has been done for a number of years now -- so that you can take the tour and walk around the outside of the test facility and look in through usually protected areas.

The test methods themselves are usually composed of a screening test, which has traditionally been immunoassay as a requirement for urine drug testing.

The screening test is followed by a confirmation test.

There is a requirement, of course, that the confirmation test be based on a different chemical principle or a different chromatographic principle than the initial screening test.

This has also been traditional GC/MS. There is no reason why it can't be LC/MS or some other acceptable substitute.

A few words about the screening test. You have to, when designing your initial testing procedure, take into consideration, of course, the differences in the matrix, how you are going to handle that matrix, how you are going to extract, if necessary, and how you are going to apply your assay to this rather unique, different biological specimen.

There is a need for at least some form of semi-quantitative response, and of course you have to define the operational parameters that are needed to assess the performance of the assay, sensitivity, specificity, accuracy, precision and so on.

You certainly have to identify the other materials that may cross react, potential interferences.

You have to make sure that there are available reference materials. The QC issue is a real problem sometimes with new matrices, and we will talk a little bit more about that.

A real trouble spot is establishing a new cut off for a biological matrix. This is probably one of the most challenging things, once you have developed an assay for a specimen, is to come up with the appropriate cut off concentration.

There are very good, solid methods, scientifically based methods, for determining what optimal cut offs are with the use of receiver operating curves, and the like.

Other considerations in the screening assay, you must think about what is going to happen after the screening is done, and consider the confirmation assay.

You need to approximate, to some degree, the analytical sensitivity of the confirmation method. You don't want an assay that is far more sensitive than your confirmation assay. You certainly don't want one that is less sensitive.

So, you must think about the development of the screening in the context of what the confirmation method is going to be.

There is certainly a need for sensitivity. Then specificity, we said this already but I mean this in a clinical sense, the clinical definition now of sensitivity, which means the ratio of true positives plus false negatives.

So, you want to maximize the identification of true positives and minimize false negatives, of course, with sensitivity. You can give up a little bit on specificity usually to

achieve this goal, where you maximize true negatives and minimize false positives.

There is some need for linearity around the cut offs so you can have good precision. Finally, of course -- we have said this already -- the consideration of what the cut off is going to be is going to be one of the major decisions that you are going to be facing.

The confirmation assay is also an issue where you have to consider -- again, even more important, you have to identify now a way to get the drug or the analyte out of the specimen, generally, if you are going to use the combined methodologies.

You have got to get it somehow into usually a liquid matrix.

So, you have got to consider, how do you get the analyte out of your unique matrix and into a form that can be analyzed by one of the combined technologies.

Now we insist upon quantitation and accurate quantitation, and of course the other performance parameters, including specificity, accuracy and precision.

You need to evaluate and make sure that there are no interferences, and define the LOD LOQ linearity and so forth.

Again, we get back to this issue of pharmacologically relevant cut off. How do we do that? That is usually somewhat problematic.

This is a slide where I have sort of tried to illustrate the overlap between the two major considerations in establishing a cut off concentration, the pharmacological issues versus the analytical issues.

They are totally overlapping in terms of considerations when you are trying to establish a cut off concentration.

You have to select a cut off or a concentration that can be detected by your assay. So, of course you certainly have to have the sensitivity from your requisite analytical assay. You also have to know what is there and how to detect it, and in what concentration in that particular biofluid are you going to find drug.

You also have to worry about setting the cut off concentration such that you can distinguish, we would hope, active use from potential environmental contamination.

You have to take into consideration other issues like, are there going to be major influences from route of administration or passive exposure and so forth.

So, what I have tried to do is illustrate this by throwing in an extra slide here. This is sort of like selecting a cut off where you have to have really sound scientific reasons for selection, when you grab the tiger by its tail. Otherwise, he might turn around and bite you.

What I tried to do here is illustrate that I think this company has done a pretty nice job of selecting a cut off that is well removed from the possibility of environmental contamination.

The noise from their assay system, passive exposure, contamination carry over and so forth, this looks like they have done a pretty nice job of establishing that cut off.

Another company didn't quite get the message. You can see we have a problem here. I would characterize this situation as the company being somewhat alligator challenged.

We have a lot of ground to cover and I am going to go through these quickly. Quality control we will spend an entire session on.

It is absolutely essential that the appropriate reference materials be available, and that you somehow devise a method whereby you can supply and utilize quality control materials that simulate or are very close to real clinical specimens.

That has been a real problem with many of our alternate matrices, is coming up with appropriate QC materials.

There is an entire science behind developing these materials. But it is a very important consideration if you are developing an assay for a new matrix or even an on-site test.

I would make a quick comment about on-site testing. There is quite often a QC check on the on-site test device.

It is my contention that most of these devices are not actual QC controls in the traditional sense of being a positive or a negative.

They are very good QC checks for the operation of the assay and the reagents. But they do not fit the traditional definition of a control.

So, in addition to those very nice checks that are built into on-site devices, you still need appropriate control materials to be run, both in an open and a blind manner.

You need to assess the performance of the assay and of course the accompanying quality assurance issues.

Reporting, the guidelines require that reporting results only go out to medical review officers. There are good reasons for that. We will talk a little bit about that in a minute, that prohibit telephonic reporting.

Again, security. The form of reporting must be secure. Certified copies are sent to the employers -- I should say MROs. The standard forms really simplify things as well.

Storage, you have to consider the requirement for not only protecting the specimens and the records with secured storage, but then they have to be retained in forensic testing.

The standard is for positive specimens for a year, and records for at least two years.

Personnel. Personnel are probably your most important aspect in laboratory operations. They have to be trained in forensic drug testing methods.

You have to have training documented for all your personnel.

As you go up the chain, the documentation needs increase for your supervisors and your certifying scientists, as well.

There must be very good personnel records throughout that document the training of your people in the laboratory.

The facility or the lab director is the ultimate person in charge, must be involved in day-to-day operations, and must have adequate training in forensic toxicology to serve in this role.

Confidentiality is an issue that we all worry about. We want to protect individual rights at the same time.

There are a number of ways to do this; by employee ideas instead of names, strict reporting rules, reporting security, and the use of MROs.

I want to say a word about MROs. They have really served an extremely important component of the urine drug testing program that is outlined by the HHS.

Their job is to review routine custody and control forms and determine whether the test result looks like a valid test result of not.

They can listen to the applicant or the employee who tested positive, and who may have a valid reason for testing positive.

They can intercede for those positives who are legitimate and make sure that they are not accused of being a drug user because of a legitimate excuse.

They have access to medical records. They can order retests, and they can protect the confidentiality of the medical records.

So, this is an extremely important function that the MRO serves. I can tell you that they have prevented several mistakes from happening over the years in the drug testing program. They have served a very important function in this role.

Finally, in the accreditation area, there is a need for consideration of accreditation, whereby an independent body reviews the drug testing.

They can utilize performance testing, inspections and remediation. Hopefully there are enough teeth in the program to allow for those labs that are not meeting standards to have their certification suspended or even revoked.

I am going to move and change my focus and go through quickly now the issue of how do you validate an assay.

I am going to turn from forensic requirements and talk quickly about what validity means.

I think one of the simplest and most elegant definitions of validity of testing has been defined by Chuck Oredetsky as the ability of an assay to detect a drug or its metabolites in biological fluids following human drug administration.

This is a very simple and elegant explanation, but it encompasses a variety of issues that one must consider, including the things that we said already -- sensitivity, specificity of the assays, the metabolic and pharmacologic variables such as dose, route of administration, concentration of the drug in the biological fluid, the pH and intersubject variability and absorption, metabolism and excretion.

Not only do you have to validate your analytical assay -- we spent enough time on that -- but you need to perform pharmacological assessments.

These are best done with clinical dosing studies. .You have to ultimately apply the assay in a real setting to determine how it works.

You need to define what detection times are and ultimately build a scientific base of what the testing positive means in terms of results.

We need that scientific base to be able to provide interpretation of positive and negative specimen results.

The validity has to go through a variety of components, of understanding what the pharmacology of the drug is, how it is distributed, and what is distributed into the different biological matrices, and most important, for interpretation, we need to understand time course and the relationship between concentration of drug in the matrix and relationship to dose.

We have to worry, of course, about the issue of multiple dosing and accumulation in tissues.

This is a slide that I put together that I think reflects my view now about how technology can evolve over time.

The development and improvement of a new technology does not stop when the test becomes available for use in the work place.

At this early stage, the scientific and legal challenges have just begun. There will be continual scrutiny by the forensic community of an assay's strengths and weaknesses over the entire lifetime of a product.

Legal challenges are likely to be made based on its use in the work place.

New applications are often discovered in products that were not imagined by the original designers. Important weaknesses may also be identified that require adjustments in product design or process design.

Ultimately, inferior products fall by the wayside, while good products get better.

The development of a new test, its commercial introduction, application and use by the public, and general acceptance by the forensic and legal communities can take a variety of courses.

Many new methods may never make it to the status of general use and acceptance. How a technology evolves is dependent on many factors. Frequently, the philosophy of a company that seeks to market a technology is an important factor in its evolutionary path. Scientific acceptance general begins with publication of studies in peer reviewed journals. These studies general establish the assay's performance and describe possible new applications.

As the number of supporting applications build, a product may attain a comfort level of acceptance within the scientific community.

In a few cases, FDA review and approval provides additional assurances of adequacy or comparability to existing products.

By their very nature, new products will face a tougher challenge in facing broad acceptance and use.

This is the final slide. It really reflects how I feel about where we stand and what we are trying to contribute to society. During the evolution of a work place drug testing product from design to marketing, developers should evaluate how their products are viewed and used.

Ethical companies strive to produce companies that employers and employees trust and rely upon to be fair and accurate.

At the same time, employees have the right to expect that their test results will be kept confidential, and that the intrusion into their privacy will be as minimal as possible.

Fairness and due process must be incorporated into the program, and there should be opportunities in which the accused can be evaluated to determine if they have legitimate explanations of test results.

Overall, the important goal of keeping the work place environment drug free and safe should be balanced against protecting basic individual rights.

The creation of accurate test methods can be an important contribution in advancing the public good and health of our people and our nation. Thank you for your attention.(Applause.)

DR. BUSH: Thank you very, very much, Dr. Cone. It is now my pleasure to introduce Dr. Yale Caplan. He received his bachelors degree in pharmacy and a PhD in medicinal chemistry from the University of Maryland at Baltimore.

After a year as a research associate and supervisor in experimental toxicology at Sinai Hospital in Baltimore, Dr. Caplan came to the Office of the Chief Medical Examiner as an assistant toxicologist, and served as the chief toxicologist and scientific director of the State of Maryland's alcohol testing program.

Dr. Caplan is now director of the National Center for Forensic Sciences and director of forensic toxicology for Quest Diagnostics, Incorporated.

Dr. Caplan has been very active in his profession. He is president of the American Board of Forensic Toxicology, of which he is a diplomate, director, and was secretary/treasurer.

He was the president of the American Academy of Forensic Sciences from 1987 to 1988, previously holding other offices, including chairman of the toxicology section.

He was the recipient of the Academy's 1989 Rolla N. Harger award for outstanding contributions to forensic toxicology, and the 1996 distinguished fellow award for a lifetime of service to the forensic sciences profession.

He was the president of Society of Forensic Toxicologists in 1981, and is a member, and was chairman, of the National Safety Council's Committee on Alcohol and Other Drugs.

He was chairman of the Maryland section of the American Chemical Society, and recipient of the 1994 Maryland Chemist Award.

Dr. Caplan is adjunct professor in the department of pharmaceutical sciences at the University of Maryland School of Pharmacy.

He has served as clinical professor in the university's school of medicine, director of forensic toxicology in the universitywide program in toxicology.

Dr. Caplan has contributed over 200 chapters, papers and abstracts to the scientific literature. He serves on the editorial boards of the Journal of Forensic Sciences, the Journal of Analytical Toxicology, and Forensic Science Review.

Dr. Caplan was a recipient of the 1986 distinguished service award of the Maryland State Alcohol Control Administration, and was a member of the Governor's Commission to study deaths resulting from building fires.

He is the director of national scientific services and serves as an advisor of attorneys regarding alcohol and drug testing issues, and has regularly testified as an expert witness in court.

He is a member of the Substance Abuse and Mental Health Services Administration Drug Testing Advisory Board, and an inspector and an instructor for the national lab certification program.

He has also served as a consultant for the Department of Transportation Office of Drug and Alcohol Policy and Compliance, and the National Transportation Safety Board. Without further ado, Dr. Caplan.

Agenda Item: Overview of Alternative Specimens/Technologies.

DR. CAPLAN: Thank you, Donna. She left out the one thing, that she was one of my students. I think she was afraid to admit that here.

It is a pleasure also to be here this morning. I understand why Dr. Cone wanted five more minutes. It is just so Donna can get through these introductions.

Here we are after a number of years with work place drug testing, asking many questions. I have been asked to sort of give an overview this morning about some of the various technologies and things you are going to hear in greater detail.

I am going to talk a bit about what the core subject for the several days will be, alternative technologies and some of the factors that are supporting just general information about what they are and some notations about them.

First, there are a few things to focus in on this conference. Dr. Cone asked you some of them.

We are going to be looking at, I think, research. Good researchers must doubt even the most widely accepted premises.

If we take this as a guiding light for the rest of the week, although I am going to show you some general information that might show some preconceived ideas, I think the idea here is to go through the next few days with an open mind.

Second, we have to transfer scientific information in a credible manner. The fact that there are various points of view, communication and the ability to both transmit and understand these scientific principles is somewhat paramount.

If we think these things -- and if you remember nothing else, there are these two cartoons and I have three at the end -- you can figure out what is in between.

In general, though, we need to look at a variety of things and Ed has already talked about a few of them.

Also, let me point out that all the slides I have are in the handout. Some of them I may not use in the interest of time, or go through them fairly quickly.

These are the specimens that we are here to talk about. I have blood on this slide. It is not one that we are going to be talking about, but it is sort of a point of comparison. So, there is some information on the handout, looking at some of the same characteristics for blood.

The characteristics that were looked at, or need to be looked at when you consider these things, include physiology, pharmacology, which drugs can be detected in the medium, the various collection methodologies, testing methodologies.

I am going to talk about what the current uses are, and how that may change, the accuracy of testing, scientific acceptability, litigation, risk, cost, advantages and disadvantages. So, within the handout, there is a characteristics slide on each one of these elements for each of the types of technologies, and I am going to be talking a little bit more about each of those.

There are some caveats. You have to have a disclaimer before you do any of this before such a wide group.

I am going to try to give you a very rapid overview of the state of testing. What you are going to hear and what is in the handout are some contemporary notations that are the result of a report that I, with Dr. Cone, prepared for HHS at the end of last year.

It has the same ideas coming along as to where these things might be going, what do we know today, what are the various thoughts.

So, the idea is that this is to serve as the basis for the meeting going forward. I am going to make the disclaimer that there are some statements in there that a lot of people may disagree or not agree upon, and others may agree very favorably with.

The idea is that they are presented as one liners, as thought processes, as thoughts that are held by some or many people in the relevant community. So, there are some caveats in what we do.

First of all, briefly in the way of background -- and Ed has mentioned a little bit about this -- there is some different physiology. The types of specimens and substances that we can deal with have some different characteristics.

We are well aware of urine, production by the kidneys, the volume and the fact that there is a stabilization of that and a validation following creatinine and the ability to eliminate creatinine. Hair, on the other hand, is a solid-type substance, has a fixed growth rate. It is going to be around for a much longer period of time. It has a crystalline structure. It can deteriorate, also, with age.

Blood, by way of comparison, is really the most dynamic of all the specimens. It is not used, or has not been used for a number of reasons.

If one wants to make assessments about intoxication, impairment, cause of death, blood is the fundamental specimen.

It is stable in the body. Its volume changes very slowly. Therefore, it gives us dynamic representation of what is going on, has a fairly fixed volume in the body.

Sweat is produced by eccrine glands. Its production, though, is highly dependent upon environmental conditions.

While it may follow or parallel a variety of drugs that are excreted, the ability to collect sweat has been a problem until recently.

It is mostly water. As you can see, the blood has got a lot of protein substances, and the urine has a lot of other substances, and is also mostly water, but sweat is almost 100 percent water.

The last of the ones that we are going to talk about -- saliva -- is secreted by various glands in the body.

Its flow is dependent upon neurotransmitter stimulation, and the flow does vary widely.

Saliva composition is dependent upon flow. If you want to come and pair off these specimens, saliva is one that might parallel blood in some regards and is the closest one that might be utilized for determination of impairment purposes, whereas each of the specimens -- I think that is what the purpose of the rest of the meeting will be -- may allow us to do something different, to ask and answer a different question.

In science, as you all know, it is not the information that you get, it is the question that you ask. If you ask the right question, then the information that the test can produce can be used validly.

If you ask the wrong question, it can be misinterpreted. So, in the end, asking the right question and getting an answer that is useful is the goal of what we are all going to try to do in the next few days.

I am going to now kind of quickly go through the various specimens or technologies with a bent on comparing them to each other along the lines that I mentioned earlier.

The drugs that can be tested in urine, we have extensive knowledge, as you all know, regulated tests. There are the five major analytes with the a couple of other drugs like 6-acetyl morphine.

Other drugs, the NRC and the military have included barbiturates, benzos, LSD from time to time. In the non-regulated arena, barbiturates, benzos, propoxyphene, methadone, methaqualone and LSD have also been included.

So, the drugs which are detectable have been clearly identified, and you will find that there are many similarities throughout the various matrices.

With laboratory based collection for urine, we rarely see observed collections. We are generally dealing with non-observed collections. It is the common practice.

There is a concern about limiting adulteration or substitution. We do collect about 60 milliliters of a specimen into a particular container.

We are using, as Ed described, custody control form documented processes, and we are commonly now today using split specimens.

I want to point out also, before I go further with this, that this might be a little more difficult to visualize here.

The best way to look at this might be as a matrix when you compare them. But it doesn't really fit too well on slides, to put multiple specimens across. So, I am kind of forced to do one at a time. With regard to the various types of testing on laboratory based urine, again, as you know, regulatory testing, screening by enzyme immunoassay followed by GC/MS confirmation is mandated.

Non-regulated laboratory testing, they follow the same methodology, or it may be different. A confirmation may or may not be included out of the regulated arena, and adulteration testing is possible.

There is much to be heard of concern with adulteration of specimens as we go forward with new technologies.

The current uses for laboratory based urine testing is widespread in the implementation and function of the current drug testing programs.

It is also used extensively, and has bene used for many years prior in treatment in correction programs.

The accuracy of this testing can be described as essentially, with the controls and situations with MROs and all that have been put in place, regulated testing is essentially 100 percent accurate.

We have extensive, in regulated tests, we have extensive PT programs available, multiple types, that have maintained a degree of credibility in maintaining the accuracy.

Non-regulated testing may be equally accurate. However, there is a choice as to whether to include various mandatory requirements or not.

Procedures utilized, screening without confirmation does occur. When in that arena, they are subject to some false positives.

So, we do have a type of testing with the current laboratory based urine which can be highly accurate if properly practiced.

It is practiced very widely inside and outside this arena, where you might have some diversion from that kind of activity. Scientific acceptability with regard to the current process,

we have a vast body of scientific literature addressing all aspects of urine testing.

I guess 10 years ago when a lot of these things started one would have said, urine is just not that important and we don't need to know that much about it.

Most of the previous forensic work was based on blood and tissues and other specimens.

I have to say, in the last 10 years we have done an awful lot of research. Every day, it seems like every couple of months a new technical scientific issue arises with regard to a specimen that maybe 10 or 15 years ago we would have almost discarded, as being relevant and having a lot of science surrounding it.

Properly conducted and evaluated urine tests are accepted and effectively utilized. That is the key underlying statement for what is going on this week.

Then an ultimate question is, are there other technologies or situations that can be equally accepted and equally effectively utilized.

This surrounds also the fact that we have passed the legal scrutiny with regard to urine testing, that has been generally upheld in various court cases, and urine testing has been documented.

In the many years that we have been doing it now, it has not been adversely affected.

In a summary of the urine specimen, here are some of the advantages to using laboratory based urine. Drugs and metabolites are highly concentrated. That is why we use the specimen in the first place.

There is a high propensity to find and detect these substances with conventional techniques. The drugs, particularly through their metabolites, are there.

The specimen can be obtained without physical risk to the donor. The main reason that blood is not widely utilized is the concerns for obtaining the blood specimen.

There is an extensive scientific basis now for the testing methodology. It has been utilized in many labs and documented extensively.

PT testing is liberally practiced. Results are frequently and regularly accepted in court.

As Ed had pointed out -- this is one of the over-riding things in looking at the other technologies -- we do have uniform testing criteria and well established cut offs.

We may want to change them for different reasons, but they have been well established and are documented.

We have methods which are of a commercial nature which are easily available so we can utilize these. We do have tests that are FDA approved as part of this process. There are many advantages to dealing with urine.

Some of the disadvantages, though, and things that might be of concern, is that the detection period is very short, two or three days.

Recent research shows that for some of the drugs that may be even less than we had originally anticipated, depending upon dose.

There are no dose concentration relationships. We can't say a great deal about exactly when the drug was used or what level of impairment there may be, if that is desirable.

Drug concentrations are greatly influenced by water intake.

The amount of water that an individual ingests, particularly for certain drug classes, will significantly affect the deductibility of those drugs.

In your laboratory based urine, we do have a relatively time consuming process, a laboratory based process with extensive laboratory costs.

You have to build a laboratory, you have to have a lot of personnel, and there is a fairly large cost involved with doing it.

Those are some of the advantages and disadvantages of doing the urine laboratory based.

As the first other modality we could look at is on site testing. We have a number of drugs which are fairly similar.

Most of the common analytes are also available and we are reasonably certain that, if there is a need, other ones could become available.

This methodology is now single drug kits for single drugs one at a time, multiple kits and various combinations thereof.

Some of the kits, because of the clinical application, do include one drug class anyway, tricyclates, which are not generally included in abused drugs.

This technology includes all the drugs and can probably be easily modified, because you are going to end up using the same antibodies and things that you use for other laboratory based urine testing, and just put them in kit form. So, the drugs would be similar.

Collection methodology would also be similar. The only difference here is that we would need to consider the acceptability of having results readily available to the collector.

In fact, if that is deemed acceptable, then there is probably going to be no difference. That will happen. We know that if we do an on-site test, that result is going to be immediately available.

There are kits which actually use the collection and testing devices simultaneously. So, there are a lot of novel items in the technology which could be very useful with on-site technology.

Right now split specimens are not common. Again, that could be altered very readily if necessary.

So, we have a collection methodology which is similar to the laboratory based, with the exception of the identification of the various subjects to the collector.

The methodologies are, again, basically immunoassays. Some have -- and I think Ed mentioned this -- built in controls.

We do have to consider, as we go forward, how you would really quality control such a program. There is some controversy, I think, about the built-in controls, actually only monitoring the reagents as opposed to monitoring the process.

There is minimum space required, though, compared to a laboratory.

Some of these processes are temperature dependent. While you use these kits, the temperature in which they are used could affect the timing and the outcome.

At least right now adulteration testing, which would still be difficult, in the laboratory based there can be other things done, other tests with the specimen when it gets to the laboratory.

The kits themselves, at least the way they are currently configured, do not lend themselves toward any adulteration testing and we will have the same types of specimens, whether it is laboratory based or on site, and the same propensity for alteration or adulteration of the specimen. So, that would be an issue.

Current use, fairly limited for work place purposes, although there are some mostly used in treatment and correction programs, and certainly used extensively in hospital emergency rooms and other arenas outside the work place arena.

Accuracy, some comments on accuracy. HHS regulated and common non-regulated analytes are all available.

Adequate procedures are not developed to ensure reliability. So, right now we do have kits out there. While most of them are FDA approved, although some in the forensic arena might not require FDA approval, there is some lack of uniformity amongst the products.

There is a concern, I think, about the end point determination being subjective in nature. When you read these things, I think all the kits except one read a positive by the absence of color. Color goes away.

When using kits, one has to be concerned about visual acuity, color blindness. The kits that are out there today have a lot of variability in their end points.

It is not that this isn't correctable. I think this is due, at least at this time principally, to the fact that we have a situation in which there are not rules.

When you make rules, manufacturers are able to adapt better. Right now there is variation in the detection end points and there is a lot of variation in cut off concentrations and how those concentrations actually operate.

So, we have a group of devices out there which we will hear a lot more about, which require principally more individual concern as to the assessment of the end result, and some variability as to the concentrations that are detected.

With on-site testing and scientific acceptability, they have not generally been used for forensic purposes. They are not supported at this time by any real court experience, and there is a relatively limited body of scientific literature.

That is not to say that there is not a large body of scientific literature about urine in general, but about the actual application or applicability of the various test kits to these processes, there is limited publication.

There is obviously some publication on the initial evaluation of the kits in comparison to their deductibility in a patient population, but that is not too extensive at this time.

The summary there, the advantages of on-site urine are fairly similar in many respects to urine in general.

Laboratory based, drugs and metabolites are highly concentrated, specimens can be obtained without physical risk.

The testing can be performed with non-scientific personnel.

That would be an additional advantage to this technology.

Minimal space and resources are required. That would be another pretty distinct advantage and we don't have to build a laboratory to do these things.

The other major advantage are the rapid turn around time of the test results, which could be desirable in any number of settings. Again, most all of these kits, except the ones that are only in a certain forensic arena, are approved by the FDA.

So, we do have a number of modalities which would make this very attractive.Some of the disadvantages of on-site urine, again, are similar to urine in general -- the detection period, the dose concentration relationship, the water relationship.

The other major ones on the bottom, the presumptive results may be acted upon inappropriately. So if one is to use this in a regulated structured arena, there have to be means to ensure that presumptive results are dealt with appropriately.

There is a possibility to compromise objectivity, from the collection process, since the individual doing the test may have access to the result.

The scientific basis, it is not as definitively established; that is to say, the actual use of these kits around cut offs, et cetera.

It has not generally been accepted yet for forensic purposes. Uniform testing criteria cut offs have not been established, although there are a lot of claims that there are various cut offs, that they meet certain criteria. Again, these have not been uniformly decided upon and tested.

There is, essentially, no performance testing program available, except that which a particular manufacturer might offer. There is no centralized performance testing program in on-site testing.

The third substance which can be looked at for acceptability and the possibility of testing for drugs is hair.

Again, a wide variety of drugs have been detected in hair. All the common drugs of abuse, including heroin, 6-acetyl morphine, actually a preponderance of the ability to find the parent drug in hair has been demonstrated.

Other drugs, metals, over the time period have also been detected.

So, the methodology and the various drugs have been established and shown to be present in hair. So, we can find the same drugs we are interested in, in different varying amounts.

Collection methodology as compared to urine would be substantially different. We are cutting hair usually at the scalp surface, at the vertex at the back of the head, and you can measure and document the length from the root to the tip out.

The hair can be stored in protective containers. It can readily be shipped to laboratories. Hair from other parts of the body can also be utilized.

We have a large source in the human body of hair. We have the ability to detect the drugs. We have some ease in getting the specimens from the collection sites to various laboratories for testing.

The methodologies, some comments here. The parent drug is often present, which is pretty different from urine situations.

Few laboratories are currently available that routinely test hair. Again, if there is an acceptance of this technology this could rapidly change.

Specially adapted amino acids are currently used for screening, GC/MS for confirmation.

One of the major points, I think, for hair testing is that it is laboratory based and we may need more sophisticated equipment to really go to appropriate sensitivity levels, as has been demonstrated by the need for tandem mass spec in order to really be able to appropriately detect cannabinoids.

So, the conventional laboratory that is doing urine testing would have to, to be doing hair testing, would have to be a little bit more sophisticated and some more costly equipment involved.

The methodology is essentially there and I think it has been extensively looked at and adapted in the arenas that are practicing it.

Current use, a lot in the gaming industry, mostly for pre-employment testing. There has been limited use in the work place, although some.

There has been some utilization in other investigations, particularly death investigations where other specimens are not available and prior drug exposure is an issue. It has been used in other forensic investigations. There are possibilities for use in various arenas with hair.

Accuracy, with regard to hair, there are currently no proficiency testing programs that exist. There have been some attempts at round robin testing with the limited laboratories. I think there are some external services that provide a type of specimens to laboratories performing this.

So, there is an attempt at that process, but there are no structured proficiency programs at this time.

There are some variations in analytical results. The matrix itself is not uniform. One could argue there is a benefit in an attempt to measure time with long enough hair.

On the other hand, if you take a random sample or a sample from a fixed portion of the hair or fixed length, as compared to other places, the matrix is somewhat different, the deposition may be different. There is a lack of clear uniformity there.

There have been some NIST surveys -- this round robin -- where some false negatives and false positives have been reported.

While this technology is clearly -- while it clearly has possibilities, there is clearly some concern from this point of view.

Scientific acceptability, there are some controversial aspects not resolved. How the drugs got into hair is an issue. Interpretation, dose, time relationships not established, ethnic bias has been of concern. The feasibility of marijuana testing is somewhat uncertain depending upon laboratory capability.

The advantages of hair, longer estimate of time of drug use.

That is a distinct advantage. Ease of obtaining, storing and shipping of specimen, low risk of disease transmission in the handling of samples. In the second specimen, one can go back to the hair and get another specimen and test it.

Disadvantages of hair, inability to detect recent drug use. It takes a while for the drugs to go into the hair. So, if it is very recent use, that is an issue.

The bias of color, ethnic origin and sex needs to be resolved or dealt with. The limited number of laboratories able to do this service is a disadvantage at this time.

Also, there is the possibility of an environmental contamination for some drug biases, lack of PT, and these procedures are not FDA approved.

I have in the handout -- I am going to skip through this.

This is just for comparison, some points about blood.

Blood is not really one of the things we are going to be looking at. So, I am going to skip through these. I put them in there mostly because they were in the report and they would allow you to compare some of the specimens to a dynamic specimen which has been widely utilized in the scientific community for therapeutic drug monitoring and for post mortem assessments.

In quick summary, some of the advantages are that you can interpret these behavioral relationships that many of the other specimens don't allow. There is a large data base of information.

The major disadvantage is that there aren't set cut off concentrations, and the concentrations of the analyte are generally much lower and the time of detection is also much lower.

Let's look at the last two, sweat and saliva. There is a little bit less to say about them from the data base.

Again, we can identify, in sweat, many of the same drugs.

In addition, alcohol is going to be excreted in the sweat. If you collect it for drug testing purposes, it is obviously going to evaporate.

The collection methodology, we have the sweat patch which is now available. So, there is now a technology to actually collect this in a documented fashion. Before it was much more difficult.

This can be worn for days to weeks, and it can be readily sent to laboratories.

The testing has been limited but essentially follows the same conventional immunoassay and GC/MS methods, and we do have the capacity to find the parent drug here.

The current use is mostly in the criminal justice system. There is interest in the use of this in the general treatment community.

There are no PT programs. The sweat patch is approved by the FDA and there has been a limited scientific evaluation and some publications on that.

The advantages are that there seems to be high subject acceptability to wearing this patch and a low incidence of allergic reactions.

So, there is the ability to monitor drug intake for a period of weeks. It is a relatively tamper proof and FDA approved product.

In the arena that this exists, there are some good advantages to possibly using it.

Disadvantages, variation in sweat production. There are a limited number of collection methods available. There is still high inter-subject variability. Risk of removal, risk of contamination are issues that have to be dealt with if we are going to use this for other purposes.

Saliva, which is the final specimen, has again -- we can detect all the drugs. You can also detect alcohol.

There are some commercial devices available for collecting saliva. The parent drug is frequently present. The immunoassay methods generally need to be targeted to parent drug.

This is another issue following one of the things Ed said, how the assays are set up. If we are looking more at parent drug or the metabolite, that may affect the nature and structure of some of the assays.

Their reactivity, conventional confirmatory procedures can be utilized, and we are looking at generally lower cut off concentrations.

Saliva is not widely used. It is limited to the private sector for insurance purposes. There is a device for alcohol. No PT programs.

There have been a fair amount of things written about the scientific acceptability of saliva, but mostly limited to purposes other than the work place.

The advantages, readily accessible for collection, may be related to behavioral performance. This is probably the only other specimen that really has this possibility.

So, if we are really looking for recent drug use, it has some major advantages there.

Disadvantages, possibility of contamination by oral smoke, internasal routes. The same disease collection concerns we have for other specimens, not FDA approved.

In the way of a summary here, urine laboratory based, we detect metabolites. We are looking at a two to three day detection window. We have laboratory turn around times to deal with. We are concerned about adulteration.

On-site testing, also detects metabolites, same detection window. The major difference there is we have a rapid turn around time.

For those industries for which that is useful, that might be valid. Also, adulteration concerns.

Hair, detects parent drug, long detection window, generally months. We do still have a laboratory turn around time and we still do have some concerns about environmental contamination.

Saliva detects the parent drug. There is a short detection window, less than a day. But it has a rapid turn around time, on site.

Contamination is possible. You can coincidentally do alcohol on the saliva sample, if that is otherwise warranted.

Sweat looks at both parent drug and metabolites. There is a long detection window in the way of weeks. It is a lab turn around time and there are concerns about variation in sweat production.

These are the other three slides. This is where we are probably today, at the beginning of this conference. We are down in the pits and we are going to look at everything that is going to happen.

Hopefully we will go up the ladder to come to some significant resolutions.

This is a slide I have never had a chance to use before, but this did seem appropriate. It says up here, this raging war with spears over there on the left. I don't have time to talk to some salesman. He has got a battle to fight.

Of course, I guess this guy has got the first model of the old gatling gun here. That would be very significant, if we had time to stop and think about using it appropriately, and I think that suggests this.

Then finally, we have to make decisions. In reviewing the data, I think we need to look at it objectively, and then ultimately decide which way we want to go, for those of us who are able to do those things.

Finally, the common parameters that we will be needing to develop as we go through the rest of the week is to look at scientific acceptability, court legal acceptability, community acceptability, whether these are FDA approved or that is necessary, establishing cut offs, quality assurance and PT and the cost/benefit to the various technologies.

Hopefully, in an abbreviated amount of time, I gave you a rough overview or at least my view of where some of the technologies are today and where we may be going with them.

With that, I will conclude and Donna won't have to put her second note up.

(Applause.)

DR. BUSH: Thank you very much. What a great kick off to this meeting. I am very, very, very pleased with your attention on a Monday morning and the number of you who showed up to join us for such a great kick off.

Now, let me do some housekeeping things. There are a lot of people here, in case you didn't notice. Everyone is going to want to get up and rush for the coffee.

If you want to reclaim your seat, you may choose to leave your handout book there. If you choose to do that, put your name on it. Okay, it is Monday, we have to start slow. The coffee break will be out the back door of this conference room. (Further housekeeping matters discussed.)

On that note, it is about five after 10. We will take a 15minute break. We will convene at 10:20. We are going to start on time because we have so much to cover. Please join us back. Thank you.

(Brief recess.)

DR. BUSH: It is time to get started. That is the Monday morning crowd. They got up, got a cup of coffee and walked around, and now they are very active.

In order to kill a little bit of time while everybody is taking their seat again and quieting down, I would like to take the opportunity to introduce myself to those of you who don't know me.

I know there are many friendly faces out there, Dr. Rohrig. Anyway, I would just like to let you know how I fit into things, where I am and what I do for a living.

Since December of 1989, I have been the chief of the drug testing section in the Division of Workplace Programs at the Center of Substance Abuse Prevention at SAMHSA.

I am responsible for planning and directing the National Lab Certification Program administered by the U.S. Department of Health and Human Services.

This certification program was established to ensure forensically and technically sound drug testing of federal employees and employees of federally regulated industries, such as the Department of Transportation and the Nuclear Regulatory Commission.

This effort directly supports the achievement of a drug free federal workplace as ordered by the President, and covers more than 10 million employees.

I am executive secretary of the Drug Testing Advisory Board, chartered to provide guidance to the Secretary of the Department concerning drug testing technology and policy implications.

In my other life, I also serve as faculty for the American Society of Addiction Medicine and the American Association of Medical Review Officers, to train medical review officers in the review and interpretation of urine drug test results.

Additionally, I serve as faculty for the Florida School of Addiction Studies and the Southeast School of Alcohol and other Drug Studies.

Prior to this job, I served with the Department of Defense as Deputy Chief and Technical Director of the U.S. Army Forensic Toxicology Drug Testing Laboratory at Fort Meade, Maryland.

I provided expert witness testimony and advice on drug testing issues for all military agencies.

I received a bachelors of science in chemistry from Loyola College in Baltimore, a masters of science and medicinal chemistry from the University of Maryland at Baltimore, a master of science in toxicology from the Johns Hopkins University School of Hygiene and Public Health, and a doctorate of philosophy degree in forensic toxicology from the University of Maryland at Baltimore School of Medicine, where Dr. Caplan had a very easy time with me as his graduate student.

I am an active member of the Society of Forensic Toxicologists, I am a diplomate of the American Board of Forensic Toxicologists, a fellow of the American Academy of Forensic Sciences, and I have been nominated for inclusion in Who's Who in Science and Technology.

(Applause.)

Now it is my pleasure to introduce Dr. Robert Willette. Dr. Willette has formed DuoResearch in 1981 for the primary purpose of providing consulting services to employers in the public and private sectors, on the design and implementation of work place alcohol and drug programs.

Dr. Willette has helped to establish policies, procedures, laboratory inspections and blind quality control programs for the U.S. Navy, Administrative Office of the U.S. Courts, the Federal Bureau of Prisons, the Drug Enforcement Administration, the Federal Bureau of Investigation and numerous private companies.

As chief of the Research Technology Branch of NIDA, Dr. Willette was responsible for overseeing the federal proficiency testing programs for drug testing laboratories conducted by the Centers for Disease Control from 1973 to 1981.

At NIDA, Dr. Willette was also responsible for multiple studies on the disposition of drugs in the body and their detection in biological specimens.

In addition to these specific efforts directed toward quality control programs, Dr. Willette is serving, or has served as a consultant on drug testing programs to the White House Office of Drug Abuse Policy, the U.S. Navy, the U.S. Army, NIDA, Federal Aviation Administration, Federal Railroad Administration, National Highway Traffic Safety Administration, the Postal Service, the DEA, FBI, Customs Service; in other words, pretty much most of the federal agencies.

This included the development and administration of the National Lab Certification Program for HHS, for which he currently serves as a certified inspector.

Dr. Willette received a bachelor of science in pharmacy from Ferris Institute, now Ferris State University, and a doctor of philosophy degree in medicinal chemistry from the University of Minnesota.

He has held research appointments at Upjohn, University of Michigan, Australian National University, and the Commonwealth and Scientific Research Organization of Australia.

He has served on faculties in schools of pharmacy at Ferris State and the University of Connecticut.

Dr. Willette, I will turn this panel over to you and adjourn to the back, and good luck to you, sir.

Agenda Item: Specimen Collection/Chain of Custody Panel.

DR. WILLETTE: Thank you very much, Donna.

It is appropriate that the first panel address the first step and subsequent steps in handling the specimens that are going to be tested.

Prior to 1981, most drug testing was done in the context of drug treatment programs, whether it was in the work place, the private sector, criminal justice or the military.

The process that was used to collect specimens and to handle specimens and throughout the entire process and the data followed fairly traditional clinical chemistry practices.

Because there was not strict control over the identification of who provided the specimen or the handling of the specimen in the laboratory with literally hundreds of people that had access to them, many errors occurred. Many people received false laboratory results.

The consequences of that were minimized by the fact that generally for the purposes that the testing was being conducted, that severe consequences didn't occur.

Usually people were either extended in the treatment and so forth.

That all changed in 1981 when two events happened in sort of conflicting motion. The Department of Health and Human Services got out of the laboratory proficiency and laboratory inspection and monitoring business, but the military got into it.

The military embarked upon a massive random testing program of military personnel. This prompted the necessity to develop a totally new system for conducting drug tests.

This was to merge together the high volume clinical practices with forensic principles. That emerged slowly during the early 1980s, spreading into other federal agencies, into the private sector, and then culminating in the development and implementation of the federal mandatory guidelines for federal work place programs in 1988.

Those looked at all of these critical steps which Dr. Cone has outlined.

It seems only appropriate to then bring together the people that have played critical roles in this. Some of you may notice uniforms in the audience.

There are at least eight people who are presenters over these days from that military background. They played a critical role in the evolution of this whole system.

The panelists that we have to start off with on collection and chain of custody represent views from each of the major subject areas that are the focus of this three-day hearing.

I have been told by the organizers that the order in which these areas are covered have been randomly selected throughout each of the different panels.

Just to prove that the lottery works, it seems most appropriate that the system selected our first speaker.

Dr. Michael Walsh was, in one of his many roles in the government, the person at the point at the lead and responsible for the development of, and the implementation of the mandatory guidelines and the national certification program.

Dr. Walsh is currently the president of the Walsh group, which is a research and consulting firm located in Bethesda, with the focus on providing consulting services and substance abuse policy research in this field, and certainly the technology.

From 1989 to 1993, Dr. Walsh served as the executive director of the President's Drug Advisory Council in the Executive Office of the President.

He also acted as the associate director of the Office of the National Drug Policy Control.

Prior to those appointments, Mike was the director of the Division of Applied Research at the National Institute on Drug Abuse, and served in many capacities, overseeing the development of drug policy, drug effects on performance and driving, drug testing programs, employee assistance programs and in AIDS treatment research.

Prior to joining NIDA, Dr. Walsh was a scientist, a civilian scientist with the Naval Medical Research Institute in Bethesda, where he conducted and oversaw research in various aspects of the effects of drugs on human performance, the areas of psychopharmacology, behavioral toxicology and hyperbaric medicine.

When Mike came to NIDA, he went from studying the effects of drugs on diving to studying the effects of drugs on driving. So, all we had to do is stick an R in that title.

Dr. Walsh received his doctoral degree in psychology here in town, the American University, in 1973. He is licensed to practice in the state of Maryland.

He has published more than 60 articles, books and chapters in various research publications, and has been in demand as an invited speaker at many, many conferences and meetings.

With no further ado, I am going to turn this over to Michael and start his timer.

DR. WALSH: Thanks very much, Bob. Bob and I worked together on the drugs and driving program at NIDA, and co-edited the WHO document.

Although they translated it into Spanish and French and German and I think Russian, they played it as if it was a new publication every year. So, we had about seven or eight citations from that one publication.

They haven't done anything since, in a long time. It is about time to do it again.

I would like to congratulate Dr. Autry and the staff of the Division of Workplace Programs and the Drug Testing Advisory Board for their leadership in convening this conference to examine alternative specimens and technology.

I would like to just say a few personal words about Joe Autry. As Joe's predecessor, I became acutely aware of the fact that no matter how hard you work, how professional you are, you still can't please everybody all the time.

As a result of this, there are times when it gets real personal in terms of some of the criticisms of the way in which you are running the program.

When I took the bull's eye off my chest and gave it to Joe, I was delighted to get rid of it, but I think that Joe deserves a lot of credit for wearing that bull's eye with pride and a lot of professionalism.

The mandatory guidelines for the federal workplace program were originally drafted in 1986. While science was the guiding principle, or the basic philosophy by which we developed the guidelines, much of the details were driven by legislative mandates and legal concerns of that time, of the middle 1980s.

Nearly 10 years have passed now since the final implementation. I really believe that federal scientists and policy makers now need to be open to new technology and new methods.

I think a desirable goal would be to incorporate the new advances in technology, to make drug testing programs more efficient and more cost effective without sacrificing any scientific integrity.

With regard to the on site technologies, I believe there is a need in many work place settings to have the capability for rapid screening -- fitness for duty, pre-performance testing, safety sensitive, post-accident, incidence and so on -- where negative specimens could be rapidly separated from those needing further laboratory evaluation.

I believe that on-site technology has reached the point where it is feasible to produce accurate and reliable results, and that procedures can be established to utilize this technology within a forensic program.

Over the next two days, the speakers representing the on-site technologies will attest to the fact that on-site urinalysis procedures, the methods, the reagents, internal and external quality control, confirmation procedures, and reporting of test results, really vary little from that which is currently mandated in the federal guidelines.

For example, with regard to specimen collections, maintaining the integrity of the specimen at the collection site, specimen chain of custody and control, we really believe that standard procedures as described in the HHS guidelines can and should be followed, whether the test is conducted on site or sent on to a lab.

Now, Dr. Vogel said all of that was nice but he wasn't going to let me get away with just saying we ought to use the HHS guidelines. He asked me to be a little bit more specific. Preparation,

I think, and set up of a collection site is a key to minimizing the problems in specimen collection.

These procedures are really no different whether the specimen is analyzed on site or sent to a lab. You have got to limit access to water and other potential dilutants.

You have got to put some dye in the toilets to inhibit adulteration. You need to check photo ID of the donor.

You need to have the individual remove excess clothing. It is good to provide some kind of security for the contents of pockets or a purse.

The temperature of the specimen has to be assessed, and the collector is going to have to check the color and the order at an absolute minimum of the specimen, to deal with the issues that Dr. Caplan so clearly laid out.

Properly trained collection personnel is also key. Training for collectors should include education about site preparation, about the responsibilities of a collection site person, ensuring how to ensure specimen security and chain of possession.

Collectors need to be trained about specimen substitution issues, dilution and adulteration, laboratory and medical review officer procedures, and proper completion of the chain of custody and control forms.

In addition, there are additional responsibilities if the collector is going to be the on-site testing person. They have got to be familiar with the package insert information for any device that is going to be used on site.

They should be required to at least have observed a manufacturers training video. They should be required to perform a number of tests and interpret the test results under supervision.

Some of the clients we have dealt with actually have a second person interpreting the results and signing off on the chain of custody, so that the one single individual cannot be so easily compromised.

If you are going to be an on-site testing person, you need to be trained to develop a standard operating procedure and so on.

In terms of the actual collection, it may differ, depending on the type of device to be used. Some devices allow for testing within the collection cup, while others require an aliquot for on-site testing.

For devices which permit testing within the cup, without introducing any kind of a foreign body into the urine specimen, there aren't any really special procedures that are required.

A standard chain of custody can be completed while the test is running. When the test is completed, the cup can be sealed with tamper evidence seals, and the donor dismissed with his or her copy of the form.

The test results can then be read, and with a positive, the sealed specimen should be shipped off to the laboratory. Negatives should be disposed of and results reported out.

For devices which require pipetting of specimen or introducing something into the specimen, I would recommend that an aliquot be poured off and the remainder of the specimen sealed for shipment, and the chain of custody initiated.

Typically in a lot of cases people use a disposable cup anyway, and then fill two smaller bottles to be shipped to the lab.

I would suggest that all of that be done prior to the specimen being analyzed. If it is a positive on site, the specimen to be sent to the lab are already sealed and ready to go.

In terms of maintaining chain of custody and control, I believe the chain of custody should be initiated for every specimen, bar coding or some form of coding should be used to label the sealed specimen and any aliquot used for on-site testing.

All specimens testing positive should be sent to a certified lab for further evaluation.

In terms of copies of control forms, I believe every donor should get a copy. The collection site should maintain a copy. Negatives, if it is a negative on site test, the remaining copies can be returned to the referring agency. If it is a positive onsite test, the standard procedure of sending the original to the lab, copy to the MRO, and so on, would be done.

The specifics of the on-site testing device is going to depend on the device. I believe that typically the actual on-site test can be done after the chain of custody is complete and the bottles are sealed.

That way, it doesn't have to be done in view of the donor.

If it is a positive, it is going to go on to the lab, all sealed and ready to go.

The Department of Transportation has established regulations for on-site screening for alcohol. Criteria have been developed to evaluate test devices, and they have established the conforming products list.

I believe that similar standards can and should be established for on-site drug testing devices, to evaluate sensitivity, specificity, precision around the cut offs, and other variables, to determine suitability for incorporation in federal programs.

Model collection and screening test personnel and procedures for quality assurance and quality control can also be established for using these devices, as they have for on-site alcohol test devices in the DOT programs.

I am going to give you a little time here, because I am going to summarize up by saying that I think this meeting really represents a significant opportunity to move the field of drug testing forward, and to broaden the application of testing into these other areas where it is so badly needed, in which the current system is either not feasible or too costly. Thank you all very much for listening.

(Applause.)

DR. WILLETTE: Thank you, Michael. We have another Michael from the draw of the hat that is going to follow.

Dr. Michael Peat joined Lab One in Lenexa, Kansas in June of 1994 as the senior vice president for toxicology, and currently serves in that capacity as the executive vice president.

In this role, he is responsible for the operation of the toxicology lab, as well as the entire substance abuse sales force and client services of that facility.

Prior to that, Dr. Peat was the vice president for toxicology for Roche Biomedical Laboratories and CompuChem Laboratories. His previous experience is with the metropolitan police forensics science lab at the new Scotland Yard in London.

He has had a long, interesting journey from London to Kansas, and yes, this is Kansas.

He served at the Center for Human Toxicology at the University of Utah and other appointments, prior to joining Lab One. Dr. Peat has an undergraduate degree in chemistry, a doctorate in pharmacology, and is certified by the American Board of Forensic Toxicology, to which he presently serves as the director of that board.

He has been a member of this Drug Testing Advisory Board until 1994, and serves as a consultant to the College of American Pathologists on their toxicology resource committee.

He is the author of many peer reviewed articles and book chapters, and is also in demand as a speaker.

Dr. Peat is going to talk to us about the chain of custody and collection procedures surrounding that of saliva.

DR. PEAT: Thank you, Bob. I had to move to Kansas. It was the only time zone I hadn't lived in.

I am going to talk today about saliva, for the next few minutes. Before I get into the collection and chain of custody issues associated with it, I would like to give you a little background as to saliva and what is currently being done with saliva.

It may come as a surprise to a number of you that there is a large amount of saliva testing going on currently in some labs in this country.

In fact, there is a lot of literature about saliva as Yale mentioned in his talk. Over 10 years ago, in the first volume of the Advances in Analytical Toxicology, Brian Cadey wrote a whole chapter on saliva testing for drugs, including drugs of abuse.

There is extensive literature available on saliva and what it means and how it can be tested for.

Saliva, however, is not common spit. We are not dealing with people spitting into test tubes and sending common spit specimens to the lab.

We are dealing with oral fluid. Even though I will revert to the term saliva during the next two days, and I am sure the other speakers will use that term, we don't necessarily mean that it is saliva. We really should refer to this as oral fluid.

There are several sophisticated collection devices -- one of which I will describe in a few minutes -- available for the collection of saliva, that refer more to that being an ultra-filtrate of plasma than being spit.

It really is an ultra-filtrate of plasma mixed with the oral fluid in the mouth.

Who has been testing saliva? Well, obviously the transportation sector is allowed to test saliva for ethanol and there are NITSA approved devices for that purposes, and a number of large transportation companies currently use those.

Laboratories in the risk assessment or the insurance testing business have been testing saliva for a number of years.

There are three such labs in Kansas City -- mine, Osborn

Labs and Clinical Reference Labs. Two of those are NIDA/SAMHSA certified at this point.

There is extensive experience in testing saliva in the risk assessment business.

There has been a small amount of testing for therapeutic drug monitoring reasons. That is not commonly done today. Certainly there are research groups that have focused on that, as there have been researchers focused on other things in terms of testing saliva for drugs.

Saliva has been FDA approved for HIV testing. That is obviously where the risk assessment labs have focused their efforts.

They also test saliva for creatinine and for cocaine metabolite. Several hundred specimens a day are done for those three tests, in my lab and several hundred more in other labs. So, again, there is extensive experience with saliva testing.

As I said earlier, there have been various reports on the detection of other drugs of abuse.

Really, when we look at the general guidelines for collecting saliva or collecting bodily fluids such as saliva, they are the same as for urine.

Dr. Cone described those protocols and the principles thereof in his first talk, and I echo them for saliva.

Correct collection, documentation, sealing, shipping specimens, et cetera, to the lab, all those principles apply to the collection of saliva.

The specimen collection would require a collector, not necessarily a phlebotomist, but certainly a trained collector.

A number of the insurance companies use insurance agents to collect saliva, rather than paramedicals to do that job. So, you need training but you don't need extensive training.

Obviously, you can seal the specimens, ship them in bags or boxes to the labs under chain of custody. That is currently being done today in the insurance business, not necessarily to the degree of documentation that workplace testing requires, but certainly the basic principles are there in terms of collection of saliva today.

This is one such device. This is the episcreen device manufactured by Epitope from Oregon. If you can look at the top of the slide, you will see a little pad on a stick.

That little pad is put between the gum and the cheek and that is left there for a period of time, inserted in the tube, which contains this fluid, then sealed. Obviously bar coded seals can be used, labeled seals can be used, and then the whole box containing this kit is shipped to the lab.

Once at the lab, it is received and the documentation. That tube is spun down, and certainly you can then go ahead and analyze the fluid that is collected from that saliva collection or oral fluid collection.

There are other collection devices available. They may not necessarily collect the same type of fluid, but they are certainly focused on the same type of principle.

That is, they have basically been developed to test for HIV.

So, they are going to collect the same type of fluids.

In summary, although I refer to it as saliva, it is not saliva. It is oral fluid. There is considerable experience available with certain types of analytes, cocaine, creatinine, and in some cases other drugs of abuse.

There are collection devices available commercially. Obviously, the guidelines for collection and chain of custody, et cetera, are very similar, if not identical, to the collection of urine specimens. Thank you.

(Applause.)

DR. WILLETTE: Thank you, Michael.Our next speaker is Neil Fortner. Neil joined PharmChem Laboratories in 1991, and serves as the director and vice president of laboratory operations.

Prior to joining PharmChem, he also served as the director of toxicology at another HHS certified forensic drug testing laboratory. Neil has had more than 15 years experience in forensic toxicology, and serves as the responsible person at PharmChem for the laboratory certification, and also serves as an inspector to that program and to the College of American Pathologists.

He is a member of the American Society of Clinical Chemistry, the Society of Forensic Toxicology and the American Academy of Forensic Sciences. He is a member of the American Board of Forensic Examiners.

As vice president and laboratory scientific director, Neil is responsible for the technical quality and reliability of the test results. This includes the whole of activities that have been described as essential for producing forensic quality results in the laboratory.

He has a bachelors degree in biology from Hiram College and a master's degree in biology and biochemistry from Western Kentucky University.

Neil is going to talk to us about the collection and chain of custody procedures dealing with sweat.

MR. FORTNER: I would like to first of all thank the Drug Testing Advisory Board and Health and Human Services for the opportunity to look at new technologies and new advancements as they may assist us in the war on drugs.

The testing of sweat for drugs of abuse is not new. It goes back many, many years.

If you look at the literature, one of the things that has always limited that forensic application is how do you collect the material itself.

These are just some of the techniques that have been used in the past to collect sweat. We are talking about different types of sweat here -- unsensible and sensible sweat, as the primary mechanism of the body to cool itself.

You can certainly go back and look at, in the literature, techniques using capillary micropipettes, literally dripping, swabs or skin wipes.

You get into some rather elaborate type of body bags, if you will, which could present some rather interesting challenges in the collection procedure.

What I am going to talk today about, and subsequent speakers in the day and a half is not really designed to be a promotional item, per se.

We are going to talk about an application that is generally termed the sweat patch. As Dr. Caplan addressed in his previous introduction, it is one of the mechanisms that is currently used to collect sweat for analysis for drugs subject to abuse.

Now, the property of the patch itself is what sort of lends itself to this application. It is a film -- a polyurethane film -- which is over top of an absorption pad.

On top of the absorption pad is a unique identification number. Under that is a release liner.

The properties are such that if you look at it from a sandwich component, you have the urethane film, a release liner, and then an absorption pad of medical grade blotter paper.

How does it work? Because of the properties of this urethane film, it actually becomes a physical barrier. It will not allow environmental contaminants through. This is based on the molecular pore size and also the structure of the pad itself. It does allow water vapor, though, to be passed through the pad.

Drugs, as they are deposited in the sweat, dissolved in that, then become entrapped within the pad itself, and then you can subsequently remove the pad for analysis.

This is sort of the underlying principle of how this patch works. You have a urethane film, an adhesive that is in there.

The adhesive literally permeates the upper epithelial layers. This is one of the properties we will talk about a little bit later in terms of a tamper evident characteristic.

As these cells infiltrate the adhesive and the pad or the patch is physically removed, a layer of those cells come with it, which makes it very difficult to reapply the patch.

Now certainly as Dr. Caplan addressed, in the application, certain areas that you want to be concerned about is going to be an external contamination issue.

If an individual is involved in illicit drug activities, there is a high likelihood that there could be illicit drugs on the skin itself.

So, the first thing that you want to do is prepare a site for the application of the patch. Now only does this physically remove possible contaminations, but it also removes oils from the skin which are essential to get good adhesion.

Following clean of the area -- typically it is with an isopropyl swab -- you can then apply a patch. You can see it here in its component.

There is a urethane film right here on the outside edge.

This is the patch itself. This is a unique number within the upper layer.

Simply by applying the patch to surface area and then physically applying pressure around the absorbent pad itself you can then remove the outer window frame and apply additional pressure throughout the rest of the upper layer.

Now, this patch can be worn for a number of days, including up to several weeks if desired. The reason we can do that is when you first initiate chain of custody there is a recording of this unique patch identification number in here.

That chain of custody then is matched up to a bar code impregnated upon the document itself. An individual can then leave, come back at a later time to be set up, and remove the patch itself.

In the removal process the first thing they are going to need to do is verify that the patch number matches the chain of custody for the patch that was originally applied.

Certainly it is conceivable that somebody may try to physically remove the patch, replace it with a counterfeit or other bar code, or replace it with another patch, if they have access to that.

So, verification of the patch number is rather critical, in terms of linking that particular patch to that particular donor at a specific application time.

Now in order to maintain this chain of custody, it is recommended that you have the donor physically start to peel the polyurethane layer away.

All they need to do is just peel the top part of it away to expose the patch itself.

As Dr. Caplan mentioned, one of the other issues can be external contamination. So, if the individual is, again, involved in illicit drug use and they happen to have drugs on their hand, it would be much better for them to remove the upper layer.

We would then take a disposable tweezers and just pull the pad itself out of the patch dressing.

This is then placed into a forensic sterile bag, would then remove appropriate bar codes from the chain of custody which has already been initiated because it would have recorded unique patch number, date of application, donor's identity.

Proper documentation, bar code identification is very similar to the urine process that is established under the HHS guidelines, using very similar chain of custody, positive ID bar code.

Again, it is analogous to the guidelines, removal of the tamper seals for sealing the forensic bag. You can go through the same procedures, documentation of donor signature or initials, collector initials and date, bar code on the bag, bar code on the tamper seal.

At that point in time, this sample is prepared and ready for shipment to any centralized laboratory that can conduct this particular analysis. I thank you for your time.

(Applause.)

DR. WILLETTE: Thank you, Neil. Our next presenter is Dr. Donald Kippenberger, who serves as the vice president for laboratory operations and is laboratory director at Psychemedics Corporation, which is a certified forensic drug testing laboratory specializing in testing hair for drugs of abuse.

Prior to joining Psychemedics, Dr. Kippenberger was with the U.S. Army. In his last assignment in the army, after a long career in toxicology, he served as consultant to the Army Surgeon General for forensic drug testing.

He was the technical director of the drug testing program, and was the Army's Surgeon General to the Department of Defense Drug Testing Advisory Board.

Don Kippenberger received a doctorate in physical organic chemistry from Texas A&M University. He is going to talk to us about the collection and chain of custody procedures involved in the testing of hair for drugs of abuse.

DR. KIPPENBERGER: Thank you, Dr. Willette. I want to thank the Drug Testing Advisory Board and Dr. Vogel for inviting me to coordinate this conference on the hair portion.

I hope in the time allotted, that we will be able to show the effectiveness of hair testing in the work place environment.

I have split my talk into two portions. The first is on the collection portion and the second is on what happens to the sample when it gets to the laboratory on a chain of custody.

When I read what we were asked to do I felt that what happens to the sample ultimately would be important.

The first thing we do is we secure the collection area. Unlike some matrices, we can use any environment we can find -- office area. It just has to be low access for people to walk through.

There are really no privacy concerns with collection of hair. People get their hair cut in public all the time.

This is a copy of a picture of our conference room. When you stop collecting samples in this room, the room can go back to being used for what it was used for originally.

Then you gather the collection supplies. Various supplies we have, we have the test request form here. We have a sample acquisition form, what we call a SAC here. This is an envelope where we put the sample.

There is some aluminum foil that we are going to put the sample in. We have a pouch that we are going to seal the SAC and a copy of the TRF in.

We have some alcohol swabs. The purpose of that is just for hygienic purposes, to clean the scissors and the hair clip.

Then we have to make sure that the person has signed a consent form, and also that we have identified the subject.

You can see here the subject has been identified with an ID and they are signing the consent form at the same time.

If there is no ID available, then we have to get someone from the company to verify who the person is that we are testing.

Then the next step is to complete the administrative data on the test request form. Examples of administrative data is the MRO address, the client information, which is usually typed in ahead of time, the collector's signature and name and phone number so they can be contacted if needed, the subject ID of the individual, where the hair is collected from if it becomes an issue in interpretation, and then the chain of custody on the sample.

Next is the administrative data on the SAC. What is on the SAC is a place for the collector to actually sign, saying they collected the sample.

There is a place for the subject ID and a date and time of collection.

Then we can take the bar code, because each TRF is uniquely labeled with a unique number, and place it from the TRF and put it onto the SAC, which you see there.

When you collect the hair, typically what you do is you take a hair clip and lift up a portion of the hair on the back of the hair usually.

You take about a half-inch-wide swatch of hair or lock of hair that is about one layer of hair deep, and you cut it close to the skin. Here is an example of that.

Then keeping grasp of this lock of hair, you place the root end, which is the end next to the skin where you cut, and you place it in the aluminum foil.

Then you place it in a SAC and it is a little easier to see. You can see the nice, even cut portion here that is being placed into the sample acquisition card.

Then you seal the SAC with an integrity seal, being shown there. Then the collector initials and dates the security seal.

The test subject then initials the SAC. They are prompted to read a statement. The statement is -- my photographic memory is running out of batteries so I have to read it:

By these initials, I can certify that I am the test subject, that the test sample contained in the envelope is my sample, and was cut close to the skin. I witnessed the sample collector seal the sample in this envelope.

Then the test subject completes the MRO portion of the form, which is -- it has all the information that I mentioned before, but it also has a place for the donor to sign their name, print their name, put their phone number so the MRO can actually contact them.

Then down in the portion right here, there is a place for the MRO when we send them the result, to either accept the result, to call it positive or negative depending on what the result was, or to cancel the test or to order another test.

Then the SAC and the top copy are placed in a pouch. The pouch is sealed. The subject initials and dates are placed on the pouch.

Therefore, at the point we are getting ready to ship in the lab, the individual has now signed the collection devices twice, once on the pouch, once on the sample, before it is even sent to the laboratory. Then it is sent to the lab.

In hair collection, I wanted to point out to folks that this is a matrix where the sample is in the complete view of the collector from the time the sample is obtained to the point where it is mailed off to the laboratory.

At the laboratory site in the chain of custody, I put this slide in to make sure that if I don't say that we put things in temporary storage, or we perform everything under chain of custody, we actually do.

On the comments that Dr. Cone said on the lab, I did not look at his slides when I prepared these. I agree 100 percent.

The entire lab must be secure. Access to the lab must be tightly controlled and documented. Visitors must be escorted, and accessioning and handling must be limited access areas.

The samples and the data must be handled with the same amount of control.

In our chain of custody we use samples that are accessioned one at a time. The person who opens the package dates and signs the air bill.

The pouch is checked for an integrity seal when it is taken out of the Fed Ex package, which is usually how we get our work, and the pouch is opened and the SAC and the TRF are taken out.

The subject identification number on the SAC and the TRF are compared, to make sure also that the information also on both forms is complete.

The bar code LAN label is placed on both the TRF and the SAC, and the tech verifies the integrity of the SAC seal. Then the tech breaks the seal to make sure there is hair present.

Some of the reasons for invalidating hair samples, the subject initials are missing, pretty obvious, no hair in the SAC -- it doesn't happen very often. TRF is missing, that happens occasionally.

Hair is from two body locations. I mean, head hair and body hair, if it is from two different locations, it is real hard to interpret and give accurate interpretation of results.

The subject name appears on the SAC ends the testing, because it has to be anonymous.

Hair is too misaligned to test. This is sometimes if someone takes it from a hair brush, it is too hard to determine time frames and make interpretation.

If it is Q and S, our laboratory requires approximately 10 milligrams in order to perform the test. Also, if the collector's signature is missing.

The TRF is then completed by the tech and then initialled as to whether or not the integrity of the seal is in place or not.

The SAC is transferred under chain of custody to a weigher. What the weigher does is they take the hair out that was in the aluminum.

They measure the distance from the head. We test usually about 3.9 centimeters. We cut it and then we take an aliquot of that and we actually then begin testing it.

We put each of the samples into LAN-labeled tubes, and that is how we control the movement of the sample through the testing.

The aliquots are placed in batches. The tubes are transferred to the screening laboratory and then we perform an immunoassay for five drug of abuse classes.

About midnight of the day that we receive the samples we have reported all the samples that have come into the laboratory.

We have gone back and have begun re-weighing the samples for the second test for the presumptive positives.

The negative data is reviewed by technologists before they are released to the client.

Re-weighed samples, which are weighed in the manner which I just stated before, are then sent to the screening laboratory and rescreened by immunoassay.One thing that I haven't mentioned is about the washing, which Dr. Baumgartner has put out on how to eliminate external contamination.

At the same time that the sample is going through and the washes are being performed, the washes are also put on a separate chain of custody that mirrors the sample, once the sample has been digested and goes into testing.

If the sample is positive, it then goes over to mass spectrometry after the second screen, where a GC/MS or GC/MS/MS is performed.

For negatives, for your information, we keep our negatives for approximately 18 months. This is done under chain of custody. At the end of 18 months, the samples are destroyed.

Positives are kept for five years, or at least five years. Sometimes we go a couple of months over.

The samples, sometimes we use these samples as some of our blind QCs. All movements into and out of storage are controlled by chain of custody. Thank you very much.

(Applause.)

DR. WILLETTE: The last presenter on this panel, and our third Michael, is Dr. Michael Baylor, who received his PhD degree in pharmacology and toxicology from West Virginia University in 1976.

His career was initiated in the clinical toxicology and post mortem toxicology laboratory in Germany and I believe that was with the U.S. Army at that time.

He has been involved in the field of forensic toxicology and drug testing for over 20 years.

As part of his service in the army, he was the director of the army's drug testing laboratory at Fort Meade from 1985 to 1987.

He retired from the army that year and became the technical director of one of the largest commercial drug testing laboratories at the time, which was located in Research Triangle Park.

He has been a laboratory inspector for the National Laboratory Certification Program since 1988, since it began.

Not to totally abandon government service, Michael then became a senior staff fellow at the National Institute on Drug Abuse from 1990 to 1993, and served an additional year as a special expert with the newly founded Substance Abuse and Mental Health Services Administration, which we know affectionately as SAMHSA.

In that capacity, he served with Donna Bush's staff in helping to further develop and implement the work place drug testing program and laboratory certification program.

He left NIDA to join the Research Triangle Institute and serves as a senior research forensic toxicologist, directly involved as a senior staff person in the operation of the National Laboratory Certification Program.

In that role his primary areas of responsibility are the laboratory inspection process and the remedial action requirements. Michael is going to talk to us about the collection and chain of custody issues which are well in place, but still remains one of the most vulnerable areas in the whole process, and that is the collection of the specimen.

DR. BAYLOR: I also would like to thank the Drug Testing Advisory Board for the opportunity to speak and to make a statement. Human drug testing has been proven to be a reliable tool in deterring illicit drug use in the work place, when it is performed and interpreted using appropriate forensic procedures.

The validity of urine drug testing results requires that consideration be given to the methods that are used for the collection as well as transportation, analysis and interpretation of results. Certainly the collection process is the initiating step.

The collection site, by definition, is responsible for collecting the specimen. The collection process places much emphasis on properly identifying the specimen, as well as labeling the specimen and packaging and shipping the specimens to the laboratories.

With today's technology, bar coding has eliminated much of the transcription and human error that had been formerly associated with the process of identifying and labeling items.

Tamper evident adhesives have again ensured the integrity of the specimen and the seal of the specimen containers, as well as lids with the proper sealing technology have minimized the ability of aqueous specimens to leak from those.

The utilization of non-leadable and non-tearable sealed bags or containers, primary devices, have again eliminated the risk of specimen leakage, should it occur.

Specimen site also was responsible for initiating the proper documentation for the process of the identification and collection of specimens.

The collection site is, lastly, responsible for establishing adequate security measures for not only the collection process, but as well the collected specimens.

A designated collection site must fulfill at least five basic requirements. It must, for the case of urine collections, have an enclosure where the privacy for urination is possible.

It must have access to a toilet for the completion of the urination episode. It must also have moist towelettes or a sink for the proper washing of the hands upon completion of the specimen donation.

It must have a suitable surface for writing for the completion of the documents necessary to complete the collection process.

It must have restricted access. Restricted access is so that the collection site is secure during the collection process.

Under ideal conditions, the collection site should be secure at all times. However, ideal is normally not what is achieved. Facilities that are used for other purposes may be secured for the purposes of specimen collection by visual inspection.

Visual inspection normally would include line of sight, that all areas of entry and exit are within the line of sight of the collection official.

It should allow for the prevention of undetected entry or exit to the area, and it should have effective restriction of access to the collection materials and supplies used in the process of the specimen collection. It should allow access to authorized personnel only.

Without the collector assuring the integrity of the specimen and adhering to the collection process by collection SOP, any specimen may not be considered as a valid piece of evidence.

The collection site procedure establishes that primary link between the urine drug testing results and the specimen donor.

It is at this point that that specimen and the results to be generated by that specimen are linked to an individual donor.

Donor identification is one of the initial steps. This is generally, by today's standards, a photo identification in most collection procedures.

Positive identification by an employer or agency representative is an alternative to the possession of a photo identification.

There are other identifications that may be allowed by specific work place drug testing plans.

It should be noted that identification by a co-worker or a co-donor is generally not acceptable, and identification by a single, non-photo ID is generally not acceptable in the collection processes.

One of the key points in the collection process is ensuring the integrity of the specimen. Certainly for any specimen collection, clean, unbreakable, leak-proof containers should be utilized.

The potential for substitution, adulteration or other attempts to sabotage must be limited. In the case of the individual

having to donate the specimen in the privacy of a stall, certainly that is a concern.

Split specimen collection is generally conceded to, again, add to the integrity of the specimen, allowing the individual to have a sealed bottle A and a sealed bottle B that would be sent on for appropriate analysis.

There are certainly processes that can significantly minimize adulteration or substitution or other attempts to sabotage a specimen.

Certainly the measurement and recording of the temperature of the specimen can be utilized a lot. This is generally performed within minutes of the collection.

You can control access to hot water as far as substitution or dilution attempt. There can be the use of bluing agent in the toilet bowls which are used for the collection process.

One can monitor dilution by testing specimens for creatinine and specific gravity, as detailed in a number of the work place programs.

There is certainly a long list of collection supplies which are utilized in the collection process. Certainly appropriate ID to identify the collection, the use of wrapped, sealed collection kits, or specimen bottles with lids, wrapped, sealed specimen containers.

They include the utilization of temperature strips or temperature measuring device, a custody and control form to document the integrity of the specimen and its identity, the tamper-evident labels and seals for specimen bottles with today's adhesive.

You can include a separate supply of tamper-evident seals as manual dexterity skills do get challenged with the utilization of plastic gloves in trying to apply adhesive labels.

Shipping containers are included, and a bluing agent for the toilet bowl.

One of the most critical aspects of the collection process, or any collection process, is establishing the chain of custody.

The chain of custody was described by Dr. Cone, and chain of custody is generally perceived to be the procedures and the documentation which attest to the identity, the integrity and the security of the specimen.

Urine drug testing has utilized what was referred to as a custody and control form which is initiated at the collection site.

This is a seven-part form. Copy one through three are white copy originals. Copy one original -- all three of those copies generally accompany the specimen to the laboratory.

One is the original. It accompanies the specimen to the laboratory. It eventually becomes the laboratory archive file copy of the specimen results and it is maintained by the laboratory for a period of two years.

Copy two, or the second original, is that accompanying the specimen to the laboratory. It is generally utilized by the laboratory to report results to the medical review officer.

Copy three is a split specimen copy. It accompanies the A and B bottles in a split collection to the laboratory.

It would generally be used to further test and further transport a specimen to a B lab for B bottle analysis.

Copy four is a medical review officer copy that goes from the collection site to the medical review officer, and it is pink.

Copy five is a donor copy. It is green. It goes with the donor upon completing the collection process and leaving the collection site.

Copy six is a collection site archive copy, and copy seven is the employer copy sent from the collection site to the employer. Collection steps generally involve the identification of the donor, and the donor verifying or completing the information in step one of the custody and control form, which indicates the employer as well as the medical review officer to which results will be reported.

The donor is generally asked to remove unnecessary clothing and secure personal belongings. These are generally bulky outer clothing or briefcases, et cetera.

The donor is instructed to wash his or her hands prior to collection so there is no adulteration contamination possible.

The donor receives the collection container. The donor takes the container into the rest room. Upon completion of donating the specimen, the donor gives the specimen to the collector.

The collector checks the volume and the temperature and pours the specimen into the transport specimen bottles.

The collector would then secure the bottles and apply any label or seal that has been properly filled out. The donor would initial the label on the specimen bottle.

The donor would read and sign the certification statement on copy four of the custody and control form. Only copies four through seven contain donor identifying information.

The collector would complete the certification step, which is step five of the custody and control form. The collector would sign line one of the custody and control form block, which is step six of the custody and control form.

The collector then gives copy five of the CCF to the donor.

The collector completes the chain of custody block.

It places copies one, two, three of the custody and control form and the specimen bottle or bottles in the plastic bag and seals the bag.

The collector places the sealed bag usually into a secondary shipping container and seals that container. The collector sends copy four of the CCF to the MRO, copy seven of the CCF to the employer, and files copy six of the custody and control form in their archive records.

In summary, the collection procedures are the key, I believe, to the success of all drug testing programs.

The collector at the collection site is the one individual with whom the donors will have direct, face-to-face contact.

It is the initial link that links a specimen to an individual. If the specimen is reported positive, the entire collection process must be able to withstand the closest scrutiny and answer all challenges to its integrity. Thank you.

(Applause.)

DR. WILLETTE: I wanted to thank the presenters for being generous with their times. They all stayed under their time limits, which will leave us a little extra time in dealing with questions and answers.

According to the procedures that were set up for this hearing, at this stage the panelists that are here on the podium and the members of the Drug Testing Advisory Board are the only ones permitted to ask questions.

As was explained to you earlier, those of you from the public may sign up and make comments and ask questions at the assigned times on Wednesday.

I think it is appropriate, because we have a few minutes, to identify the board members. Now, you have met Dr. Autry and Dr. Donna Bush and Dr. Yale Caplan, who is a member of the board. Certainly Dr. Cone is eligible to ask questions as well, although he is not currently a board member.

I think it would be good to introduce the others that are here. In the first row we have Dr. Alan Skip Jones from the University of Mississippi, Dr. Dick Pinder from the State of Connecticut, Dr. Tai Kwong -- maybe you can identify yourselves. If they ask questions, you will know who is asking. Why don't you stand up so they can see who they are.

We have a PhD doctor, but also Colonel Jacobs from the Army Lab in Fort Sam Houston. Diane Wilkins is here, and Melanie Mallory from Scottsdale, Arizona.

Is Dr. Holland here? We didn't see him earlier. Dr.

Holland is another board member. He is an MRO with the services. These are the board members who we are assigning this monumental task of refining and expanding the mandatory guidelines and giving us advice on work place testing.

So, who is going to address the panelists with the first question or going to make the first comment?

DR. CONE: I just wanted to return to something Mike said about specimen collection, which I fully agree. I want to make a quick comment and then maybe have a couple of the panelists respond.

I hope you will take my comments as constructive comments. Mike said that without the collector ensuring the integrity of the specimen and adhering to the collection process, the specimen may not be considered a valid piece of evidence.

I guess I have a question for the sweat people and the hair people. In both the application and removal of the sweat patch, of course there is some risk. You outlined the steps that you would take to avoid picking up environmental contamination.

It looks like it is critical to make sure that the skin is clean and that you apply the patch over clean skin and that you not touch the patch.

I do see that you are trying to use forceps, I believe, to apply and to remove the patch.

I am still a little concerned that the person, if the collector or the applicator were not wearing gloves, that there could be inadvertent touching of the patch.

I know that sounds trivial, but I truly believe and I have seen evidence that you can transfer millions of nanograms of cocaine from surface to a patch or to anything else that you touch.

I am concerned that you could do more in possibly eliminating the risk of environmental contamination.

In the same vein, I guess I am shocked to see the collector collecting hair handling the specimen. That is tantamount, to me, to seeing the collector for a urine device sticking his finger in the cup and checking to see whether it is warm or not. I am concerned, obviously.

A very simple fix, of course, is to wear gloves. Why don't

you guys use gloves?

DR. WILLETTE: Thank you, Dr. Cone. Who wants to be first?

DR. KIPPENBERGER: We wash the hair, so we don't see it. External contamination, that is not how you can contaminate hair.

DR. WILLETTE: Does anyone else have a response?

MR. FORTNER: In response to your concerns, certainly the potential for external contamination exists, which is why you would go through the procedures of gloves and disposable tweezers.

I just can't help but think that in your scenario, most collection sites that I have visited don't have a lot of free cocaine floating around, but that potentially could exist.

I think the precaution in the use of disposable gloves and disposable tweezers in application and removal of the patch are reasonable precautions to address external contamination concerns.

DR. WILLETTE: Let me just interject here a minute, I neglected, in my responsibility here to identify the ex officio members of the board.

I see Mr. Warren Bush in the audience here from the Nuclear Regulatory Commission and Ken Edgell from the Department of Transportation and Ron Dison and Aaron Lucas from the Administrative Office of the U.S. Courts.

I think -- is it Captain Jeminic(?) or Colonel Smith from the Department of Defense? This is Captain Jeminic from the U.S. Navy who is the Department of Defense representative.

Any other questions or comments from the board members or panelists?

DR. WILKINS: I just have a quick question actually for each of the members of the panel, specifically Dr. Peat, Mr. Fortner, Dr. Kippenberger and Dr. Baylor.

As we are all aware, under the federal guidelines there are accommodations for collection of split specimens. Certainly with each of these matrices it would be possible to collect more than one specimen, presumably.

Do you have any information about the agreement between split specimens that might occur? Has anybody looked at that?

DR. PEAT: For saliva, not that I know of. It is in the QC talk.

MR. FORTNER: There are no provisions for that in the sweat patch. It is conceivable to develop that procedure.

DR. WILKINS: So, the collection of the hair at this point in time, do you in some circumstances collect split specimens at this point in time?

DR. KIPPENBERGER: We have at this point a safety net program. If there is a challenge, we get another sample, another collected sample. We also like to get enough hair to repeat the initial assay.

As part of the safety net program, we test a newly collected sample along with retesting the sample that we kept up in permanent storage. All of that is in Ann Marie's talk, or some examples.

DR. WILLETTE: Neil, just to follow up on that, has anybody attempted to apply two patches, one serving as the primary and one as the secondary?

MR. FORTNER: Certainly in some of the clinical trials individuals wore multiple patches. While there is not currently a program in place that does that, it would be easy to address.

DR. WALSH: What was the concordance in your clinical studies there, in terms of left arm, right arm patch. Are you getting the same result out of both patches? I think that is what the basic question is.

MR. FORTNER: Certainly very comparable results, if they are applied to the same portion of the body. Upper arm, there didn't seem to be significant difference right arm versus left arm.

Now, if you were to apply back versus arm, you might see some differences.

DR. JONES: Dr. Kippenberger, I would say also, to add to Dr. Cone's comments and concerns -- and I know we are going to hear more about external contamination later -- but multiple use scissors are a concern, too, to me.

I realize that you are preparing these scissors for cleaning, et cetera, during preparation, but that would just be a comment I have.

I would have one question to you concerning hair testing, though. We frequently are challenged with identity of specimen, et cetera.

I know even in urine collection the substitution issue is there. Are you doing anything -- it is trivial, but are you doing anything to allow the donor to see the hair cut in a mirror or anything like that?

It can always be claimed that a hair specimen has been substituted, if they do not see that specimen removed. Are you doing anything like that in your program?

DR. KIPPENBERGER: No, sir, we don't, to my knowledge.

DR. JONES: I was just curious. We see these challenges and we have to be concerned about these challenges. So, the integrity or the identity of the specimen was a concern.

DR. KIPPENBERGER: Plus, we get to collect a new sample.

DR. JONES: Mr. Fortner, I did have one question. You made

a comment about the patches, that they can be on for extended periods of time.

What are the data that you have about the life of the patch?

How long are you leaving it on, your collection of this sweat, et

cetera?

MR. FORTNER: It depends upon the program itself. We know from the clinical studies and from the controlled studies where drug was administered that it shows detectable on the patch within 24 hours.

Most of the programs are either on a one to two-week cycle, depending upon what type of program, how often they might want an individual to report.

Personally I have worn patches up to 30 days. So, it depends upon the duration of the program and how they design it.

DR. JONES: So, you are saying they will stay on up to 30 days, is the type of thing that you would typically see as an outside window.

MR. FORTNER: I think in general, I mean, you get some subject to subject variability off of that.

DR. WILLETTE: Following up on that, Neil, what is the incidence of either adverse reactions or patches coming off as part of the inappropriate application of them, or in the collection process?

MR. FORTNER: Certainly if the application site has not been cleaned, while it is removed you will have a higher incidence of fall off, or failure to comply with adhesion.

There are some individuals who, although the material is hypoallergenic, may show an allergic reaction. In our experience that number has been very, very small.

DR. WILLETTE: What about showering and washing in terms of the integrity of it?

MR. FORTNER: Simply because it is a one-way barrier, individuals can carry out normal functions -- swimming, showering -- without any danger.

The biggest problem, I think, has to do with either active removal of the patch or the physical abrasion which would cause it to try to peel off.

DR. CAPLAN: A couple here and there. To Mike Walsh, you mentioned you didn't think it was necessary for tests to be done in the presence of the donor.

Could you elaborate on that a little bit more, whether you think it is really a good practice or not?

DR. WALSH: The on site test?

DR. CAPLAN: The on site test being done in the presence of the donor.

DR. WALSH: I would recommend that it be done in the presence of the donor. I think if you have gone ahead and sealed and initiated a chain of custody, that the test does not have to be done in the presence of the donor, if the only action on a presumptive positive is to send the sealed specimen and full chain of custody on to a certified lab for the standard confirmation procedure.

DR. CAPLAN: You are talking about in a situation where you have the sealed specimen and the test specimen on site is separate.

DR. WALSH: That is correct. That was the caveat.

DR. CAPLAN: For Dr. Kippenberger, you mentioned, and I was not clear, that you do not test unaligned hair. Did you say that? Is that correct?

DR. KIPPENBERGER: That is true.

DR. CAPLAN: What is the probability of a specimen coming in and not being tested because it was unaligned.

DR. KIPPENBERGER: Very slim.

DR. CAPLAN: I can visualize a lot of people with short hair that you couldn't align --

DR. KIPPENBERGER: If it is less than 3.9 centimeters, obviously then it doesn't matter if it is unaligned. I am talking about hair that comes in like Donna's length of hair, and it is taken out of her hair brush, not a nice, clean cut next to the skin.

DR. CAPLAN: You would recommend testing hair which is otherwise smaller than designated length.

DR. KIPPENBERGER: It doesn't matter which end is the skin end, because it is within the 3.9 centimeters that we test.

DR. CAPLAN: So, it really doesn't matter whether it is aligned then.

DR. KIPPENBERGER: In that case.

DR. CAPLAN: This is a general question for everybody about training. You all mentioned it to varying degrees and I thought it would be useful to comment.

As to the level of training you think is needed for each of these types of collection, the necessity for documentation which I think we all agree is there, but to what degree.

Although we don't have it and we all know it is the weakest link probably in the urine collection process, the need for certification of training processes.

DR. WILLETTE: Let's do them in reverse order. Let's let Michael start, the other Michael, at the other end.

DR. BAYLOR: Certainly I think training for the urine drug testing is necessary, and certainly those individuals that function as collection site officials should be trained in the technical as well as the administrative aspects of the collection process, and to adhering to an SOP, a standard procedure that is utilized in each and every instance.

DR. KIPPENBERGER: We have a lot of training aids. We send out people to train. We have a collection manual that just about anybody in the United States, I think could follow very easily.

We have a collection video that shows in depth for anyone on how to collect. We have a test that we give all of the people that collect samples, to make sure that they know how to collect them correctly.

Those tests are faxed in to us or are given to us. We correct it and then, based on the performance on the test, we give the collector a certificate saying that they are a certificate collector.

This data base is kept at the laboratory and, if required, could be shown to anyone that this collector was trained or not.

MR. FORTNER: I think certainly training is very critical, whether it is urine or hair or sweat. There are procedures that need to be followed, techniques that need to be taught.

That is something that is easily done through a variety of mechanisms -- personal, on-site training, on-site training coupled with videotape for subsequent retraining, and then very specific standard operating procedures, not just for the application, but for the removal and documentation therefore.

It is something that is done quite readily, but it is quite critical. I would certainly concur.

DR. PEAT: I would agree with the general comments. I think the experience in the urine testing arena is not that the specimens are collected incorrectly. The documentation is completed incorrectly.

I think training in the documentation aspects of it is very, very important, no matter what the specimen collected may be.

I think actual collection of specimens is not such a big issue as training in documentation.

DR. WALSH: I think for the on-site testing training becomes an even more critical issue. I think some standards for the certification of on-site testing personnel can easily be established.

I think 10 years ago this was a whole area that we sort of avoided dealing with because we were building a system from scratch.

I think at this point in time with the DOT model in place, with the various organizations now certifying MROs, I certainly don't think there needs to be two and three day courses to do this.

A minimum amount of training specifically in the documentation procedures and the use of the device, good knowledge of the particular device to be used on site would be critical.

DR. CAPLAN: Since the last part was avoided by three out of five people, one quick liner. Do you think there should be a certification process for people doing any feature of the collection that you are representing?

DR. PEAT: The answer to that is yes, that would be ideal if you could do that. The problem is that the turnover of collectors at clinics, et cetera, is extremely high.

Certifying collectors is not such an easy process as certifying medical review officers or lab directors or some other people where there is not such a high turnover.

DR. CAPLAN: I agree that it is difficult. I was just trying to get a sense of whether it would be a useful thing to look at, or whether it is impossible.

MR. FORTNER: I think it is a useful thing. I think that certainly some of that can be done as part of the training process, where they are certified and provided with a certificate.

Whether it has to be a national program I think is maybe a question.

The challenges that come in, independent of what the biological fluid, are always going to have some component at the collection site, and was the person trained and how did you document that particular training.

DR. WILLETTE: We will give Dr. Bush the privilege of the last question.

DR. BUSH: Thank you very much. This one will be directed specifically to Mike Walsh concerning on-site drug testing.

I noticed in the discourse how you spoke about how an individual, a donor of the specimen, could be tested right there.

You know, the collector collects the specimen, the result comes back, and the individual right there, then, sees the result.

This is extremely effective in treatment arenas, as we all know. I am going to relay to you grave concerns expressed to me quite recently by collection site individuals, who have described their experiences with collecting alcohol specimens, saliva specimens, by onsite techniques, and then the individual sees that result right there and they have to go on to a confirmation site.

As I understand it, Bubba isn't always well receiving this information. As I understand it, the presence of drugs in the urine may be occurring at a greater frequency than alcohol in the saliva.

So, I have had collection site people express grave concern to me, especially on drugs like amphetamines and opiates where there may be cross reactivity to other substances, other than just the focus substances for which the device was made.

I don't know. How would you suggest collection sites, on-site testing sites, dealing with an immediate cocaine positive or amphetamine positive, showing this to Bubba and saying, Bubba, what do you think of this, we have to move on and this and that. Help me out with this, Dr. Walsh.

DR. WALSH: I should mention that I hired Donna and she has never forgiven me for that.

DR. WILLETTE: This is pay back time.

DR. WALSH: I think it really depends on the reason for the test. If it is a post-incident test or if it is a random test where someone is in a safety sensitive decision versus a pre-employment test is going to make a big difference.

I think most employers tend not to run the test in view of the employee, although I am finding more and more that especially in larger especially manufacturing companies where the test is being done by the staff nurse in the company medical facility, they will do it in front of the employee.

The way in which they typically handle it is, I am going to need to send this on to the lab for additional testing, and not that you are positive for cocaine or whatever.

They have been well trained as to the fact that an on-site result is a presumptive positive and it should be sent on for further evaluation.

I think particularly in personnel departments where the on-site technology is being used, the applicant specimen is sealed, chain of custody initiated.

Then the test is run either on an aliquot or in self-contained devices. The test can be run and the results are read after the individual has been asked to step out into the other room.

The same story is told to the individual. Thank you, we will call you in a day or so and let you know the results of the test. The negatives can be sent on with a negative result.

DR. BUSH: What would concern me there then -- that is exactly where I want to go. If the test is negative, you tell Bubba, hey, this is negative, so Bubba is not all that out of it.

Bubba knows these technology. So, Bubba one comes in, Bubba two, Bubba three, and they are all being told oh, negative, negative, negative. Bubba four, well, we will just have to send this on to the lab for some more work.

DR. WALSH: Right, and a lot of companies like the idea of having a third party deal with it and not getting the results back for 48 to 72 hours. Then the results go back through the system and personnel and somebody else deals with it.

I think this is the whole issue that DOT is now dealing with, of having to confront the individual immediately and deal with it, get them to an evidentiary breath testing device within 20 minutes.

There will be procedures that will evolve on how to deal with people who are belligerent.

I think, again, the individual who is actually conducting the on-site testing is going to have to develop those kinds of skills. I don't have a magic answer for you.

DR. WILLETTE: Thank you. We have accomplished our mission for this first panel. We have addressed many of the topics and issues of concern dealing with collection and chain of custody.

We have also stopped at the appointed hour of noon. You have one hour and 15 minutes for lunch and the program will start again promptly at 1:15. Thank you very much.

(Applause.)

(Whereuponn, at 12:01 p.m., a recess was taken until 1:15 p.m., the same day.)

A F T E R N O O N S E S S I O N (1:17 p.m.)

DR. BUSH: We will reconvene the open meeting of the Drug

Testing Advisory Board.

We will now convene this session entitled Initial Test

Reagents and Procedures Panel. The moderator is Dr. Marilyn Huestis. I have a few words I would like to say about Marilyn.

Marilyn A. Huestis received a bachelor's degree in biochemistry from Mount Holyoke, a masters degree in clinical chemistry from the University of New Mexico, and a doctoral degree in toxicology from the University of Maryland at Baltimore.

Dr. Huestis has been working in the fields of forensics and analytical toxicology and clinical chemistry for more than 20 years.

She was chief toxicologist of Nichol's Institute, San Diego Laboratory, and directed the emergency toxicology, therapeutic drug monitoring, analytical toxicology and urine drug testing operations. She was a child graduate.

Dr. Huestis was nominated for the research staff fellow award of the Addiction Research Center, NIDA, where she studied the pharmacokinetics and pharmacodynamics of marijuana use.

Dr. Huestis has published important papers in the area of marijuana pharmacokinetics and pharmacodynamics, evaluation and validation of immunoassays, and on the use of alternate testing matrices for drug analysis, including hair, saliva and sweat.

Dr. Huestis was awarded the American Academy of Forensic Scientists Irving Sunshine Award for outstanding research in forensic toxicology in 1992, and was elected to the Phi Kappa Phi honor society for outstanding academic achievement.

She currently serves as secretary of the Society of Forensic Toxicologists, is 1998 program chair for the toxicology section of the American Academy of Forensic Sciences, is an active member and former secretary of the California Association of Toxicologists, serves on the therapeutic drug monitoring and clinical toxicology committee of AACC, and is a member of the International Association of Forensic Toxicology.

Dr. Huestis was co-chair of the joint TF SOFT joint congress held in Tampa, Florida in 1994. Currently Dr. Huestis is a senior research scientist at the Addiction Research Center NIDA, where her research interests include the pharmacokinetics and pharmacodynamics of drugs of abuse in humans, evaluation of the effects of drugs on behavior, including correlations of blood levels of drugs and concurrent changes in performance, evaluation and validation of analytical techniques for detection of drugs of abuse in different biological fluids, and medication development projects, including the use of bupronorphene as a pharmacotherapeutic agent in opioid dependence.

Dr. Huestis is interested in evaluating diverse routes of administration and determining the effects of different routes on the abuse liability of drugs, effects on performance and on the onset, peak and duration of action and the resultant effects.

In her spare time she does things like this. So, Dr. Huestis.

Agenda Item: Initial Test Reagents and Procedures Panel.

DR. HUESTIS: I don't get mad. I just get even. Enough is enough. Is there no body fluid or tissue that can't be analyzed for drugs.

Alternative specimens provide unique information about an individual's drug use history, and describe different segments of the drug disposition story.

Each of the alternative specimens considered today has proven to be a valuable analytical sample for other diagnostic clinical chemistry or toxicology tests.

For example, we have analysis of chloride in sweat in cystic fibrosis patients, therapeutic drug monitoring of theophylline and antiepileptics in saliva and determination of in utero drug exposure in hair.

For the next three days we will have the opportunity to evaluate the use of alternative specimens and on-site testing methods as a means of monitoring employee drug use in work place drug testing programs.

This segment of the program deals with initial test reagents and procedures. The goal of the initial test is to identify specimens that contain drug and/or metabolites of interest at a concentration above a designated cut off.

The initial test separates negative from presumptively positive specimens and must be based on a different analytical principle than that employed in the confirmation test.

As with any procedure, a thorough method validation is necessary. Controlled drug administration and pharmacokinetic studies document the presence, magnitude and time course of detection of drug and metabolites in biological tissues and guide selection of relevant pharmacologic and analytic cut off concentrations.

Important factors to consider include assay sensitivity, specificity, accuracy and precision. Analytes of interest vary in the diverse sample matrices.

Sensitivity is enhanced when assays target the appropriate drug molecules. Cross reactivity with other drug analytes and metabolites may complicate interpretation and decrease concurrence between initial and confirmation test results.

Although the dynamic range of the assay may be limited, linearity around the cut off and assay precision greatly contribute to an accurate separation of positive and negative specimens.

Reference materials should be prepared in the same matrix to adequately assess assay performance.

Sample preparation prior to initial testing and potential for contamination and adulteration vary considerably between the matrices.

The design and frequency of quality control analyses varies with differences in cut off, stability, batch size and drift across the batch.

Other analytical factors include, but are not limited to sample size, through-put, cost, training and education of personnel, and instrumentation requirements.

We are fortunate to have with us today some very knowledgeable and experienced individuals to help guide us through an analysis of the pharmacologic and analytical validation of each method.

Clearly, alternative specimens provide different windows of detection of drug use. Hopefully the next three days will provide us with the necessary scientific knowledge to adequately judge the windows of opportunity that analyses of these alternate specimens and alternate technologies may offer.

Our first speaker today is Dr. Anthony Costantino, who is the director of toxicology at American Medical Laboratories in

Chantilly, Virginia.

Dr. Costantino has a bachelors in pharmacy degree, a masters degree from Duquesne University in Pittsburgh, Pennsylvania. He has a PhD in forensic toxicology from the University of Maryland.

He is an inspector for the National Laboratory Certification Program, a diplomate of the American Board of Forensic Toxicology, a fellow of the American Academy of Forensic Sciences, member of SOFT, TIOFF, AACC and the American Pharmaceutical Association.

As I said, he is currently the director of toxicology at AML. Dr. Costantino.

DR. COSTANTINO: Under the federal guidelines for work place drug testing, each urine specimen must first be analyzed by screening with an immunoassay kit that has been approved by the Food and Drug Administration.

Those kits, as used in the work place drug testing program, test for either a specific drug, as in phencyclidine, or a drug class, as in the cannabinoids, cocaine metabolites, opiates and amphetamines.

One point about the FDA approval process, the FDA approval gives some semblance of equality among the different kits for their specific, intended use.

So, if they are approved to be used for testing for urine a particular cut off for a particular matrix they would be considered to be somewhat equal for that purpose.

The different techniques that are currently employed in the laboratory for work place drug testing include enzyme immunoassay, radio immunoassay, florescence polarization, and kinetic interaction of microparticles.

The initial test is defined as an immunoassay test that is used to eliminate negative urine specimens from further consideration and to identify presumptively positive specimens that require confirmation or further testing.

A negative specimen is any specimen whose apparent concentration of analyte is less than the pre-established cut off concentration for that drug or drug class. That is, the term negative is not restricted to drug free specimens.

Immunoassay technology is based on the ability to produce antibodies to drugs. These antibodies may be monoclonal or polyclonal.

The polyclonal antibodies are formed by binding a protein to the drug and injecting it into the animal, and actually having the animal's immune system develop antibodies to the drug.

Monoclonal antibodies are grown in organic cultures in the laboratories -- not the testing laboratories.

The antibodies must have a label for detection. These techniques rely upon a competitive binding between the antibody and the label drug.

In the kits, the label drug is supplied by the manufacturer and it competes with the unlabeled drug which would be present in a positive urine specimen.

There is label on the drug that determines the detection system that will be used. For enzyme immunoassay, FPIA KIMS and not RIA, would be a spectrophotometer.

These techniques are also considered to be homogeneous because no separation step is needed to separate the bound from the unbound drug complex.

RIA is heterogeneous. It uses iodine 125 and requires a gamma counter for detection. Because the radioactivity from bound and unbound drug can't be differentiated, it must be physically separated.

The sensitivity of the initial screening method as it applies to work place drug testing must approximate that of the confirmation method.

The specificity must be such that the reactivity of the assay to compounds which will not be determined on the confirmation assay are not detected.

There are chemically similar compounds which, at high enough concentrations, will cause a presumptive positive on the initial screening test, but will not be identified as one of the targeted

analytes on the confirmation test.

Each FDA approved immunoassay kit will provide a list of compounds which were tested for cross reactivity, and the concentration at which they were tested.

There are always new compounds coming out into the market for medicinal therapeutic uses, and their cross reactivity to a particular assay will be discovered in the laboratories. That information is made known through the peer reviewed literature.

Substances that interfere with the immunoassay reaction or its detection system are also of concern. Some therapeutic compounds may cause a high background absorbance, which interferes with some assays.

The addition of adulterants to samples may cause false negative results in a number of ways. Some may destroy a particular drug in the sample, while others may affect the technology itself.

The initial testing process is not only affected by chemicals added directly to the urine, but it is also affected by the concentration of the urine itself.

Attempts to produce a dilute specimen include consuming

large quantities of fluid or the ingestion of diuretics.

The laboratory must be aware of adulterants that affect the immunoassay test.

Most laboratories have procedures in place to determine the integrity of the specimen. The procedures may be applied during the initial receipt of the sample, prior to testing, or during the initial testing process.

These include, but are not limited to, visual inspection of the specimen when it is received in the laboratory, looking at unusual colored foreign bodies, detection of an unusual odor, pH or the acid content of the specimen, specific gravity creatinine and osmolality are all measures of the concentration of the specimen, chloride content or unusual spectrophotometric reactions, which of course are determined after the testing is completed.

A laboratory may conduct multiple initial tests prior to GC/MS confirmation. The secondary initial tests may be used because of improved specificity for a particular analyte, or because it is less susceptible to an interferant affecting the first initial test.

The second initial test is subject to the same criteria as the first initial test. That is criteria with respect to cut off and quality control.

Most amino assays have been adapted to automated analyzers which allow for the screening of large numbers of specimens in a short period of time.

Automation helps to minimize human error, and creates a system which may be easier to control.

The automated analyzers act as pipetting stations for both the specimens and the reagents.

The package inserts or the instructions provided by the test kit manufacturers may specify the volume of the sample -- that would be the specimen calibrators and controls -- and the volume of each reagent that must be dispensed by the instrument's pipetting system.

It is important that the pipetting system is evaluated on a regular basis to ensure that proper volumes are dispensed for each of the components.

The laboratory processing staff will typically provide the aliquot for the initial screening. That aliquot will be applied directly to the instrumentation.

The instrument will quantitatively pipette the required amount and it may also allow for the proper mixing of reagents and will monitor any time critical steps.

The instrumentation also requires a means of measuring the end point of the testing reaction. As I stated before, this would require a spectrophotometric detector for some of the homogeneous assays, or a gamma counter in the case of radio immunoassay.

The analyzers may be operated in what is called a batch mode or a random access mode. Most laboratories will run the analyzers in the random access mode, which allows for the testing of multiple drugs on one aliquot.

The batch testing mode would allow for the testing of one analyte per aliquot, so multiple aliquots would be needed, where that same aliquot would be tested repeatedly.

Many automated analyzers have bar code reading capabilities, and are interfaced with the laboratory's information management system.

The bar code label has information which links the aliquot to the initial specimen. The interfacing eliminates manual result entry.

The critical operating characteristics of the pipetting and detection systems must be checked routinely.

All immunoassays must be calibrated at the mandated screening cut off. Calibrations may be single point or multi point. Single point calibration typically requires analysis of

other standards to determine the linearity of the screening assay. Multi point calibrations use a series of standards to determine the concentration response curve for the assay.

If the assay is drug specific, it will be calibrated against that drug. If it is a drug class test, it will be calibrated against a representative analyte in that drug class.

For example, EIA, as I mentioned before, phencyclidine is used to calibrate the assay. Delta 9 THC carboxylic assay is used to calibrate the cannabinoids assay in EIA.

The screening cut off is chosen based on the sensitivity of the assay, as well as its pharmacological relevance.

Cut offs that are too low will result in an excessive number of GC/MS negatives, and cut offs that are too high will result in too many false negatives.

These are the cut offs that are currently mandated under the HHS guidelines. As we know, when the program started, the cannabinoids were set at a cut off of 100 nanograms per milliliter.

There is a proposed guideline to increase the opiate cut off from 300 to 2,000.

The accuracy and precision of the calibration is monitored throughout the screening process by analysis of quality control samples.

A minimum of 10 percent of the batch must be QC samples and must include the following: samples certified to contain no drug; positive controls, four to five, with drug or metabolite; at least one positive control with a drug metabolite at or near the cut off; a

sufficient number of calibrators to ensure and document the linearity of the assay over time near the cut off; and a blind quality control sample.

The preparation and evaluation of QC samples will be discussed, I believe tomorrow by another speaker.

As you know, the work place drug testing program was founded on urine testing. We have a lot of precedent with the use of immunoassays for the initial screening test.

Whether EIA, KIMS, RIA or FPIA has been employed in the program, it has been demonstrated that the initial screening of urine samples by immunoassay is an effective means of identifying the negative donor samples as defined under the federal guidelines for work place drug testing. Thank you.

(Applause.)

DR. HUESTIS: Thank you, Dr. Costantino. Our next speaker is Dr. David Armbruster.

Dr. Armbruster has over 20 years of clinical and forensic drug testing experience. He is certified by the American Board of Clinical Chemistry and certified as a chemist by the American Society of Clinical Pathologists.

Dr. Armbruster has received a bachelor of science degree in chemistry from the University of Missouri, a masters in biochemistry from Rice University, a masters of administration and health service management from Webster University, and a PhD in clinical chemistry from the Medical College of Virginia.

Dr. Armbruster is an inspection team leader for the DHHS and NLCP programs, and a member of the American Association of Clinical Chemistry, SOFT and Clinical Lab Management Association.

Dr. Armbruster recently joined PharmChem Laboratories in 1996, where he is the scientific director for PharmChem's Texas Division.

He is responsible there for the technical quality and reliability of test results. Dr. Armbruster is going to talk to us this afternoon about initial test procedures in sweat testing.

DR. ARMBRUSTER: Thank you, Marilyn. I will be talking about the initial test reagents we use for screening sweat patch.

We are particularly talking about PharmChek, which is the sweat patch that we provide through PharmChem.

As Neil Fortner has already noted, PharmChek is our product, but we don't mean this to sound like a commercial. It so happens that it is the only sweat collection device that is FDA approved and it is the one that we have the most experience using.

When we talk about collection of urine testing, basically the only thing we need is a plastic cup. We need something a little more sophisticated to collect sweat.

Certainly you have seen an illustration from Neil of the sweat patch and it is a step above the plastic cup.

I would leave you with one caveat. Neil has already shown you all the various devices that could be used to collect sweat, going from the sublime to the ridiculous.

If you were to use one of those sweat collection devices, the findings might be different from what we note with PharmChek.

Then again, sweat is sweat, and I would be kind of surprised if you found anything too radically different.

To give you a little bit of background, we consider sweat testing to be an adjunct to sweat testing; in other words, not a replacement for it, but just a little helper to the urine test.

It basically provides you with a long term monitoring device versus the snapshot that you have when you collect a random urine specimen.

We like to think that the sweat and the urine specimens will complement one another. There may be cases where one is negative and one is positive. That doesn't mean that they are contradictory. You have to take them in context.

We do like to try to follow the standardized SAMHSA guidelines as appropriate for sweat testing. Of course, we do initial screening and we do confirmatory testing, the standard two-tier approach.

The initial testing is via immunoassay, and I think other speakers have already indicated that the immunoassay is a screening test designed for high volume testing.

We want to eliminate the negatives and identify the presumptive positives and provide either a positive, yes or no, black or white sort of answer.

We needed to establish testing cut offs or thresholds. As Tony Costantino just noted, SAMHSA has standard, mandatory cut offs for urine testing.

We don't have any such thing yet for sweat testing. It is a new frontier. We had to come up with our own cut offs. That was exciting for us because we could set the pace and establish our own limits.

On the other hand, then we have the onus of choosing wisely what we want for a cut off. We went about it by choosing cut offs derived from an empirical performance based approach, a statistical interpretation of the clinical data.

I have to thank Dr. Cone for indicating that the clinical studies are probably the best approach for coming up with cut offs.

Dr. Cone also mentioned the receiver operating characteristic or ROC curves that we happen to use with the sweat patch to derive sensitivity, specificity, predictive value and thus establish the cut offs for the screening assays.

A little bit more background. The sweat patch did go through the FDA approval process. It was originally submitted for FDA approval based on data from RIA and GC/MS data, which is good.

There was no FDA approved screening assay for a sweat patch at that time. They just weren't out there. So, FDA approval of a sweat patch required clearing a screening assay concurrently with the sweat patch to be used for drug testing.

STC -- and Dr. Niedbala is a representative of STC -- agreed to work with the patch manufacturer to modify its assays, which are microtiter well plate assays for sweat patch eluates.

The FDA approved the co-dependent submission, meaning that one product depends upon the other. So, really, the product here is a combination of the sweat patch and of the STC EIA assay.

It is a little bit different approach, but that is how the FDA wanted to handle it.

This comes from the publication noted at the bottom here.

What we are looking at is some data for the cocaine assay.

When I say cocaine I mean cocaine, the parent drug, not benzoathenine. That is one of the characteristics of the sweat patch. We detect the parent drug.

That is not to say we can't detect metabolites, but it is primarily the parent drug.

What you are looking at is replicates, namely 72, at each one of these levels, concentrations of cocaine, tested by the EIA STC assay.

They are judged as to whether positive or negative by setting the cut off at these various concentrations.

There are no surprises here. If you were down here at five nanograms per ml, it is kind of hard to get a positive, because you have such a small concentration of analyte. You don't have much analytical signal.

If you go up here around 50 nanograms per ml, there is much more concentrated analyte, much more analytical signal to measure. We can do a better job, precision-wise, accurately, and you get a lot of positives.

We chose a cut off of 10 nanograms per ml, and you have about a 50/50 chance of getting a positive screen at 10 nanograms per ml. So, we are sort of pushing the envelope. But that is okay.

The other thing I should point out is the slope of the lines here are intended to represent the analytical sensitivity of the assay.

In other words, the change in analytical signal per change in analyte concentration. You will notice the slope is greatest in the area of the 10 nanogram per ml cut off, which is where you would want maximum analytical sensitivity, to get a good separation between negatives and positives.

The real reason we chose 10 nanograms per ml for the cocaine sweat patch assay for initial testing is on the next slide.

This is the ROC plot for the EIA assay with cocaine in sweat. I want to point out how to interpret this.

Here on the Y axis we have sensitivity. This is not analytical sensitivity. This is the clinical sensitivity that Dr. Cone alluded to. It is the number of true positives divided by true positives over false negatives.

Down here on the X axis we have one minus specificity. This is clinical specificity defined as the number of true negatives over the number of true negatives plus false positives.

The name of the game is to be up here in the upper left hand quadrant. We would like an assay that is 100 percent sensitive, that detects all the true positives and no false negatives and is also 100 percent specific. It detects all the true negatives and no false positives.

Guess what. In reality you don't find these kinds of assays, whether it be in forensic toxicology or the clinical laboratory.

It is a very hard beast to come by.

So, what we try to do is optimize sensitivity and optimitize specificity. Again, this is not that hard to interpret.

If we chose, say, 50 nanograms per ml as the cut off for the assay, we would have so-so sensitivity. The good news is that our specificity would be 100 percent.

Well, this makes sense. If you have got 50 nanograms per ml of cocaine, by golly, you are going to be a true positive if you test positive at that cut off. By golly, it is going to be kind of hard for you, if you are negative, to get up to a 50 nanogram per ml reading.

This sensitivity, about 72 percent, is not really what we consider ideal. So, we start decreasing the cut off, backing off and going to lower concentrations for a screen cut off.

As you can see, as we do that sensitivity climbs. However, as sensitivity climbs, specificity starts to drop. We lose something. We are faced with the real world situation. We have to make

a compromise. We have to try to optimize the number of true positives we detect, and the number of true negatives we detect.

However, in the process of doing that, we have to give on both ends. So, we are going to detect more and more of the true positives, but we are going to lose some true positives. They will be false negatives.

Likewise, we are going to drop off on our specificity and we are going to have some false positives. Now, we don't like false

positives in forensic drug testing, but we are going to have to accept a few of them in order to optimize both sensitivity and specificity.

That is what we did here. That is why 10 nanograms per ml was chosen as the initial screening assay cut off.

Again, if I could borrow a term from Dr. Cone that I really liked, he mentioned pharmacologically relevant cut off. I like to think that this is a pharmacologically relevant cut off.

Now, where do we get the data? Well, that is down here, 501 specimens from 240 subjects. Some of these individuals are drug free, so their sweat contained no drug. Others were known users and others were controlled administration subjects; the clinical data that, again, Dr. Cone mentioned before in his talk.

So, sensitivity comes out, at 10 nanograms per ml, to be 94.5 percent. Specificity, 99.1 percent. So, again, we give the benefit of the doubt to the individual that is being tested.

We are willing to give up some sensitivity in order to try to eliminate or certainly minimize the number of false positives that we may detect.

Going to the next slide, another publication, this one about amphetamines, we have seen this before for cocaine. Much the same thing.

It just so happens that at about 10 nanograms per ml, we have about a 50/50 probability of screening a specimen positive.

Again, if you visualize the slope of the slide as being the analytical sensitivity, it is maximized up there at about 10 nanograms per ml, which is again where we would want it.

Here is the ROC curve. Again, you choose a high cut off and you are in trouble because your sensitivity is low. But your specificity is good. But you really want higher sensitivity.

So, you start backing off. Your sensitivity increases, but you are giving up specificity again.

If you were to choose five, well, gee, you would have nearly 100 percent sensitivity, which is good, but look, you are somewhere at maybe 80 percent specificity, maybe 20 percent false positives. That is bad.

Again, we just so happened to choose 10 nanograms per ml, because that seemed to do the optimal job for specificity and sensitivity.

The data, again, 229 specimens from 50 subjects, again drug free, users, controlled administration studies.

The sensitivity is not as good as for cocaine, 85 percent roughly. Specificity, 93 percent, not bad. Again, it is a give and take, a compromise situation.

I think what is exciting, though, we didn't pull these values for cut offs out of thin air. These are not arbitrary. They are empirical. They are based upon real clinical data.

A different set of data might suggest a better cut off, but I think we have a reasonable number of data points here.

This is just a diagram of a standard STC microtiter well plate assay. I am not going to spend a lot of time on it.

It is very similar, although the antibodies attach to the wall of the microtiter, to the standard immunoassays we use for urine testing.

So, we add the specimen calibrator control, an enzyme conjugate, incubate. We have competitive protein binding.

As Dr. Costantino indicated, it is the basis of all typical immunoassays in urine testing. We wash away the unbound material. We add a substrate, incubate in the dark for 30 minutes, add a stopping reagent and we measure the color.

It happens to be an inverse relationship in this case, less absorbance, more drug in the specimen. Interpretation, if the absorbance is less than or equal to the cut off, it is positive. If it happens to be greater than the cut off, it is a negative because it is an inverse relationship with the color development.

We have talked about methamphetamine. We have told you already that the cut off is 10 nanograms per ml. If you are interested, horseradish peroxidase happens to be the enzyme that is part of the conjugate. We use benzidine in the assay as the substrate.

We talked about the cocaine assay already, cut off, again, 10 nanograms per ml. Here is the opiate assay. We also chose 10 nanograms per ml as the cut off, on the basis of 215 specimens, 95 subjects. Sensitivity 86.9, somewhat similar to the amphetamines, specificity 92.8, also similar to the amphetamine assay. They are reasonable figures.

Here is the PCP assay. We chose 7.5. Surprise, it is not 10 nanograms per ml. Why did we choose 7.5? Because the data on the ROC plot suggest that 7.5 was the optimal value to choose for sensitivity and specificity.

The 235 specimens, 94 subjects, sensitivity is not as good as with the other assay, 70 percent. The specificity is very good, though, 98.9 percent.

Once again, as with urine testing, we try to give the benefit of the doubt to the individual being tested, try to minimize the number of false positives that the assay may generate.

With all these assays we have calibrators and controls, at zero, at half cut off, at the cut off, and at five times the cut off which, again, mimics pretty much what we see in many urine immunoassays.

THC I didn't mention before. THC is a switch up here. Now we are using -- I want to give DPC credit. Diagnostics Product Corporation, we used their RIA for the sweat patch for THC.

Why don't we use the microtiter well plate from STC? It is a matter of the antibody. The antibody in the DPC assay tends to have better binding with THC.

Once again, it is the THC, not the carboxylic acid metabolite, that we are measuring here. It is the parent active drug that comes out in the sweat.

So, we needed an assay with an antibody that would react preferentially with THC, the parent compound.

We found in the DPC RIA, it is a typical double antibody RIA which I am not going to spend a great deal of time going through the mechanism of it.

The cut off here is five nanograms per ml, not 10 and not 7.5. Again, why is it at five? Because the clinical data of 242 specimens, 71 subjects, indicated that that was the optimum cut off, a pharmacologically relevant cut off.

Sensitivity, 68 percent, specificity 89 percent, maybe not the best, but a reasonable compromise, and we can go well below that. Final slide, this is how we view what we are doing with the

sweat patch. Down at the bottom we have standard urine testing. Urine is the traditional specimen.

We are dealing with an alternative specimen, namely, sweat. Micturition, the way of traditionally collecting a specimen for urine testing, we need something different. We need the sweat patch. It is a more involved device. It is new technology.

Nevertheless, we have immunoassay and GC/MS, the established methodologies, for both.

Here, of course, for urine testing the SAMHSA regulatory guidelines which are mandatory for urine testing, we apply them as best we can for the sweat patch.

The idea is to come up with reliable, scientifically valid, forensically defensible drug testing with an alternative specimen source. Thanks a lot.

(Applause.)

DR. HUESTIS: Thank you, Dr. Armbruster. Our next speaker is Dr. James Towt. He has a doctorate in pharmacology from the University of Medicine and Dentistry of New Jersey in Rutgers University.

He has been with Hoffman-LaRoche for more than 13 years as a research scientist. He is currently the research leader for decentralized testing systems at Roche Diagnostic Systems.

He is going to talk to us today about on-site testing. Dr. Towt.

DR. TOWT: The one good thing about not having a long and distinguished career is that the introductions are mercifully short.

Before I get started with the reagents and procedures for on-site testing, I just would like to address Donna's comment about Bubba.

That is that the testing does not have to be confrontational. You can set up systems where random negatives are sent out, unbeknownst to the people being tested so that the presumptive positives, the people don't know if they are presumptively positive.

Alternatively, if you choose to be confrontational, it has been shown in other testing situations, that these people do want to appear to be smart, typically, and when faced with a positive result, readily admit to using the substances that are being tested for.

The on-site drugs of abuse tests are competitive immunoassays for screening urine samples. The reagents used in these on-site immunoassays consist of the antibodies, drug derivatives and various buffers and stabilizers.

These constituents are similar, or the same as the instrument based assays now being used in the HHS approved drug testing laboratories.

These on-site assay systems are approved by the Food and Drug Administration, as are the assays that are used in the laboratories.

The on-site assays are manufactured under good manufacturing practices established by the FDA. Some of the developers and manufacturers of these on-site assays are, indeed, the same manufacturers making the laboratory-based assay systems.

In preparing this talk I didn't realize that virtually the entire forensic toxicology community would be here, so please bear with my simplistic overview.

As in the laboratory based assay, drug in the sample being tested competes with drug derivative in the on-site assay system for binding to anti-drug antibody in the assay system.

As in the laboratory based reagent systems, you label one of the components of the assay. In the on-site assay systems, either the drug derivative can have a label, or the anti-drug antibody having the label.

In the laboratory based reagents, the label consists of either an enzyme, a microparticle, a fluorescent molecule or a radioactive isotope.

In the laboratory based reagents, these labels are read by an instrument and not by the human eye.

In the on-site assay systems, the label typically consists of an enzyme or a microparticle, and instead of being read by an instrument, it is read visually.

In the on-site assay systems, the reagents are set such that a visual difference is created - this is usually color formation -- between those samples containing drug at or above a minimum concentration -- and those are the SAMHSA established cut off concentrations -- and those samples containing no drug or drug below the minimum concentrations.

Like the laboratory based screening assays, these results are interpreted qualitatively as negative for drug or presumptively positive for drug.

In some of the on-site devices, the drug derivative is labeled, such as here. If the sample does not contain drug analyte, the antibody binds up the label derivative, leaving no label to be detected, indicating a negative result.

On the other hand, if the sample does contain drug, the drug preferentially binds the anti-drug antibody, leaving the labeled drug derivative free to be detected, indicating a positive result, or a presumptively positive sample.

In the alternate format, where the anti-drug antibody is labeled, the drug derivative is typically immobilized on some support system.

In this case, when there is no drug in the sample, the anti-drug antibody, which has the label, is bound to the immobilized drug derivative, producing a color formation and indicating a negative sample.

In the case where the sample does contain the drug, again the drug preferentially binds the labeled antibody, preventing it from binding the immobilized drug derivative. In this case, no color is formed and it is indicative of a positive sample.

Now, talking about the on-site testing procedures, you have heard about the chain of custody and all the documentation that is necessary.

Compared to the training needed for the documentation and controlling all that paperwork and sample integrity, the actual training and performing and interpreting of on-site testing is very simple.

The procedure for on-site testing typically involves introducing an aliquot of the collected urine sample into the test device containing the reagents.

After waiting approximately five to ten minutes, a visual cue appears, which Dr. Cone referred to as an internal control.

This indicates that the reagents have run and results may be interpreted.

The urine sample is determined to be negative or presumptively positive for the drug analytes by the presence or absence of a visual label. This visual label shows up in a results window.

Presumptive positives would be sent on to a laboratory for confirmatory testing, where negative specimens could be immediately disposed of.

Since the reagents of the on-site tests make use of the same immunochemical reactions used by the laboratory based tests, they screen for the same analytes as the laboratory based tests.

The reagents are set to detect the same level of drug in the urine as the laboratory based tests. These again are the SAMHSA established cut off concentrations shown in the slide.

Likewise, the units of measure are nanogram per milliliter, which are the same as the currently approved testing methods.

Part of the FDA approval process for the on-site drug test is the demonstration of the specificity and the sensitivity of the test. Just as in the HHS laboratory-based assays, the specificity

is determined by testing individual urine samples spiked to contain known quantities of related and unrelated drug compounds.

Sensitivity is similarly determined by testing individual urine samples containing known concentrations of the analyte to be tested, and actual clinical specimens.

Performance of the on-site tests are compared to the laboratory-based tests by testing clinical specimens using both methods. The results of the specificity testing and sensitivity

testing are reported in the manufacturer's package inserts.

Representative copies of the on-site testing inserts were available, but I understand they were not put into the book.

In addition, there are a number of toxicologist and noted clinical scientists. I think most are here in the audience, who have evaluated on-site devices and have published the results in peer reviewed journal articles.

Just as a summary, I would like to say that the on-site tests are based on similar competitive immunochemical reactions as the laboratory based reagents now being used.

They make use of the same antibody drug derivative interactions. They test for the same drug analytes at the same concentrations and they have similar specificities and sensitivities as the laboratory-based reagents. Thank you.

(Applause.)

DR. HUESTIS: Thank you, Dr. Towt. Our next speaker is Dr.

R. Sam Niedbala. He is the executive vice president and founder of STC Technologies in Bethlehem, Pennsylvania.

Dr. Niedbala has a masters in clinical chemistry and a doctorate in clinical chemistry from Lehigh University.

At STC Dr. Niedbala's responsibilities include management of all diagnostic products and business activities.

STC develops, manufacturers and markets proprietary immunodiagnostic products, capable of detecting substances in various body fluids, for use in commercial labs, physician offices, hospitals and on-site testing.

Dr. Niedbala has a number of patents and has authored over 31 publications on various topics. Dr. Niedbala, please.

DR. NIEDBALA: Thanks, Marilyn. Let me just start with a little bit of an introduction. Marilyn had given you a background on STC.

I think it is important to note that part of our mission has been to target the development of technologies for use in detecting drugs of abuse in alternate fluids.

Mike Peat had earlier talked about codeine and cocaine in saliva, and Dr. Armbruster has already talked about some of our work in contribution to the sweat patch.

As a company we have watched saliva testing -- and that is in particular what I am focusing on now -- change over the last few years.

In the early 1990s, there was a meeting down in Florida specifically to address the use of saliva as a diagnostic fluid. There were some very simple conclusions that had come out of that.

Since that time there has been a tremendous amount of work using this fluid, not necessarily with drugs of abuse.

As I was thinking about the meeting and preparing, I felt like the junior member of the alternative fluids we were looking at, in terms of the amount of data and publication, although it has been talked about for years.

There have been an awful lot of publications with some of the other fluids being discussed. Then I stepped back one more step removed, and I thought about all the work and publications out now using saliva or oral fluids for things like HIV.

Over the counter, we are looking at people like Smith-Kline, who have bought significant equity stakes in companies like Epitope, with a future goal -- and we can all just wait and see when this happens -- of bringing saliva testing out to the general population as well as the lab population.

I think the opportunity is here for saliva as well as some of the other fluids. To go back to our T shirts back at that meeting in Florida, and taking a quote from that, truly, spit happens.

I think we have got some significant things to look forward to with this fluid. I am sorry, after lunch I have got to watch half of you dozing off, and I had to throw something in here, to lighten this back up.

What is saliva? I think we need to start at this point.

There are some significant yet discrete differences, as we look at this fluid.

Really, whole saliva is a combination of several fluids from three different glands, some major, some minor -- the parotid, the sublingual, and the submandibular.

As you know, when you get up to make a speech such as this, your mouth goes dry. Obviously what is left is more the mucous materials, versus some of the parotid fluids, which may feel real comfortable at cocktails later on today.

However, when we think about what we are measuring, I really think it comes down to the bottom here, which is mucosal transudate, which is really the fluid that permeates back into the oral cavity from the blood.

This is where I believe we need to focus our efforts, as Dr. Peat had said earlier.

If we look at saliva as a fluid for testing, we see that saliva has several advantages. In this particular case, I am comparing it to urine alone.

Collection of saliva is easy. It is very simple to walk

into a room with someone and witness that collection.

The devices, as you will see noted in a few minutes, vary in terms of how they work. It is very easy to imagine sort of a lollipop stick with an absorbent pad on it, put into your mouth, witnessed, taken out, put into a container and mailed back to the laboratory for analysis.

This is really the stage of testing that saliva is at right now. There have been publications about those working on on-site tests for saliva, and I believe that those will come with time.

However, as we step through technological development, the simplest, most straightforward way to start is with a lab-based test.

So, as you look at collection, you can envision this being very easy.

Chain of custody, again I think that falls into an easier category versus some of the things we have to build in for urine testing.

Safety is very high with saliva. There have been many, many reports looking for, for example, HIV transmission with oral fluids. So, the safety of saliva itself is very good.

Training, I think, in both cases I had graded as low. It has been said over and over again, if you have a good protocol, if you follow it everything always works.

It is usually when we push the validation of that into people's hands who may not follow the protocol, how much better is it to work with one fluid versus another is what we would have to consider.

Overhead costs for saliva is relatively low versus urine, but I do have to say this. I forget who it was earlier who talked about the simplicity of a urine cup, which costs maybe a few pennies.

The people who make saliva devices have not yet quite caught up in competition. So, the cost of these particular devices is still fairly high.

Their rationale for that is that the overheads is not as high. Therefore, I believe they think they can charge more for these particular types of devices.

As we look at saliva as a fluid -- I think our meeting on saliva years ago had concluded this fairly consistently -- is that saliva is fairly useful when you want to qualitatively identify the presence or absence of materials such as drugs of abuse, hormones, certain proteins, infectious diseases.

The fluid has been looked at for things like therapeutic drug monitoring, but there are variations. In some cases we suspect that, because of the PKA of drugs, we need to consider this fluid very carefully as we walk through the analysis and analytical evaluation of any screening methodology.

If we look at requirements for all the immunoassays in alternative fluids, saliva is a great example for us to start with. First of all, we need to have again the right cross

reactivity of our antibodies with the analytes or metabolites found in that specimen.

As has been said already, in many cases the parent drug is found in very high concentrations in some of these alternate fluids, saliva being one where this is true.

GC mass spec absolutely has to be the gold standard for us to follow up with confirmation. Again, our cut offs must be set based on the best sensitivity, specificity and efficiency, and the ROC analysis that was discussed earlier for sweat, I think, clearly is appropriate for this as well.

Finally, we need to do this within the populations that we are looking at. The predictive value needs to be tied into how we evaluate the testing methodology.

If we look at some of the initial test reagents and procedures, the possible menu includes the common list of drugs of abuse that we are very familiar with and Mike Peat had mentioned earlier.

Codeine is an additional one that is analyzed very often in saliva. As many of you know, we have been involved as a company with alcohol testing, not immuno based obviously, but very much in practice with saliva.

Again, our key factors as we look at saliva for these types of analytes, is that we must target the metabolite or parent.

We must then select the correct antibody or develop it. We need to keep in mind, as we look at the test, what are the differences here.

I think that theme is coming out more and more as we go through today, and I am sure tomorrow, in that different fluids provide different levels or different qualities of information, which I think can be very, very useful when we look at time since exposure, time since dose, under the influence situations. This is where saliva is very different from some of the other fluids.

Finally, the sensitivity of the screening method that is being used has to be considered. In many cases, my general experience has been that as we compare to the urine cut offs, these alternative fluids, we must be somewhere between one tenth and one thirtieth the cut off that are currently used with urine.

So, it is a challenge for us as far as screening technologies are concerned. Our job with that screening technology is to, as I term is, interrogate these specimens accurately and quickly.

What was discussed earlier with the sweat patch is that in presenting to the FDA for clearance of these products -- I am trying not to use the term approved, but rather for the clearance of the products - what was determined through our experience there is that there must be a marriage between a collection device and a screening device such that they become a system.

There are different collection devices for saliva that I have seen over the last several years. They really do run the gamut.

If you go back into the literature you will see things like dental cotton simply packed in the mouth and then extracted.

In recent years we can look at devices such as the Epitope device such as Dr. Peat had shown earlier. Saliva Diagnostics is another one that you generally do not find within the United States, but does have an FDA clearance for sale as a saliva collection device alone.

There is no marriage with a screening test. So, its utility is really limited.

Then most recently there is a new company, a new player called Avatar. Their particular collection device I have not tested or seen.

In addition to these collection devices onto pads, there is the idea of collecting whole saliva, which is essentially collected into a tube, stored, and then must be diluted or processed somehow, from Sarstadt, called the Salivette. Some of you may be familiar with it.

This is an overview of the different collection devices that are available. None of them at this point in time have any clearances from the FDA for use with drugs of abuse testing, but I can tell you that this will be coming.

If we look at now, after collection, the procedures for testing, we have talked earlier about different types of competitive EIA.

Again, based on the selection of antibodies, I put possibly RIA or EIA. Of course, we could go back to more manual methods of gas chromatography, mass spectroscopy, HPLC or GC.

If I give you one example from one of the microtiter plate assays which we are using for sweat as well as saliva testing, you will see that if we compare cross reactivities, what we have done here is develop antibodies that will pretty much evenly cross react between benzoylecgonine and cocaine. In addition, it has some very good cross reactivity with cocaethylene.

Methodologically, this is a very simple test, in that sample and reagent are mixed and incubated for 30 minutes. There is a wash step, addition of substrate 30 minutes later, and absorbance is measured.

In the years that we have been doing this, we have developed instrumentation and software which can screen very high volumes using this method.

So, in summary, the last point I wanted to make is that, for saliva we are at kind of a cross roads with this particular fluid.

We need to marry the collection with the testing. If we look at saliva as a fluid, we know that unstimulated, we can collect about .32 mls per minute. Then daily we are somewhere between 1.5 liters would be available. When I said spit happens, it is available.

So, if we look at the steps -- and this would be using the OraSure device from Epitope now called Episcreen -- step one would be to collector in the mouth.

Number two would be to collect a saturated lollipop in this particular case, place it into a preservative fluid, seal it, send it to the lab. That fluid is then removed and analyzed.

It is very easy in terms of processing and, again, I think generationally this is where we are with saliva testing to date. Thank you.

(Applause.)

DR. HUESTIS: Thank you, Dr. Niedbala. Our last speaker today is Dr. Verner A. Baumgartner. Dr. Baumgartner received his PhD in physical chemistry from the University of New South Wales, Sydney, Australia.

He has been associated with the Department of Veterans Affairs since 1967, where he has held a variety of research positions.

He is the founder of Psychemedics Corporation and is current scientific director.

Dr. Baumgartner's main research interest over the past 20 years has been the analysis of hair for licit and illicit drugs. Dr. Baumgartner today is going to talk to us about initial testing with hair as a specimen.

DR. BAUMGARTNER: Given our time constraints, my main objective in this talk today is to discuss the clinical specificities and the clinical sensitivities of the RIA assays.

That is, how cut off were defined so that hair analysis is safe against interpretive false positive due to internal passive drug exposure.

In other words, I am talking about Ed Cone's alligators and tigers mostly.

As the problem of urinalysis has shown, and the more recent studies by Dr. Cone that the passive ingestion of one milligram of cocaine can cause positive urinalysis results, a real problem for a nondrug using spouse of a drug user, a passive internal drug exposures is certainly one of the main forensic issues of drug testing.

It is therefore important to establish how hair analysis stands on this critical matter.

Our safety measures, again, interpreted false positives due to external contamination will be discussed in the talk on data interpretation.

I want to make it very clear that specificity cannot simply be defined in terms of mass spectrometry but in terms of passive

internal exposure and passive external exposure.

Conversely, we don't want to make the test so safe against internal passive drug exposure that it only identifies heavy drug user.

Consequently, our hair analysis balances clinical specificity against clinical sensitivity that will also be discussed in my talk.

These two opposing test parameters are defined by the well known relationship shown in the next slide.

However, before covering these topics, let me briefly put you in the picture of how we do the hair assay.

We use a three-month detection window, which we access with a one-and-a-half inch of the most recently grown hair cut close to the scalp.

All our variations are produced by Psychemedics. Hair is extensively washed -- and I will talk about it in great length later in my second talk. The efficiency of washing is evaluated by wash kinetic procedures.

Drugs are liberated by enzymatic digestion of the hair specimen. A patent has been granted for the performance of the EIA on the hair digests.

Before digesting, the melanin fraction is removed by centrifugation to avoid the possibility of any hair color or racial bias effects, in case any special binding to melanin should occur.

Independent blind field studies at the University of Southern Florida with 315 African Americans and 846 Caucasian arestees have demonstrated the absence of hair color racial bias effects by our methods.

That is, self reported urinalysis and hair analysis data all indicated that the African American users had twice as many cocaine users as the Caucasian group.

Except for marijuana, RIAs are performed with one milligram hair samples. With marijuana we use four milligrams. The detection limits are set in three standard deviations with the matrix effect being the zero values of negative populations.

Matrix effects are minimized by removing interfering substances by washing and keeping a constant protein concentration in the hair digest.

Except for marijuana, no major cross reactivity effects occur. Cross reactivity effects which occur for essentially all nonmarijuana users are removed by a patented procedure.

The RIA assay utilizes an uncharacterized marijuana analyte. Positive RIA screens for marijuana are confirmed by the measurement of carboxy THC by a metabolite which does not occur in marijuana smoke.

My paper written for this conference describes all the above and the other aspects of my talk in greater detail.

As it turns out, the three standard deviation detection limit of RIA is at a much lower drug concentration than can be obtained by forensic quality mass spectrometry.

This is shown in the next slid by an order of almost 10. The question arises, therefore, what is the quantity of drug that, upon ingestion, will produce drug levels in hair which correspond to the cut off levels defined by the limit of quantitation of forensic quality mass spectrometry.

Obviously this value differs for different individuals due to biochemical individuality effects.

With respect to this question, hair testing has a tremendous advantage over urine, blood and saliva testing, meaning the setting of cut off levels.

That is the kinetically non-fluctuating nature of the analytes in a hair specimen.

The method used for ascertaining the minimum drug dose is based on the well documented fact that within the same individual, there is -- the individual is his own control -- the concentration of drug and hair is directly proportional to the ingested dose.

This linear relationship is clearly illustrated in this slide. It was obtained with C57Bl6 mice, very black mice. They are identical twins, essentially, and you see this in the melanin.

Of course, the experimental demonstration of this relationship requires effective wash and extraction procedures.

It should also be noted that this experimentally verified relationship is expected on theoretical grounds, and it applies also to sweat patch testing, where it has also been experimentally verified.

The minimum detectable drug dose giving rise to cut off level drug concentration in a particular individual's hair can be calculated from the linear dose relationship, and the individual self report of the amount of drug used, and the measured drug concentration found in the individual's hair for the reported drug dose.

These calculations are then performed on a population of drug users with self reporting. The average drug dose corresponding to the mass spectrometric cut off level is then calculated for the population using a particular drug.

What such data looks like is here. On the bottom is what the self report is of use, and on the Y axis is how much was found in hair.

By way of illustration, how the minimal detectable dose cut off concentration is determined, let me select three individuals, C, D and E, from the previous slide.

Let us start with the individual C. OC1, the line OC1, represents the experimentally verified linear dose correlation.

The amount of drugs that individual C would have to have ingested to produce drug levels in hair equalling the cut off value is determined by the perpendicular from the point where line OC intersects with the cut off line, AB; that is, point C3.

Of course, this diagram is nothing more than a graphic illustration of the very similar proportional calculation on the basis of the linear dose relationship.

The different minimum drug doses, C3, D3 and E3 are obviously caused by biochemical individuality effects.

The average value of the minimum drug dose for the different drugs is shown in the next slide. These are for cocaine, 264 cocaine per month of cocaine. For morphine, 1.9 bag per week, for marijuana .5 joints a week and for PCP, .44 sherms per week, and for meth 330 milligrams.

These values clearly indicate that hair analysis results are safe against passive internal exposure to drugs.

It is important to note that this evaluation is quite independent of whether there is a correlation between drug dose and hair levels existing in the population of drug users.

These evaluations of the safety of the hair cut off levels was subjected to further independent testing. This was done with the most sensitive assay there is with cocaine where cut off levels are closest to passive drug exposure levels.

One such blind study performed at the University of Southern Florida involved 50 under cover policemen who were extensively exposed to cocaine as part of their drug detection work.

The negative results of this study clearly shows that passive internal exposure to cocaine is not a problem for hair analysis, and this has been published in the February issue of the Forensics International Journal.

In another evaluation, we and Dr. Sachs independently performed studies with poppy seeds. Dr. Sachs studied the ingestion of 250 grams of poppy seeds containing 173 milligrams of morphine per kilogram of poppy seed. It produced no interpretive false positive test results in his test subjects.

All these studies referred to in my talk, of course, have been published.

Clinical sensitivity. The question arises, how is the exceptional safety of hair testing against passive internal drug exposure related to the clinical sensitivity that is the detection efficiency of the test.

Although clinical sensitivity can be accurately measured by controlled dose studies, when therapeutic agents are involved part of the controlled dose studies, in contrast to negative controlled dose studies, do not provide an accurate measure of the clinical sensitivity of an assay in toxic illegal substances such as drugs of abuse that are being investigated.

The reason for this is that the controlled dose studies, if they can be applied at -- for example, in the case of PCP -- can only cover very low -- that is, non-toxic dose ranges -- in short time periods.

All the controlled dose studies are most useful only in the evaluation of hypersensitive tests such as urinalysis; that is, tests that are at risks of producing interpretive false positives due to passive internal drug exposure.

Certainly controlled dose studies provide no information on the effects that invasive tactics have on the clinical sensitivity of forensic tests.

Because of these problems, we decided to evaluate clinical sensitivities of hair testing by two different and more effective approaches.

One involves blind studies where the detection rate of hair analysis is compared to self report of drug use.

Naturally, self reports were obtained under non-punitive research conditions.

The second approach consists in blind side by side comparisons of the detection efficiency of hair and urine testing.

We proposed the term relative sensitivity studies for this latter approach.

The relative sensitivity studies were performed under three conditions. One was where evasive tactics against urinalysis could not be applied -- e.g. with arrestee populations.

Two, under conditions where evasive tactics against urinalysis were not easily applied -- e.g., unannounced observed urine testing of parolees.

Third, under conditions where evasion of the urine testing was easy, e.g., pre-employment testing.

The relative sensitivity studies would be illustrated with marijuana and cocaine assays. For doing this we compared the most challenging hair test -- that is, the marijuana assay -- to the most effective urine test.

On the other extreme of the testing spectrum, we have the most effective hair test, cocaine, compared to the urine test, whose three-day detection window is considerably shorter than that for marijuana.

Let me illustrate the self report method with a blind study that we have performed with Dr. Tashkin at UCLA. Here we see that the very light marijuana users, taking between .15 to .85 joints per week, were detected with an efficiency of 29 percent by RIA.

Those using one to three joints per week were detected with 89 percent efficiency by RIA and mass spec. All the three heavier use categories were detected with 100 percent efficiency.

The overall sensitivity for the marijuana assay was 88 percent. The detection efficiency or clinical sensitivities of the other assays are shown in the next slide.

They are, for cocaine 88 percent, opiates 83 percent -- they are 83 percent for PCP, 88 for opiates and 96 percent for cocaine.

A blind study evaluating the clinical sensitivities of the meth amphetamine assays is currently in progress.

With respect to the relative sensitivity studies we see, as we expect, that as the detection opportunities increase, that you get worse results for urine.

So, in the case of marijuana, the marijuana assay in arrestee populations is about the same as for urine. It is 35 percent better under moderate evasive conditions, under probation conditions. Under applicants it is 600 percent more efficient, meaning the hair assay detection.

When it comes to cocaine where no evasion is possible, 92 percent is the advantage of cocaine. With moderate evasion, 272 percent

and with applicant it is 1,580 percent. Thank you.

(Applause.)

DR. HUESTIS: Thank you, Dr. Baumgartner. We will now have the panel available for questions from the Drug Testing Advisory Board and other panel members.

DR. WILKINS: Actually, I will just ask one question and then give up the microphone and then probably come back for some more later, if that is okay.

I have one question for Dr. Baumgartner. Several times in your talk you mentioned that you had literature references for some of the data that you were providing.

I wondered if it would be possible if you could give us a list of those so that we could then go and collect those and review those at our leisure.

There was a lot of data to look at in a very short period of time and I think it would help us review it appropriately if we could just have the names of the articles and then we could go find them.

DR. BAUMGARTNER: The literature references can be found in a review article that I wrote. It was in the book on hair testing edited by -

DR. WILKINS: Is this the KIMS Drug Testing and Hair?

DR. BAUMGARTNER: Yes, that is right. It is the CRC thing

that came out about a year ago.

DR. WILKINS: And the data presented in the tables are also referenced in there?

DR. BAUMGARTNER: It is obtained from there, yes.

DR. KWONG: I have a question for Dr. Niedbala about saliva.

If you call oral fluid and saliva a mixture of secretion from three different glands and the constituents could vary with stimulation and flow, and there are a number of different devices out there, has anyone looked at whether different devices can give you a different saliva composition and whether that would affect you clinical outcome?

DR. NIEDBALA: I can tell you, we have done some work with saliva, for instance, that has been collected from individuals dosed with -- I will use cocaine as an example.

We have done both random studies as well as controlled dose studies. The results appear to be similar.

There is a lot of work that needs to be done yet looking at single dose versus chronic dose with every one of these drugs of abuse.

With time we expect to do that, one analyte at a time. So, for the moment I have to say yes, at least for a few of the devices. Mostly whole saliva versus the Epitope device versus the

Saliva Diagnostics devices. But as far as Avatar or dental cotton, I have not done studies to compare all of those.

DR. KWONG: A related question on collection. Analyte concentrations are lower than what we are familiar with in urine. You mentioned that your GC/MS confirmation procedure will be more sensitive, too.

My question is, are there any limitations on how much those devices can collect to give you a sufficient volume?

DR. NIEDBALA: Yes, generally it is somewhere between -- again, I will use Epitope because I have the most experience with it.

You will collect saliva, a few hundred microliters, and then put that into a test tube which has a diluent. The diluent has preservatives, some detergent, some colorant.

What you will get back at the lab is somewhere between I believe 1.2 and 1.5 mls.

So, the challenge is not only in the screening test where you need to use the minimum amount of fluid to perform your panel of tests, but also in your confirmation. I know Tim Rohrig will be addressing some of that in the next presentation.

DR. CONE: For the saliva and sweat testing, one of the problematic things that I think you face is the simplicity that toxicologists like myself like to think in terms of, like expressions of cut offs and things in terms of concentrations of nanograms per something.

In your case you are saying, I think in some respects, it is nanograms per device. That is a tough one, I think, in terms of setting realistic cut offs, especially as more products proliferate and you want to do cross comparisons between them.

I think you need a much better standard for setting cut offs than nanograms per device.

DR. ARMBRUSTER: Dr. Cone, when you say nanograms per device, we do use nanograms per ml. I guess I am missing your --

DR. CONE: But that is not milliliter of sweat. That is milliliter of solution you add.

DR. ARMBRUSTER: That is a good point. Unfortunately, with only 15 minutes, I have had to delimitate the discussion of the elution of the sweat off the sweat patch. You are raising a good point there.

I guess what you are getting at is that we follow our protocol and somebody with a different protocol for a different sweat collection device, it may be a matter of comparing apples and oranges.

DR. CONE: That is part of it. Inherent in the device that you are working with, of course, is that you don't have a volume.

I think this is something to think about and to struggle with. I don't know what the immediate solution is. I just know that it is problematic when you start comparing nanograms per your device versus nanograms per somebody else's device. The same is true for saliva as well.

DR. ARMBRUSTER: Just as spit happens, sweat happens. We don't all sweat the same amount. It is true. Thus, with our device, if you are a heavy sweater, and you are also a drug user, presumably you will accumulate more drug, et cetera. So, it will be easier to detect an individual that is a drug user and a sweater.

DR. NIEDBALA: I think there is one other point that goes along with that and I had written it down and forgot to talk about it, which is some sort of testing to make sure that the sample is adequate.

With urine we do pH, creatinine, specific gravity. We have choices to make sure that we have a valid specimen. With saliva, we can look at things like human IgG, which is a simple test to do, especially in a microtiter plate format, to make sure that you have a specimen that is valid, that has collected a sufficient amount of material.

Actually in the literature there are a few publications, and saliva in particular, which deal with what is the amount of IgG that you would normally find in saliva.

DR. CONE: Sort of like a creatinine correction.

DR. NIEDBALA: Yes.

DR. ARMBRUSTER: And for sweat at one time we were looking at lactic acid byproducts.

DR. CONE: A comment for Dr. Baumgartner and perhaps a response from him. Your description of how you cite cut offs does not give me a warm, fuzzy feeling.

The use of self report data is one way of approaching the problem, I suspect, but it is not a very satisfying approach.

I do think that you need controlled dosing studies to evaluate what is the time course of detection, individual variability response and so forth.

Your self report data doesn't do everything that it needs to do, I think, and let me just tell you about some personal experiences with drug addicts over the years.

We have had documented cases of people in treatment who we were monitoring. There were no punitive actions taken as a result of them telling us the truth.

We consistently got people who just simply lied for apparently the fun of it. We have a classic example -- this is not an isolated example. It happens many times.

I had a lady in treatment who was tested three times a week over a period of six months. We had a total of several hundred positive cocaine specimens.

Each and every time she was asked she always said no. I guess she was following somebody's advice. But it is just not a satisfying approach to setting a cut off, to be asking people and correlating what is asked.

DR. BAUMGARTNER: Yes, I would certainly like to add to this, Ed, first of all, we have correlations within populations. So, we do pick out outliers.

We also can get self reports which you can rarely get when you do controlled dose studies.

Thirdly, with the hair analysis you notice that we are not a hypersensitive test. I think ethical considerations would totally prohibit giving people dosing anywhere near our cut off level. Those are three considerations.

DR. JONES: A couple of questions to you, Dr. Armbruster.

Were the data in figure one and three generated with the same microtiter plate, no changes in the assay?

The assay was there and you were just varying concentrations of analyte in the specimens?

DR. ARMBRUSTER: Yes, that is something that I wanted to bring out, as we get into the confirmatory assay. All of those specimens were tested by GC/MS, the gold standard for reference.

Thus, they were determined to be positive or negative based on GC/MS. So, we knew the GC/MS values. Then we could assign cut off points.

DR. JONES: That is what I assumed. I just wanted to clarify. Could you clarify for me what the analytes were as you go from class to class of drugs in those?

DR. ARMBRUSTER: Yes, for cocaine it was cocaine. For meth amphetamine, it was meth amphetamine. For the amphetamines, it was meth amphetamine. PCP was PCP. THC was THC.

For the opiates, if I recall correctly, it was heroin and morphine, so not just one but data from both analytes.

DR. JONES: Thank you. A quick question from you, Sam.

What do you know about adulterants in saliva. I pop a Certs in my mouth or I throw a piece of candy in or something during that collection.

DR. NIEDBALA: For adulterants in saliva, actually the way I deal with that with us is all the experience we have had with saliva alcohol.

The QED on site test was the first one cleared by the DOT.

We had to answer that question 10 different ways. That is a different technology, so I have to throw in that caveat.

Knowing what can, in that particular case, affect that technology, all we had to do were the experiments to design a protocol by which people waited before their specimen was collected.

In that particular case, it is 10 minutes after they had drunk or eaten anything. After that period of time, we were very safe in collecting the specimen and getting an accurate result.

Also, in the QED technology, if we did allow them to take the specimen prematurely, it always caused a false positive. Never did we have a false negative.

I thought that was rather interesting. In the bigger picture, especially with alcohol, what we are trying to do is keep people from getting into trouble by hurting other people or themselves, in many situations.

What that did is, it gave us a little bit of confidence that we could not find a way to cause a false negative.

With immunoassay, because of the technology that we chose to do the alternate fluids is heterogeneous, meaning that there is a separation step.

In many cases, any adulterants that have been tried -- with the Epitope device there has been anything from soda to foods to maybe 20 different adulterants that we tested.

We believe that because we used a heterogeneous technique -- meaning, we wash it before we add in the color development step -- we remove the effects of the adulterants.

DR. JONES: I wanted to ask from a one liner from each of the panelists, if I may. As we are looking at the potential of multiple immunoassays for multiple devices, what are your feelings on attempts to standardize the responses of these immunoassay based systems in terms of

their ability to discriminate.

I realize that we have got cut off issues, and that is not what I am asking you about. What I am asking you about is the ability of these immunoassays to discriminate reliably between a positive and a negative at or around whatever that cut off may be.

Should we try to establish standards there or how should we respond to those types of questions in the various matrices that we have.

We have got something from your end, so Tony, you are off the hook in one sense.

DR. ARMBRUSTER: I think we have work to do in sweat testing. We have tests that are FDA approved. We can use them. I think we can probably further improve the discriminatory ability of our assays.

I am always concerned with being able to definitely get enough signal generated for change in analyte concentration that we have clear separation from a negative and a positive.

That is the same thing we have in urine now. When we get into trouble, it is when that assay collapses and we don't have that big leap between negative and positive.

It is going to be a trial and error process, I think, and I am sure we can do better and we need to keep working in that direction. DR. TOWT: Since we are on the urine based screen, we would

be very similar to that. Obviously it is important to establish concentrations that are meaningful.

Again, it depends on the purpose of your screen. Is it to catch everyone or is it to be very safe and not produce any false positives.

DR. NIEDBALA: If I had to write a statement or something I would vote for something like the phrase of free market interrogation of specimens based on good, scientific studies.

DR. BAUMGARTNER: My point is that the radio immunoassay screens that we are using are by far more sensitive than the mass spectrometry that we use at the cut off.

We had absolutely no problem identifying drug users. But the key issue is to confirm them with forensic certainty.

So, in one respect the immunoassay serves really nothing more than a commercial purpose. You have some cross reactivity just on confirming.

Our assays, as you notice, are far more sensitive. Of course, we only use ultra sensitive mass spectrometry. It is just that we were lucky to have ultra ultra sensitive radio immunoassays using one milligram of hair. We can use two or three milligrams and it makes it even more sensitive.

DR. WALSH: I have a quick question for Dr. Baumgartner about how you characterized the clinical sensitivity of the assays.

The way I interpreted the slide is that if I kept my heroin use down below 1.9 bags a week and cocaine to less than 250 milligrams per month, that I would get a negative result?

DR. BAUMGARTNER: Correct. In spite of that, this hair identifies so many more drug users because you have the big gaps in the urine tests being randomly supplied. It just simply misses the target.

DR. PEAT: I have a practical question, and Tony is off the hook on this one, too. Obviously the trend in the drug lab is to run thousands of specimens today for cost effective, economic reasons.

What are the cost effective ways of automating the sweat patch, saliva and hair testing?

DR. ARMBRUSTER: In automating the screening?

DR. PEAT: Yes, in automating the immunoassay techniques. DR. ARMBRUSTER: Currently with the microtiter well plates it is a 96 place well. You are dealing with a fairly large number of specimens that you are testing at once.

With RIA you are dealing with large batch sizes, too, theoretically, depending on how many you have.

Automation that we use includes automated pipetting devices such as T-CANS. We could further attempt to automate some of those processes.

Already with T-CANS there is a minimum of manual testing that is going on. The biggest thing is physically removing the plates from the T-CAN and sitting it on a counter or in a drawer to incubate in the dark.

DR. NIEDBALA: My experience also comes because STC deals in the blood bank and plasma industry, where microtiter is very readily used.

We have technology that we use and instrumentation systems with software and hardware that is FDA cleared that can test up to 1,200 specimens an hour, to give you an idea of through put.

DR. BAUMGARTNER: Our procedures are similarly automated, meaning that we use automated pipetting. That is really all that we can do, because we have to centrifuge and so on, like most competitive immunoassays using double antibody methods.

DR. SELAVKA: A lot has been said today about pharmacologic relevance. That is the first time I really heard that applied, that it is a necessity.

I think if we look historically at urine drug testing, there is some -- I guess I will put Tony and Dr. Towt on the spot on this as opposed to the other matrices.

Can you tell me something about, or I challenge you to describe the pharmacologic relevance of PCP at 25 nanograms per ml in urine or amphetamines at 1,000.

DR. COSTANTINO: While I am not all that familiar with the historical determination of the PCP cut off at 25, we do know that any PCP at all is important.

I don't know about the enzyme immunoassay technology as far as how it has been challenged with regard to sensitivity. I think that is one assay where any sensitivity that is accurate and precise will be useful.

DR. TOWT: My understanding is that some of these cut offs were based on what could be detected at the time they were made.

Then as far as amphetamine, you had all the broad over the counter amphetamine related products that were pharmacologically relevant.

So, you wanted to make sure you didn't screen for those particular compounds.

DR. SELAVKA: I guess just to be clear on the answers and the question I am asking, it seems to me that if you use ROC plots, those are affected by the matrix in which you do it, the analyte package in that matrix, and in fact the immunoassay that you choose and its titer.

Every time you do one of these ROC based cut off selections, every other laboratory would either have to use the same cut offs -- every other laboratory would have to follow the procedure with the same analyte package in mind and the same matrix with the same titer on the same immunoassay or those cut offs will not carry across different laboratories.

You actually have to have standardization. I think that is maybe what Dr. Jones was trying to bring up in his question.

DR. CAPLAN: One semi-announcement on behalf of the board, is that those who have given presentations today, if you have not made your slides or text available, please do so. It is going to be very difficult to follow some of the things that were suggested without having the opportunity to look at the figure again.

I just had a couple of quick other things that were not clear to me. One with sweat. When you were going through those things, it appeared to me that you were looking for THC and BE.

You mentioned cocaine. Are we looking at a mixture of the two types of drugs?

DR. ARMBRUSTER: Are you referring to the ROC curve in particular?

DR. CAPLAN: I am not sure. I think you had it listed on that curve, but when you just answered the question before you said you had -- that the assay response was all parent drugs.

I think the whole list of drugs was parent drugs. Is that the case, or do you find a mixture of drug, metabolite --

DR. ARMBRUSTER: Basically, the major drug that we find in sweat is cocaine, the parent drug. There will be metabolites of cocaine present in the sweat.

The peculiar nature of the sweat is that we do see primarily parent drug rather than metabolite, just the opposite of what we see in urine.

So, the assay is really -- it should be really optimized for cocaine.

DR. CAPLAN: That is a point. The question is whether or not the assays that we currently use need to be optimized for what we might call cross reactants today if we switch to other matrices.

DR. ARMBRUSTER: That is exactly one of the problems we had with the immunoassays, is that they are geared to be optimal for the metabolites that we see in urine.

It is a different picture, a different story in terms of the analytes that we see in sweat. That is why not everybody was willing to jump and develop an immunoassay for us, because they would have to modify their antibodies perhaps.

DR. CAPLAN: I think it is important that we differentiate and understand, since this session is dealing with the initial screening and its capability.

That ability is going to be dictated by the assay and the assays may not all be out there that we need today. We may have far better assays that have greater sensitivity for these analytes that might make some of the products which now have low concentrations more amenable to be useful for a longer period of time, and I don't think we have answered that question.

If anybody else has a comment on that in the group, I would like to hear that. I think it is a fundamental question of apples and oranges.

As we shift technologies and use a particular assay, what is that assay really measuring and can we make assays which will do a better job.

That brings me, if nobody has another comment, back to hair.

I am not clear on -- I know you have made your own antibody mixes and you are talking about various sensitivities compared to an epidemiological identification as opposed to an analytical one.

Can you clarify exactly what you think the THC assay is actually measuring? It was unclear to me from the slide you put up what we are actually detecting when we look at THC.

DR. BAUMGARTNER: With the RIA we are measuring a substance that has not been characterized. It is not carboxy THC in THC because the patented method removes these in order to remove the universally present cross reacting substance.

We have a very high statistical correlation between the positives that we get by RIA and of course what we confirm by MS when we measure carboxy THC.

It is somewhat similar to what is happening in urine. In urine an RIA positive doesn't correlate too well with what the carboxy THC level is that you are obtaining.

As we know, the 500 or so -- 50 or so or the 500 compounds that are present in marijuana have been identified. So, it is a substance that has chemical properties that allow the separation from the cross reacting substance to be made.

DR. CAPLAN: Are you saying that on the one hand you have confirmed a number of these with mass spec, but there is another population that is not confirmed, but only statistically or epidemiologically confirmed with other studies?

DR. BAUMGARTNER: Just like with urinalysis, yes.

DR. CAPLAN: I don't think that is the case with urinalysis. DR. BAUMGARTNER: My understanding is that when you have an RIA positive, that the RIA positive value doesn't particularly well correlate with the carboxy THC levels.

DR. CAPLAN: That is true, but we wouldn't call something a positive unless we had the other confirmation. You are suggesting that if we are to use RIA in hair testing, that we should call something positive without a GC/MS or MS/MS confirmation.

DR. BAUMGARTNER: Oh, absolutely not. Absolutely not. It is simply, we call only those positives that confirm for carboxy THC. DR. CAPLAN: You don't offer that for purposes where it is not confirmed.

DR. BAUMGARTNER: That is right. We don't do that.

DR. PINDER: Referring to the question with saliva, and possible interfering substances, I wonder if you looked into materials that could change the pH of the mouth, perhaps sodium bicarbonate or something even more alkaline, that might affect the concentration of drugs and saliva, and could be held in the saliva in the mouth for quite a period of time.

DR. NIEDBALA: I have actually addressed that issue slightly differently. I don't have enough human studies to have looked at different pHs in different mouths at different points in time.

What I can control is from the moment of collection through the initial testing on to confirmation.

With saliva, during the collection process -- again, I have to use Epitope as the example because I haven't studied each and every one of these collection devices -- the pad has some buffered materials on it that are patented to encourage mucosal transudate collection into the pad.

Then that is placed into a buffered solution. So, by the time the screening assay sees it -- which is really where I have spent a lot of my time -- everything is in a very nice, neat matrix for me.

So, there is really no way exactly to answer your question without I guess going back to some basic physiologic studies.

The PKA calculations that have been done in the past are based, in part, exactly on calculations versus how could we do a controlled study of that. That is never really been done.

I know from some of the clinical -- again, controlled dose as well as randomized studies -- that it doesn't appear to be quite true in all cases, or at least we are not seeing a lot of aberrance that we can't explain in any way, shape or form.

I don't think that the pH of the mouth is really varying so much that there is not enough drug for us to pick up, as far as interrogating those specimens by immunoassay.

DR. PINDER: One question in regard to hair, Dr. Baumgartner. I believe you said that following enzymatic digestion of the hair melanin is removed. I wonder if you could go into that process a little further.

Is the analyte removed during that process? Is the analyte closely associated with the melanin and is some of it lost when the melanin is removed?

DR. BAUMGARTNER: Yes, that is a very interesting question.

First of all, melanin can trap the analyte within the structure, or it binds very strongly to melanin.

Now, if you have a strongly binding interaction, this would not be removed. With Dr. Cone's study that he published, you get loose binding. Dr. Cone didn't wash it off. That is one of the points.

So, the issue of if you have strong binding to melanin within the structure, our method, it does not produce morphological changes to the melanin.

As I said, strong binding is strong binding. In other words, the material is not liberated. It is strongly bound.

DR. PINDER: I am not quite clear on that. Are you saying that in some cases there is strong binding and in other cases there isn't?

DR. BAUMGARTNER: No, we have not observed any strong binding. Dr. Cone has found some binding to it, but it is weak binding.

You can wash it off.

To give you another example, if you take hair samples that have different porosities, you can get 10-fold more when you do soaking studies. You can get 10-fold more material into per unit time however

long you soak it. Then when you apply a wash technique, you remove it. The question is, what do these artificial studies have to do, when you do the washing, with what goes on in the real world.

When we looked at racial bias or hair color bias in the real world, Dr. Michkowski's studies with real populations showed no racial bias effect.

We have to deal with this question really to evaluate it in the real world. We have done that with cocaine.

Therefore, Dr. Cone's data has no bearing at all to the real world. It is an in vitro study and it has some interest and I think Ed probably will agree with me on this.

DR. PINDER: I see.

DR. CONE: Just to add to the comments, we characterize both high affinity and low affinity of melanin to cocaine.

We used classical drug receptor techniques to evaluate the binding characteristics. What we were most interested in was the question Dick Pinder asked.

If drug is bound to melanin, will it be released during the digestion step. We found that 95 percent of the cocaine that is bound to melanin is released in the digestion step to the supernatant, such that we see no value, in essence, in spinning down the pellet particle and saying that it is removing any potential binding as such, because it is released into the supernatant during the digestion step.

I will give you an opportunity to comment on that if you want, but I would like to ask you a different question.

DR. BAUMGARTNER: The release on the digestion step that you pointed out, first of all, your high affinity was low. I think your paper showed non-specific binding and some binding. You didn't develop any equilibrium constants on this.

Secondly, you did an adsorption study on the surface. We are talking about the trapping and of course you have got to come up with an equilibrium constant.

Then of course the final proof of the pudding is the real world studies with these 1,200 samples, and not seeing -- getting the same results for urinalysis, then hair analysis.

I think that is the final answer to the issue, you know, when these three different approaches show nothing is going on between racial groups with cocaine, I think that settles the matter, no matter what in vitro studies and sophistications we can play.

DR. CONE: Just to follow up, I would like to ask you a different question. I am really confused. I thought I understood what you said, but I believe you were describing the marijuana screening assay.

I thought I picked up, and then I thought you confirmed that you remove the carboxy acid metabolite during your washing step?

DR. BAUMGARTNER: No. In order to remove the interfering cross reacting substance that is universally present in hair, in the removal process, in the screening acid, this THC and carboxy THC go.

We can't come up with methods that remove the two interfering substance with the other. The reason is that it is a lipophilic(?) substance.

DR. CONE: My real question is, what is the unknown substance. Did I not understand you to say that in the screening assay for THC that you do not know what you are testing for, or that you have characterized it but you haven't said what it is.

DR. BAUMGARTNER: No, when I say an unknown substance tautologically it means an unknown substance is an unknown substance, of which there are many, as you know.

The point about this substance is that it has a high affinity for the carboxy THC and THC antibody that we are using.

DR. CONE: But you have not characterized it. You do not know what it is.

DR. BAUMGARTNER: I understand there are still several hundred uncharacterized things. As far as we know, the main thing is that it does the job, that we analyze it and that it correlates with use.

Just as there seem to be a lot of unknown substances in urine that we haven't characterized, characterization doesn't help.

The important thing is that the substance has a cross reactivity with the antibodies.

DR. CONE: Is it related to THC or to the marijuana plant or do you know what it is related to?

DR. BAUMGARTNER: I like to dream about it and say it is the unknown substance -- okay, the cross reacting substance, I speculated, could be a natural ligand. It is just a fantasy.

What this comes from? It is found in marijuana.

DR. CONE: Or possibly produced as a pyrolysis metabolite, or not metabolite but pyrolysis product.

DR. BAUMGARTNER: I don't know. It is internalized and then it ends up in hair.

DR. CAPLAN: I am going to ask the same question as I asked before. If that is the case, when you do the test in the laboratory and you get a presumptive positive and you confirm it by GC/MS/MS, what are you confirming by MS/MS?

DR. BAUMGARTNER: Carboxy THC.

DR. CAPLAN: So, you are saying that every time this unknown substance appears, carboxy THC is confirmed.

DR. BAUMGARTNER: It is a very strong correlation, very strong, just like with urine.

DR. CAPLAN: Very strong correlation; is it every time or you don't report it?

DR. BAUMGARTNER: No, it is not every time. But we correlated also with cannabinol, cannabidiol and so on, so there are other things. But we don't use those in confirmations.

But just like with urine, not every RIA positive can be confirmed with mass spectrometry.

DR. CAPLAN: I understand that and that is very clear. The question is whether or not you report on the presumptive, unknown entity, where every time you get this unknown entity or positive initial response, then you go to, before you go to report a response to a client, you go to GC/MS/MS and get a confirmation of carboxy THC.

DR. BAUMGARTNER: Absolutely. DR. CAPLAN: That always occurs?

DR. BAUMGARTNER: Absolutely, 101 percent. That is the safety net part. The one percent is the safety net.

DR. PEAT: I am going to continue the discussion. What is the correlation between the RIA reactive and the GS/MS confirmation or the GC/MS/MS confirmation?

DR. BAUMGARTNER: You saw that. It depends on the dose.

DR. PEAT: Not in the controlled studies, but in the random population that you looked at. What is the correlation?

DR. BAUMGARTNER: That is a variable, depending on those factors which we don't quite understand.

DR. PEAT: Fifty percent, seventy-five percent?

DR. BAUMGARTNER: It is on the order of 60 percent.

DR. JONES: One quick question or one quick comment, particularly to you, Dr. Baumgartner. As a board member, I would find it extremely useful if I had some form of a document that describes the procedure.

I thought I walked into here knowing a little bit about hair testing, and you have successfully totally confused me. That is for sure.

Your slides, you had a series of slides that we did not have. If you could supply us with those slides and also with some sort of documentation that describes the hair testing procedure that you are purporting to be utilizing, I would find it very useful.

DR. BAUMGARTNER: I would like to make a point. Obviously 15 minutes is a very short time and a lot of my papers are out there, all of which I would be happy to supply.

DR. JONES: I would like to interject. I think this was titled test reagents and procedures. We heard a lot of documentation about the hair testing protocols there.

I would just like procedures. We have another session, I believe, that talks about interpretation of results.

DR. BAUMGARTNER: Yes, I have a lot to say in the interpretation of results section, too.

DR. HUESTIS: Okay, our time is almost up. I have a couple of comments. I think there has been a lot of interesting information presented.

A few of the things that I am still interested in are the issues that Dr. Jones approached, the issue of cross reactivity profiles between the different reagents.

I think we didn't talk about precision, which maybe we will get to in the quality control, but certainly the precision of the assay will affect the ability of the assay to separate positive from negatives.

Another issue is the stability of analytes of interest in the matrices, in the different matrices and in the different collection devices.

Certainly I think issues about windows of detection and also the amount and type of QC that is required, which we will discuss later on.

Another issue is, someone has lost a wallet. So, if you look around as you get up and if you would return it to the people outside at the registration desk?

I would like to thank all of our presenters very much for their interesting presentations.

(Applause.)

DR. BUSH: I would like to take this opportunity, before we break, to ask Dr. Niedbala and Dr. Baumgartner to share your slides. Give us your slides today.

There were a couple in there, Dr. Niedbala, at the very front end of your talk that we did not have. I would like to make copies of them. Thank you, sir.

Dr. Baumgartner, if we could copy your slides, we will have them back to you tomorrow, so that we can include these in the discussions for the advisory board.

(Brief recess.)

DR. BUSH: This is the last session of the afternoon and we are a few minutes late in beginning. So, while you are quietly resuming your seats, I would like to take the opportunity to begin the last panel session for the day.

This is the confirmatory test procedures panel. Our moderator is Dr. Richard Hilderbrand.

Richard Hilderbrand received a PhD from the University of Illinois, Medical Center Campus, Chicago, Illinois.

He is currently the director for administration at the UCLA Olympic Analytical Laboratory, and is involved with the analytical programs to reduce the abuse of performance enhancing drugs by athletes.

Dr. Hilderbrand has worked in various areas of biochemical and toxicological research. He has edited one scientific volume, contributed chapters and published a number of scientific articles and abstracts. He is an inspector for the National Lab Certification Program.

Dr. Hilderbrand began his career with the U.S. Navy in 1972 at the Naval Medical Research Unit One, University of California at Berkeley.

His experience with drug abuse reduction includes managing a navy drug screening laboratory and being a program manager for the navy drug screening program at the Bureau of Medicine and Surgery in Washington, D.C.

In August 1991, Dr. Hilderbrand became the science and testing officer for the Secretary of Defense. He certified service drug testing laboratories and advised the military services on drug abuse reduction programs.

With that, I would like to thank and introduce Dr. Hilderbrand.

Agenda Item: Confirmatory Test Procedures Panel.

DR. HILDERBRAND: Thank you, Dr. Bush. I would like to take this opportunity to thank the Drug Testing Advisory Board, Dr. Autry and all the members of this august panel. Thank you for this opportunity.

Welcome to all visitors, and I am really glad to see this level of interest in the work place drug testing.

The interesting thing about the work place program in confirmation procedures is that there is really no standard other than an unequivocal result, when that is reported as a positive.

One can talk about optimizing the false positive and false negative results of a test in a screening assay, but in a confirmation there is no real opportunity to have a false positive from a work place specimen, in the regulated side, especially.

A total system has been developed over the years to maintain quality in the drug testing laboratories. This total system has been outlined in a fairly detailed manner this morning, so I won't go over that.

The entire process is designed to ensure that there is no false positive result going out on a positive.

There is more latitude given on negatives. In other words, if a result is reported negative, there is not a serious penalty for that. There can be program impact, but the real penalty comes with reporting a false positive.

The mainstay of this ability to ensure the positive result is gas chromatography mass spectrometry.

This particular procedure is specified at this time by the regulatory directive that covers the drug testing program for regulated industries.

This provides various types of information, not the least of which is the analytical evidence that is used to document this result as a positive.

Also, it provides the assurance that the level in the specimen is above the specified testing level to report a positive.

It provides information that nothing else in the specimen is going to result in a positive report. So, we are looking for specificity as well.

What I would really like to say is that there are a number of parts of the confirmation test. This is not just the gas chromatography mass spectrometry.

There is the preparation of the sample, starting with the hopefully unadulterated clean sample that was collected for this drug testing.

This specimen is subjected to an extraction process. That extraction process is going to result in a generally concentrated and purified drug that can then be prepared for the gas chromatograph by derivitization, so that it is amenable to analysis in a gas chromatograph.

Ultimately, then, it is subjected to the analysis in the mass spectrometry for the unequivocal identification.

The entire set of procedures must be considered when one is talking about a confirmation procedure. It doesn't help to use gas chromatography mass spectrometry if the quality of the specimen is not sufficient or if the preparation is not sufficient.

So, it is imperative that each sample be tested in an appropriate manner.

The current federal regulations require a confirmation method that is independent of the method of screening used and based on a different chemical principle than the initial test.

We have heard a great deal about immunoassays. I am sure that you are now going to hear a great deal about GC/MS and extractions. In addition, this procedure must be capable of quantitating

over a reasonable range. Generally this quantitative range is going to be the pharmacologically relevant value or the arbitrarily set value, however one wants to view the determination of the cut offs.

The analysis must provide a quantitative range around that level, and this must be reproducible.

One must be able to take this same sample to another laboratory and come up with a reasonably close analytical result. This, of course, is monitored by the proficiency testing.

If this entire procedure, as applied in a laboratory, does not provide a plus or minus 20 percent or plus or minus two standard deviations range in the analytical procedure, it is not going to be satisfactory. At this time, GC/MS does provide this.

It has to -- I will add these. These are not out of regulatory writing. But it has to provide a sufficient -- the sample has to provide a sufficient quantity of sample to allow for a confirmation, a retest, for repeats.

Say that there is an analytical problem. So, there are some real sample requirements that are required for the confirmation procedure.

It has to be of a reasonable cost and of a sensitivity to allow a reasonable window of detection.

The National Laboratory Certification Program now enjoys such well developed methods that we can rely on the results that are reported, and we are capable of providing detection and deterrents and it is working very effectively.

There are essentially no false positives and a tolerable false negative rate.

The integrity of the program demands no less. So, as we evaluate any alternative sample, any alternative technology, we must keep in mind that this is a process and not a single test and that there is a lot of quality built into a drug testing laboratory.

With that, I will introduce the first speaker Dr. Timothy Rohrig is currently vice president and director of toxicology of Osborn Laboratories.

Prior to joining Osborn, he was chief toxicologist and laboratory director for the Office of the Chief Medical Examiner, State of Oklahoma.

His previous positions include toxicologist for the Office of the Chief Medical Examiner, State of West Virginia, and Chief Toxicologist for the Kansas Bureau of Investigations Forensics Sciences Laboratory.

He has also held academic appointments at the University of Oklahoma and Rutgers College.

Dr. Rohrig holds a bachelor of science degree in chemistry, a doctorate with an emphasis in pharmacology toxicology, and is certified by the American Board of Forensic Toxicology.

He is active in a number of professional society, including being past president and current secretary of the Southwestern Association of Toxicologists and a fellow of the toxicology section of the American Academy of Forensic Sciences.

He has authored 15 peer reviewed articles and given numerous scientific presentations in the field of toxicology.

Dr. Rohrig will address confirmation methods in saliva testing.

DR. ROHRIG: This afternoon I would like to spend some time sharing with you some of our data on oral fluid or saliva or spit or something that comes out of the mouth that we test for drugs.

Since we don't have a lot of time, what I would like to do is focus in on cocaine testing, and that is something that our laboratory and actually the insurance industry has some experience with.

Saliva testing has evolved from the necessity of the life and health insurance industry to find a simple, cost effective way to assess risk in individuals.

What this fancy statement basically means is that it is a new, cheap way to do HIV testing. Since you could test for HIV they thought, what else could we test for to help us in the risk assessment process in issuing life insurance policies or health policies for individuals.

When we talk about testing of oral fluid or saliva in the insurance industry, basically we are talking about testing for HIV and it is actually HIV-I antibodies, cotinine, a marker for cigarette usage.

The laboratories will also do a validation specimen adequacy test to make sure that we don't have a sample from Rover, or if we have a negative HIV, that it is negative because the person is HIV negative and not that we don't have enough sample.

Then also as a tag along, cocaine testing is performed in the laboratory.

Now, we have heard some talk earlier today about the predominant analyte in cocaine. If you will recall just the previous slide, I had benzoylecgonine in brackets as far as what we test for in an oral fluid sample or a saliva sample in the insurance industry.

This next slide will kind of explain or help explain why we look for benzoylecgonine instead of parent cocaine.

This is some data from Ed Cone's group. On the top we have cocaine itself, tracking the concentration of cocaine with various dosings.

You will see the concentration rise and fall, rise and fall.

The time frame here is very short, on the order of hours.

The question was asked earlier, is there a difference between devices or stimulated or unstimulated. This data here from Dr. Cone shows that at least with IV cocaine, there is a significant difference between saliva, stimulated and unstimulated saliva.

This is what I call the spit in the cup type collection, which is a little bit different from what we use in the insurance industry.

Dr. Peat shared with us the Epitope device, or the Episcreening device, the stick, lollipop, with a pad on it. That is just one of many devices.

Currently we support four or five oral fluid devices that either originate in the United States or are sent in from international clients.

SDS is one manufacturer that Dr. Niedbala mentioned. Right now we will get in three different SDS devices.

The buffer may be slightly different. One device may have an indicator. All this indicator is, is a wick that goes up the device that says we have collected enough spit and the end of it turns blue to say that we have collected an estimated volume of oral fluid for testing.

Back to the data, you will see cocaine rising and falling fairly rapidly. If we look at benzoylecgonine in the dosing of cocaine, it actually rises.

Now there is a significant concentration difference. Here we are like at 300 and here that is like 2,000. So, there is almost an order of magnitude or at least a five-fold difference in concentration, but still fairly high concentration of benzoylecgonine that we are looking for.

As far as the analytical procedures, these are common procedures that we use on urine, that we can use in sweat, really any biological fluid. I use these on brain tissue, for example, in our post mortem work.

You can do a liquid liquid extraction or a solid phase extraction. Since we are looking for benzoylecgonine, we have to derivatize.

You can make a butyl derivative or a propyl derivative.

Some people can make a TMS derivative. There are a variety of derivatives one can make to get benzoylecgonine through the mass spectrometer for your analyses.

Good confirmation, we do use gas chromatography mass spectrometry. The State of Kansas actually mandates that you use that as the confirmation technique.

Recently they kind of revised it and said that you can use some sort of mass spectrometry. So, they would allow MS/MS or LC/MS, but they want a mass spectrometric method for the positive identification of the drug that you are looking for, in this case, cocaine.

Typical cocaine parameters, the people who are doing cocaine analysis probably have something equal or similar to this as far as HP-1 type column or a DB-5 or an RTX-50 or some semi-polar to non-polar type

column.

A typical ramp program, since we are -- this particular set up is for the butylated derivatives. These are the ions that we look for and we monitor the 224 as our quant ion for BE using deuterated internal standard for the analysis.

As far as the identification of the analyte, again, traditional type rules that we have put in place to identify the presence of benzoylecgonine, 2 percent plus or minus retention time. That is compared to the calibrator.

Plus or minus 20 percent of ion ratios, compared to the calibrator. Like quantitation, one can do single point, you can do linear regression. We happen to use single point.

We normally report out either yea or nay. That is, benzoylecgonine or cocaine is present or it is absent.

As far as the calibrator, we use our calibrator at 10 nanograms per device. Given Dr. Cone's comments, I will come back here in just a second and talk about why we report per device.

We have controls, eight nanograms and 12 nanograms on either side of the cut off, if you will, as our internal QC samples.

Now as I stated earlier, we support a variety of devices.

Also, if you go back and remember, one of the very first slides I showed you is what we do in the insurance industry.

Toxicology is kind of the stepchild department in the risk assessment laboratory. So, when an oral fluid sample comes in for testing, it is first tested for HIV or IGG or whatever you use as your adequacy indicator.

Then it is screened for cotinine and then it is screened for cocaine.

If the HIV is reactive, and/or the cotinine is positive, they will either go back and do another ELISA and if it is still reactive, they will go on to western blot.

If the cotinine is positive, some labs will go back and repeat the cotinine assay on another aliquot of that buffer, if it is a collection device that happens to have buffer on it.

Then after all this testing is completed, then toxicology finally gets the sample for the confirmation part of the assay.

Even if we know that this particular device collects 1.5 mls of oral fluid, by the time we get it, we don't have 1.5 mls of oral fluid left to test.

We have opted to report out, this drug is present on a perdevice basis. What we try to do to somewhat normalize it is, if we have an Epitope device come in, our calibrator and controls will be made from Epitope devices.

If we have an SDS third generation device come in, we will make our calibrator and controls from the SDS third generation device. There is another device that comes out of Canada that we get every now and then that is called a dry saliva device that has no buffer. It is basically a wick with a pad on it and it goes into an empty tube.

If we get that sample in, we will make the appropriate calibrators and controls from that material.

Ten nanograms per device, that is our cut off. I have heard a lot of new terms today as far as pharmacologically significant cut offs, scientifically significant.

Well, in the insurance industry a lot of this is what I call market driven cut offs. Insurance boils down to dollars and cents. You have these actuaries or bean counters that will determine that for a certain population there is so much risk of mortality in this population.

They have decided that a certain percentage of that population are cocaine users and those are a bad risk. If you are a cocaine user, the chances of getting life insurance and health insurance, if you are identified as one, is slim and none at all.

They get really upset if they switch from urine, which they have been testing for drugs all along, and those cut offs being kind of set in stone either by regulations by the state or the federal government, but they are kind of comfortable with this 300 screening cut off and 350 nanogram cut off for VE for cocaine confirmation.

So, they say, all right, in the population that we have tested -- this is in hundreds and thousands and millions of samples, a lot of pee goes through these laboratories -- we expect about a .3 positive hit rate.

If you go above or below that tremendously, they get a little anxious because they are worried if they are throwing away good business, not taking policies that they should, or accepting a lot of undue risk.

So, the cut offs are somewhat driven by what cut off can we use that will mimic or match what we are seeing with urine.

That is, the industry wants a cocaine positive hit rate around .3 percent, to be consistent with the urine sample.

So, it is an arbitrary decision, but basically we meet the needs of the industry. That is, we come up with a cut off that is close to what we expect to yield a positive hit rate that is consistent with urine.

Some data. That sounds like a very, very low cut off and it is a low cut off. Can we still have good data on these devices at that low of a cut off.

What I thought I would do is throw a couple of example chromatograms up here, showing you some of the raw data that we actually achieve with this assay in our hands, at least.

This is our below cut off QC. This is an 8 nanogram per device QC sample for benzoylecgonine. Composite total ion chromatogram here, the two ions from benzoylecgonine deuterated internal standard, and then the three ions that we use for quantitation and identification of benzoylecgonine in the sample.

There is really a great signal to noise ratio, the chromatography is nice, fairly short run times. So, we can get through a fair number of samples and have good deductibility of these low levels of BE.

At again the 10 nanogram cut off, we are running consistent.

Actually we run a little bit higher hit rate in the oral fluid versus the urine, but there are some other demographic factors that come into play that might influence that.

So, the bottom line is I tell the industry that yes, our oral fluid hit rate, our saliva hit rate, is consistent with or equal to what we see in urine. That makes the actuaries happy and the insurance is happy because they think they are still making money.

Here is another presentation of data. This is just a different software package, so the presentation looks a little bit different.

This is from a real live applicant sample, showing the internal standard ions here, and again the quantitative and identifying, qualifying ions on the left for BE.

Let's see, this particular one comes out to about 40 nanograms per device.

Linearity assay, the assay if fairly linear. This is data from an Epitope device. We have these same linearity characteristics on the three SDS devices, the dried saliva device.

Dr. Niedbala had mentioned another newer device that is out on the market, but we haven't started receiving those yet. I haven't had a chance to challenge those as far as linearity.

The bottom point on this -- obviously you can't see it, but we can go down to about two nanograms per device and still have good signal to noise ratio. It still falls on the curve.

Now naturally this is a slide, so I picked the best one, but the R-square is one in this particular one. We usually hit close to that, but you didn't expect me to put anything less than that if I had it.

The last piece of data that I will share with you is just a distribution. This is percent of the positives. So, if we take 100 positives, what is the distribution of concentration that we see.

I am using concentration, again, loosely, but how much cocaine is left on the device.

If we talk about the ones that are greater than 50 nanograms per device, that accounts for about 35 percent of the positives.

Between 26 and 50 are about 27 percent, and when we go down to the 10 to 15 nanograms, about 22 percent of all the positives fall in that low range.

We haven't had one, to the best of my knowledge, about 2,500. Most of them will fall below 1,000 nanograms per device for BE.

I think that is about the end of my time, so I will thank you and stop here.

(Applause.)

DR. HILDERBRAND: Now I would like to introduce Dr. Thomas Cairns, as vice president for research and development for PsychoMedics Corporation in Culver City, and adjunct professor of pharmaceutical sciences in the School of Pharmacy, USC. That is the University of Southern California. He serves as an FDA science advisor.

Formerly a senior research scientist at FDA for 22 years, Dr. Cairns served as the director of the National Center for Toxicological Research, a member of the FDA senior science council, a national expert in organic mass spectrometry.

His published research studies have involved the application of mass spectrometry to ultratrace concentrations in the area of food and drug research, in support of a regulatory program to provide consumer protection.

Dr. Cairns received his PhD in analytical chemistry from the University of Glasgow, Scotland. In 1990, the faculty of science of the University of Glasgow in Scotland conferred upon him the degree of doctor of science, for work of distinction in the field of mass spectrometry.

Now we have actually a special treat for Dr. Cairns. He has just been notified that he will be receiving the Hammer Award.

This is the award that was established by Vice President Gore at a special awards dinner at USC next Sunday.

The award recognizes individuals who contribute significantly to the reinvention process in government.

Dr. Cairns was part of a four-member team who reinvented how FDA field labs plan, execute and report research reports. I think that is a tremendous honor for Dr. Cairns, and I think we ought to give him a hand.

(Applause.)

DR. CAIRNS: I understand that award is actually a hammer in a presentation case with a note from the Vice President. So, it is pretty cheap, but it does bear the intention of the program, which was to reinvent and make government a little more efficient.

Of course, I had to leave government to make that transition.

What I am going to talk about today is really a three step process. I don't want to confuse, but I want to educate that in mass spectrometry you are dealing with three steps; identification, confirmation, and quantification.

Now, as a little bit of preamble to this, I would like to take you through the logic behind doing mass spectrometry of hair samples.

By far, the highest level of confirmation is undoubtedly a total scan. Under electron impact, you can in fact match full scan versus full scan.

Traditionally, mass spectroscopers have said those ion ratios should be within five percent, provided the standard and the unknown were recorded on the same day on the same instrument within a reasonable amount of time.

When it came to chemical ionization, the mass spectrometry community -- meaning the Society for Mass Spectrometry -- realized that we were dealing with an ion molecule reaction and that we had pressure fluctuations.

So, they extended the window for ion ratio compliance to 20 percent.

Now we come to the techniques commonly used in the 1980s and the 1990s; that is, single or multiple ion monitoring, where traditionally you choose two or three ions to identify the compound.

That was established while I was at FDA in 1978, and that was done purely from a data base that said okay, here is my molecular weight for my compound. How many compounds have the same molecular weight. It turns out it was a whole slew of compounds.

If I give one fragment ion, it cuts it down to two compounds. If I give it two ions it cuts it down to three and if I give the fourth ion I am down to one unique compound, the compound of interest.

There is no retention time or column, no relative abundance ratio, just purely on the mass of the ions.

Positive chemical ionization is a little more specific for analysis, since it favors the molecular ion. I think the proteinated molecule ion is highly significant of the compound you are dealing with.

It is its molecular weight, after all. So, it gave you increased specificity.

Then along came negative chemical ionization, which was a neat way of saying, okay, if I incorporate fluorine into the molecule through some derivitization procedure, I have greater sensitivity of detection because I have increased the cross sectional area of the molecular for negative chemical ionization or the molecule's ability to sustain a negative charge instead of a positive charge.

The next logical sequence in confirmation technology was that introduction of what we know as product ion chemistry, often referred to as daughter ion chemistry.

It is a simple process of using three mass spectrometers in tandem. The first mass spectrometer is to select the ion required for interest, be it the molecular weight of cocaine or benzoylecgonine, separate it from every other ion in the spectrum and the source, take it through to the second mass spectrometer where you collide with a gas such as argon, and cause a fragmentation and then scan out the molecular weight of that daughter ion or product ion in the third mass spectrometer.

Hence the terminology, TSQ, tandem stage quadripole mass spectrometer. The first one selects, the second one collides, the third one scans and detects.

Now, apply these criteria to hair analysis and you look at it from a number of important points. The first important point is the instrument performance check.

For hair analysis we make sure that we have a limit of detection below our cut off, well below our cut off.

We do limit of quantification and we do that in advance. We also use standards and controls and also blanks.

The standards are used in triplicate within each batch, so that we get a response factor or it behaves as the calibrator.

The criteria that we are going to use, retention time must be within plus or minus two percent of the standards and controls run contemporaneously within a batch of samples.

The ion ratios, plus or minus 25 percent, particularly because we are dealing with hair, we are dealing with a matrix, and we are dealing with a type of mass spectrometry that is chemical ionization oriented and using a type of instrumentation known as an ion trap.

The number of ions, we try to stay within the traditional concepts of three ions, but in certain cases with chemical ionization we have two very strong ions, the specificity being we have one of them equal to the molecular ion or the proteinated molecule ion.

Our quantification is done by a technique as isotope dilution analysis, where we use the retention time to separate the target analyte from its deuterated analogue, usually by three or more scans or at least a little bit of time between the two.

Why is that? Just because a molecule contains deuterium, it does not have the same retention time as the target analyte D-0. You know you have high resolution on your capillary GC column, where you can clearly separate and delineate D0, the target analyte, from its

deuterated analogue, D3.

On batch performance, we are going to do standards and controls. The standards at the cut off are usually in triplicate, the controls one slightly below the cut off, the other above the cut off, and insert the necessary blanks and quality control samples.

The first assay is PCP in hair. Traditionally that is done by electron ionization, known as EIMS. We use an ion trap. We take total scans from mass 88 to 250.

We use a standard curve approach to the analysis of PCP; that is, using a curve that is calculated from two nanograms, five nanograms, 10 nanograms and 20 nanograms to 10 milligrams of hair.

Now, what does the raw data look like? Here, as you can see, we have the three ions belonging to D0 for PCP and the three ions belonging to the deuterated analogue D5. This calculates out at about five nanograms per 10 milligrams of hair. A cut off for this assay is three.

Dealing with amphetamines and hair, we are looking for two target compounds, meth amphetamine itself and amphetamine.

Again, electron ionization is the choice for this assay.

The scan is from mass 88 to 260 mass units. We derivatize to make both compounds volatile for gas chromatography, and we use a mass spectrometer.

Here is what the raw data looks like. Here you can see the meth amphetamine. In this case we are dealing with a deuterated analogue that gives us a 254 versus a 258. So, it is a D4 difference.

The spiking for the deuterated analogue at this level is 10 nanograms and our cut off is five. As I said, we calibrate at five using triplicate analysis, and use that as a calibrator or a response factor.

Linearity studies are conducted from way below the cut off at .5 up through two, five, 10, 50 and 100 nanograms.

As you can see, the spike is at 10, which is slightly above the five nanogram cut off.

Now, how good is this assay on a production level? We have heard a lot of science, but what happens when you do this day in and day out.

What I have got for you here is a typical month period running methamphetamine standards and controls in a batch environment. As you can see, the five is run in triplicate. So, you can see the variance in the five batch by batch as you proceed along the month -- zero to 28 days. Here is the 10 control and here is the two control.

So, we have gone much the same route as urine perhaps testing, but we have gone beyond it. We use the calibrator in triplicate and we are at much, much lower levels.

So, we have gone a little bit beyond, say, the standards required. We also run blanks.

This represents, I think, a good picture of reliability on 14 batches run over a month's period.

Opiates, we are looking for three compounds. What are the advantages, I think, of hair? We are looking not only for codeine and morphine, but you are also looking for the stable metabolite of heroin, 6-acetyl morphine, which is stable in hair over long times relative to the very short half life that is experienced in urine testing.

Here again we derivatize. We are running scans from 275 to 440. Again, the same scenario. We are calibrating three times five nanogram controls.

The significance, I think, is that this compound clearly identifies a heroin user and the poppy issue disappears.

Here is what the spectra look like. You can see the ion abundance ratios are in conformity. Here is D0 versus D3, and the amounts are significant here.

This is codeine at the cut off of roughly five and you can clearly see signal to noise. Here is morphine at roughly 24, and you can see the ratio of 6 acetyl morphine is about 8.4 as derived from heroin. So, you can identify heroin users directly.

How good is this assay? Again, here is the triplicate calibrator at five nanograms. Here is the 10 nanogram upper control and the two nanogram lower control, again representing 14 batches run over a 28-day period. That is a couple of hundred samples.

Moving on to cocaine, here again what is the advantage of hair? You are getting the parent drug, cocaine, the benzoylecgonine metabolite that is found in urine, of course, but we are also getting another metabolite, cocoaethylene.

The advantage of cocoaethylene is the synergism it adds to the cocaine use, indicating that perhaps this individual is at least a frequent user of alcohol.

How does this assay stack up? Here again I produced roughly one at the cut off to show you good signal to noise, again three times five nanogram calibrator.

The cocaine is at five nanograms. The ratio is D3 to D0 giving the amount. The spike is at 10 nanograms.

Here is benzoylecgonine at roughly 1.5 nanograms per 10 ml per 10 milligrams of hair. Here is cocoaethylene at roughly one nanogram.

Mass spectrometry wise, these are very, very significant and outstanding results from the point of view that the on column injection that these represent are in the low femtograms because of the volume of the extract prepared.

When I look at the performance of this assay on a production level over a typical month, here again is the triplicate at the cut off, five nanograms, the high control, the low control. It is pretty impressive for production over a long period of time.

Perhaps the assay that demands the most accolades because it is modern technology at use is, of course, the marijuana assay where we concentrate our attention on carboxy THC.

For obvious reasons, it is not found in smoke. The passive contamination issues disappear. We have a problem going to low levels, and that is solved by tandem mass spectrometry.

The tandem mass spectrometry is done basically on this parent ion after derivitization, incorporating the fluorine so that I can use negative chemical ionization.

Here is the structure of the 670 ion, which is the ion I am going to select to collide with the argon and produce a daughter, or product, ion.

This is the raw data. If you are not familiar with MS/MS, these are the ion profiles for the deuterated internal standard, as well as the D0 for carboxy THC.

These represent individual data points over the envelope.

If you are used to smooth curve, look at this curve. But look at the number of data points acquired for collision.

Each one of these apex points here is basically a collision experiment. So, we get roughly 10 collisions over the elution profile, or the chromatographic portion of elution.

Here are the data points to show you how good the assay is on a production level. I am talking about Psychemedics, doing roughly, basically -- since I have been there two years I have accomplished more than 100,000 of these analyses.

That is spectacular in a production mode. When you consider here is the cut off, there is the two and a half picogram, there is the five and there is the 10, the technique is very suitable and reliable for precision and accuracy over the long term of a month.

You can tell that I do a lot more batches of carboxy THC than I do meth amphetamine. So, it may give you a clue as to drug usage out in the real world.

What were the challenges? The challenges were that for hair analysis, drug and metabolite concentrations were much lower than those found in urine.

Our cut offs, basically we get more positives by hair analysis, say, for cocaine than urine. We are dealing with low nanograms per picograms injected on column. That translates to femtograms.

A digest procedure preserves the drugs, removes the components and our extraction technique enriches and derivatizes for gas chromatography and, as you can see, GC/MS confirmation at the femtogram level.

In closing, I am pleased to say Psychomedics meets the challenge, that it has innovative hair digestion technology with no loss of drug of interest or drugs of interest, efficient extraction to separate the drugs from their matrix, derivatization for GC/MS, and the successful use of ion trap technology to concurrently identify drugs and metabolites using full scan data, and of course tandem mass spectrometry.

You all referred to GC/MS as the gold standard. Might I suggest that you refer to GC/MS/MS as the platinum standard, since it is able to go to extremely low levels for carboxy THC.

I am reminded by Primo Levi that it is not the mere introduction of emerging technologies that advance science. It is the creativity and innovative spirit of the researchers. Thank you.

(Applause.)

DR. HILDERBRAND: It is always a pleasure to introduce the military members of this group, primarily because of the impact that I have known of in the past 15 years.

Dr. Smith is one of those premier members, so it is always a pleasure.

Dr. Smith is currently a U.S. Army colonel and chief deputy medical examiner over the forensic toxicology laboratories in the Office of the Armed Forces Medical Examiner, Washington, D.C.

He is a native of Buckland, Kansas. He received his bachelors degree in chemistry from Kansas State Teachers College, Emporia, and a PhD in bioanalytical chemistry from Purdue University.

He has been awarded diplomate status by the American Board of Forensic Toxicology. Past positions include being director of clinical research at army medical centers and commander of the United States Army Forensic Toxicology Drug Testing Laboratory, which served the European theater in the late 1980s.

Dr. Smith has 120 professional publications and abstracts.

The forensic toxicology laboratories he currently directs include the Department of Defense, Central Post Mortem Human Performance Laboratory, the DOD Drug Testing Quality Control Laboratory, and the DOD Forensic Toxicology Research Branch.

His current research interests are in the development of analytical methods for drug analysis. Dr. Smith will speak on confirmation methods in urine testing. Dr. Smith.

DR. SMITH: Thank you, Rich. My presentation, as you can see, is going to be on confirmation of drugs in urine.

I would like to thank DTAB, first of all, for inviting me.

I see Ed left, but I would like to thank him for turning his back long enough that I stole this slide out of his collection.

Actually, in less than 15 minutes, I am going to outline general testing principles for confirmation of drugs in urine, selection of target molecules, and then go into extraction and GC mass spec requirements.

I am also going to talk a little bit about a few analytical problems and interpretation problems that have arisen in the past decade of urine testing.

In general, confirmation procedure requires a qualitative drug identification. Then once that is done, we have to demonstrate that the drug quantitation is at or above an administrative cut off. I call them an administrative cut off.

Also, I put up here at this particular point, as a reminder, that you also have to be able to interpret the results.

I think a good example of these principles, or general principles of confirmation testing I can select from the military.

In the Department of Defense, it is illegal to use drugs.

It is not just against some regulation. People get prosecuted and they go to jail.

If we consider the example of a soldier who is being accused of using drugs and prosecuted, based solely on the policy of urinalysis result, an expert witness who comes into court to defend the laboratory's data has to first of all prove -- let's say this guy is being accused of using marijuana or hashish -- has to prove that there is 11 nor delta 9 tetrahydrocannabinol 9 carboxylic acid in his urine.

After that particular proof, the expert witness is also challenged to show that the quantitation is equal to 15 nanograms per milliliter or above.

The proof does not stop at this particular point, because the person is not charged with having this particular substance in their urine. They are being charged with using the drug.

So, based on known literature data, the expert witness also has to tell the finders or fact, or convince them, that measurement and identification of this particular metabolite can be tied back to the ingestion of marijuana or hashish.

In the work place programs, both SAMHSA and DOD test for the five drugs shown up here, which include the classes amphetamines, cannabinoids, cocaine, opiates and PCP.

In addition, the Department of Defense also looks for barbiturates, LSD and steroids and selected designer drugs.

Currently the designer drugs that are authorized are MDA, MDMA and MDEA. I should also mention that some other regulated agencies, such as those that are controlled by the NRC, have approval to test for other drugs.

One of the first problems to solve in the confirmation process is to identify target molecules. For the drug selected, the metabolism must be known so that we can pick an effective testing method and also we can interpret the results.

Usually long detection times and analytical feasibility are also desirable.

The choice of a particular target metabolite can be as simple as, in this particular case, for PCP where PCP itself, the parent drugs is one of the principal, long-lasting metabolites.

It can be a little bit more complex, such as in this case with heroin, which is shown in the lower right-hand corner, which metabolizes to 6 acetyl morphine, then to morphine, and then finally to the conjugated metabolites of morphine.

Any one of these particular metabolites could be selected for a drug testing program, but there are many factors that could be considered.

This is one of them. What is shown up here in blue is actually a representation of the conjugated morphine in a particular sample from an individual who was given a known dose of heroin.

The smaller amount of free morphine is also shown in red. Down at the bottom you can see that little bit of 6 acetyl morphine.

It might be a good idea to use 6 acetyl morphine, since it is much more specific, but it is very low in concentration.

This is why one of the general approaches in testing for heroin use is to first screen for the total amount of morphine in the sample, set an administrative cut off.

Then all those samples that are above that particular level are re-investigated with more refined technology to look for 6 acetyl morphine, the specific metabolite.

These are the current metabolites or target molecules and cut offs. I won't go through them. You can read them later in the handout.

I will point out that the cut offs for codeine and morphine for SAMHSA I believe are going to change on the first of July to 2000 nanograms per milliliter. Is that not true, Donna?

DR. BUSH: We have not adopted that policy yet, so there is no way for me to comment.

DR. SMITH: Never mind. I just thought I saw that in the bulletin.

At some point in the future, they are considering raising them to 2000 nanograms per milligram. (Laughter.)

I might also point out for general interest, you can see how much lower the LSD cut off is and why it is such an analytical challenge.

At this particular point, we are going to talk about the analytical process. As this cartoon says, the husband says to his wife, you know, do you want me to guess what is in it? What am I, a toxicologist? That is what we are here for.

The first step in the confirmation procedure is actually extraction. We forget to talk about that because GC mass specs are kind of sexy and more interesting.

But extracting the target analyte out of the urine matrix and getting rid of all that junk is a very important part of the process.

This can be done by either a solid phase extraction or a liquid liquid. Actually, both of these are used and, when optimized, get rid of most of the interfering substances.

Also, GC mass spec analysis usually follows this, once we have got the target analyte into some kind of reconstituted fluid, and electron impact selected ion monitoring is a very popular type of GC mass spec that is used, also ion trap.

There are also tandem mass spectrometry techniques that are used. A case in point might be LSD. At least two of the laboratories do LSD in the Department of Defense system using a TSQ.

I am not going to focus on those because the most common technique is actually electron impact, SIMS. So, when we go into the actual requirements for these confirmation techniques, I will principally focus on just the GC mass spec EI SIM.

As I said, I can start with the solid phase extraction. I won't go through this. This is a teaching slide. I am just using it here to show that this is a first step.

You end up getting a chromatogram. I thought at this point I should probably show a chromatogram, since that is what we are talking about.

As you can see, there are some impurities in the solid phase extraction, this particular one for PCP. But they are well separated from the analyte of interest.

I am going to go through just the GC mass spec criteria.

You have seen many of these before from the other presentations, but a lot of these were developed for urine testing.

Certified capillary columns essentially means that once you get a capillary column, which is the most common type of column used in urine drug testing, you have to condition the column and then you have to put some kind of check samples through it, to show that you get good analytical separation.

The GC mass spec procedure itself has to be tested with known samples and controls, to demonstrate that you can get expected values. Some of the actual requirements for that you will see later.

You must determine an LOD, and LOQ, and to test the linearity not only for your instrument, even though this is an instrumental check, but for the procedure itself.

We often forget that the operators have to be certified and in most laboratories that is done. Records are kept of their certification and continuous education is provided for most of those people also.

A very important component of the laboratories which has developed over the past decade is the reviewers certifying scientists saying, this result is good to go.

At this point I should probably point out that most of these techniques are good laboratory practice. Many of them are actually requirements within the SAMHSA system and within DOD. Some that aren't I will try to point out as I go along, if I happen to remember that.

A daily autotune is required in most of the laboratories and also that you have frequent septum and insert changes. In the Department of Defense it is required to be daily.

If we go on to the analytical quality control requirements, we have to use an extracted standard in the testing process, which means it cannot be a non-extracted standard in some kind of a solvent.

In general, deuterated internal standards are used. There are some exceptions. One I will mention again is LSD. We have not found an acceptable deuterated standard for LSD. We still use a congener, lampa, in that particular assay.

The standards and controls themselves, once they are prepared or purchased must be certified to have the right target concentration and to meet other parameters.

There are open and blind quality controls in the confirmation process. This is one difference I will point out.

I don't think SAMHSA or the NLCP program currently requires blind controls in the confirmation process, even though some laboratories do carry their blinds through to confirmation.

The Department of Defense right now requires both a positive and a negative blind control.

Some of the controls that are nice to have in the process are a drug free control, a cut off control, something that is 40 percent of the cut off, and at 125 percent of the cut off.

Controls like this are similarly required by both the NLCP and the Department of Defense.

Also, the basic chromatography, that thing we learned a long time ago, has some requirements. There are certain peak height and area requirements.

I haven't put any specific things up here because it depends on the instrument, the laboratory and the procedure.

But nearly every laboratory has some minimum requirement, which includes minimum abundances signal and noise criteria, symmetrical peak shape.

Now the last two things listed up here are specific objective requirements. In most cases the retention time has to be plus or minus two percent of the extracted standard.

In the Department of Defense we have a resolution requirement that is listed here at the bottom, which essentially says that the resolution of the peak of interest, whether it is the principal ion or any of the associated ions, has to be separated from other things with a resolution greater than one. If there happens to be a shoulder on the peak, the shoulder has to be less than 10 percent of the peak height.

GC/MS criteria are listed up here: three drug ions, two internal standard ions, ion ratios must be plus or minus 20 percent of the standard.

This is what an ion chromatogram looks like, three typical ions for PCP.

These are some of the analytical problems that I mentioned before. Some of those people who have been in the program a long time will know that some common interferences with amphetamines are sympathomimetic amines. This was a problem in the system.

SAMHSA addressed this problem principally by making the laboratories prove that they could separate things like ephedrine, pseudo ephedrine, et cetera, from the target compounds, methamphetamine and amphetamine.

There was a heroin and poppy seed problem that arose. This was solved by medical review officer requirements.

The last two up here are specific for DOD. Essentially the one for cocaine was that benzoylecgonine was said to not be a good enough metabolite. It was later shown that it was acceptable in most courts of law.

This is a specific problem, just a total ion chromatogram.

I should point out that the PPA there should be ephedrine.

You can see some extra peaks in this methamphetamine ion chromatogram tracing, but well separated from the methamphetamine which is on the left.

Not so for the particular ion selected for amphetamine.

Even in the total ion chromatogram where there didn't appear to be interferences, there was some.

This is s typical problem for DOD. This particular screened positive. Sixty percent of the samples then confirmed negative.

One of the reasons is a contaminant, isol LSD. That is something that needs to be addressed in the Department of Defense.

I put this anecdote up here because it essentially shows that we have to know something about the science of the molecules in order to establish good policies.

This is my summary. We need, for GC mass spec confirmation, an optimized extraction, defined chromatography criteria, acceptable MS criteria, adequate quality control, and we need to interpret. We need to have criteria to do that.

Once we have done all these things, we are ready to fly.

Thank you.

(Applause.)

DR. HILDERBRAND: Now an opportunity to re-introduce Dr.

Armbruster. I won't go through the entire bio that he has provided. What we will hope is that he in the last hour is still the designated RP at PharmChem, and leads their forensic drug testing for the National Laboratory Certification, certified laboratory.

He is the laboratory scientific director and I welcome Dr. Armbruster.

DR. ARMBRUSTER: Thanks. I was kind of hoping to waive all that fluff in the bio and take that time and add it to the presentation.

I feel kind of pressure to go into verbal after burn to tell you everything that I want to tell you.

DR. HILDERBRAND: One minute.

DR. ARMBRUSTER: It can't hurt to ask. I feel a little bit guilty. One thing I didn't talk to you about in the first presentation was our elution procedure.

Yes, we get this sweat patch and we don't have liquid here.

We have to elute the drug from that patch. What we do is take about two mls of a methanol acetate buffer and we have a half an hour elution procedure. We wind up with maybe 1.5 mls of actual eluate and that is what we actually test.

Going back to Dr. Cone's point, when you determine a cut off in terms of concentration, how should you standardize or establish that. Well, we use it in terms of nanograms per ml, because we do

have a certain quantity of liquid that we deal with.

That is also a problem, because if you have a multiple positive, you can run out of sample before you can confirm everything that you would like to confirm.

I should mention that we have stuck with the sacred NIDA five. It seems like a practical thing to do. We can look at other drugs, and with the other sources of specimens you could look at other drugs as well, but for practical purposes, we stick with the NIDA, now SAMHSA, five.

I have explained about the cut offs, where we got them from.

What I again wanted to emphasize is that data that I showed you before, that clinical data, all those specimens were analyzed by GC/MS as well as the immunoassays.

So, we do have known, quantitative -- for our system -- nanogram per ml values for all of that data.

Finally, of course, as you recall from the discussion of the immunoassays, we know that we are going to get false positives. That is just something that goes along with choosing the optimal cut off.

The false positives are still the bete noir, the hobgoblin of forensic toxicology. Our ace in the hole is that we do have confirmatory procedures and we can use them to knock out those false positives.

This is our amphetamine GC/MS assay for the sweat patch.

Two analytes here are amphetamine and methamphetamine, same as in urine. We use internal standards, deuterated internal standards, just as with the urinalysis for GC/MS.

I am expecting a collective gasp because this is the notorious CB derivative. You remember this bad boy from a few years ago, the false positive methamphetamine in urine samples. That is proven and it is a concern, and I will get to how we address that concern in just a second here.

We use a 58/90 59/70, a standard GC/MS system here, electron ionization, DB1 column, again three ions for both drugs, two drugs for each one of the deuterated internal standards, the standard approach for confirmatory testing.

The cut off for GC/MS is the same as for our immunoassay, 10 nanograms per ml, and again, you saw that data, the clinical data, the ROC curve.

We can go down to two nanograms per ml for an LOD LOQ,

decent linearity for what we are dealing with here. Of course, recognize, in comparison to urine, it is a 500 nanogram per ml cut off.

So, we are quite a bit below that, and keep that in mind when I show you the chromatograms.

The control scheme and calibrator scheme follows what we use at PharmChem for urine testing. So, we have a zero, we have a low control, we have basically a cut off control.

We throw in a 20 control, twice cut off, just for the heck of it in this assay. Then we have what we call a carry over control.

We will follow this carry over control with a negative. We look to make sure that negative is below LOD. Otherwise, we may have a carry over problem.

Okay, so we are using this CB derivative. We know that ephedrine and pseudoephedrine and compounds like that in large concentrations in urine can cause a false positive, because in the injection port under the right conditions -- and there are a lot of discussions about what the right conditions are, phases of the moon, et cetera -- you can create methamphetamine, it seems.

We tested ephedrine and pseudoephedrine at 5000 nanograms per ml, and yes, I know you are going to say but a milligram per ml is what we use in urine for interference, but our cut off is a lot lower, too. We feel that this is not unreasonable in terms of challenging the system with the amphetamine-like drugs that could cause a problem.

Another thing that seems to be common when you have the false methamphetamine positives is a high injection port temperature.

We run at a relatively cool 200 degrees centigrade. We don't think that we are creating a problem for ourselves.

To date we have had no amphetamines -- that is either amphetamine or methamphetamine -- to test greater than 1000. So, we are dealing with fairly small concentrations of the amphetamine compounds here.

This is a typical chromatogram for amphetamine at 10 nanograms per ml, at our cut off. I think you would agree this looks pretty decent, what you would expect for a urine specimen, but with a cut off of 500 nanograms per ml. So, we are kind of proud of that.

Here we see the typical split methamphetamine peak, and here is the methamphetamine, again at our cut off of 10 nanograms per ml. I think this looks acceptable and much like what you would expect with urine testing.

Here we have the cocaine sweat patch assay. Here is where we split company somewhat with the urine testing.

We look for cocaine. We look for BZE, benzoylecgonine, and EME or ecgonine methyl ester, all three of them.

Of course, under the SAMHSA program this is verboten and this is verboten. You can only look at benzoylecgonine. This is the blessed drug, the metabolite that you can actually acquire data for, as SAMHSA likes to say.

We look for all three because we can find all three. Again, cocaine is really the predominant form of the drug that we find in sweat, however.

Internal standards for all three, typical trideuterate internal standards. We use a derivative for the benzoylecgonine and the ecgonine methyl ester. The cocaine comes through underivatized.

We do use solid phase extraction. If I don't mention solid phase extraction it is because we use liquid liquid.

Here we have a 58/90 59/71. The 59/71 is not unusual in laboratories doing confirmation for urinalysis testing, too.

A typical EI in the DB1 column, we look at the typical EI SIM approach, three different ions for each one of the analytes and two ions for each one of the internal standards.

Our cut off remains, as for the immunoassay, at 10 nanograms per ml. We have established LOD LOQ for each one of the analytes.

I should mentioned these are not statistical. We kind of favor an empirical approach, which I personally think is good, rather than just multiplying a number to come up with a statistical kind of LOD LOQ.

Decent linearity, again when you consider our cut off is 10 nanograms per ml. Again, we use a zero, a low cut off, a cut off and a carry over control.

Here we have cocaine, which is the longest retention time. Again, I think it looks halfway decent, acceptable, followed by BZE, benzoylecgonine, the shorter retention time, right here. Again, this looks like decent chromatography, acceptable results. Finally, the ecgonine methyl ester, the shortest retention time. I don't think many people would have a problem accepting that kind of data.

On to opiates. Opiates gets even more interesting. Now we have four analytes. We can detect all four of these. We now can detect heroin, the parent compound, better than even 6-MAM. We have the mother of the drug right there, plus morphine and codeine.

So, following the standard protocol, however, for urine testing, we have a separate internal standard for each one, trideuterated.

We do a derivative. We use a 58/90, 59/72 and the 59/72s are not all that unusual either these days either in urine testing laboratories.

EI SIM, three analytes again for each one of these four.

Two ions for each one of the internal standards.

The cut off for the GC/MS assay is at 10 nanograms per ml as for the immunoassay. Yes, LOD LOQs for all four analytes. Linearity 5 to 500, again decent when you consider our cut off. Controls, the same standard pattern that you have seen for the other analytes.

Here we have heroin out here, the longest retention time.

This maybe doesn't look quite as pretty as what we have seen before, but it is usable. It is workable data.

Going on down, we have our 6-MAM. Again, maybe not the prettiest, but certainly acceptable. Cut off for morphine, next in line here, I think that is all right as well.

Finally, codeine, again, at the cut off of 10 nanograms per ml, not textbook perfect here, but again, it is acceptable data, when you consider again that a urine cut off for GC/MS is up at 300 nanograms per ml.

I have to apologize. I don't think you will find this slide in your handout. I don't know if that is a faux pas on the part of us at PharmChem or if something was lost in translation further down the line, but in my copy at least I don't have this, but it is not that big of a deal.

This is PCP. The analyte is PCP, as with urine. Of course, we use this typical internal standard. No derivitization. We are back to a 58/90 59/71 system, EI SIM, three ions, two ions for the internal standard.

We use a 10 nanogram per ml calibrator cut off. The cut off, however, is at 7.5. So, maybe we fudge a little bit there. We don't feel like making up a 7.5 cut off calibrator.

This is, of course, the same cut off that we use for the immunoassay, LOD LOQ down at one, linearity one to five, same standard control.

This is our cut off at 10 nanograms per ml, looking pretty good here for PCP. Of course, this isn't that dramatic a change, when you consider it is a 25 nanogram per ml specimen for urine specimen.

Now just for chuckles and grins, I thought I would take you down to the one nanogram per ml LOD here. Not bad, it is getting a little bit fuzzy there, and not super. We are pressing the envelope, but we can still get usable data down at our LOD LOQ for PCP.

I have saved the best for last, which is THC. Here is where we do kind of break company with the traditional approach in urine testing.

The analyte is THC, not the carboxylic acid metabolite, but the parent drug. We do use the parent drug trideuterated for internal standards.

We derivatize it. We have solid phase extraction. We use 58/90 but now a 59/89 MS engine, not typical for urine labs.

We use negative chemical ionization. Why do we use NCI? Well, the big advantage of NCI is that it allows you to increase your sensitivity.

Estimates are 10 to 100 times increased in sensitivity. I don't really want to say sensitivity. I want to say lower your limit of detection. I hate using those synonymously.

Anyway, we have to go down pretty far here. So, we couldn't get by with the typical system. We had to go to the engine. We had to go to NCI in order to get the limit of detection that we needed.

Our cut off now is not five nanograms per ml as we had with the immunoassay, but .5 nanograms per ml or 500 picograms per ml.

Here again is a little differences. All right, it is a big difference maybe. We use only a single ion for both the analyte and for the internal standard.

Then again, remember, chemical ionization is a kinder, gentler way of getting ions as opposed to EI. So, we don't typically see as many ions either.

So, we are running with just a single ion for both analyte and internal standard. Our LOD LOQ us down there at .02 nanograms per ml at 20 picograms per ml, pretty low.

Our upper linearity is 15 nanograms per ml, which is the cut off for the urine testing system. Our control is zero down there, well below the cut off, right at the cut off, and about 10 times, twice the cut off there, and a carry over control as well.

What does it look like? Like this. At cut off, .5 nanograms per ml or 500 picograms per ml, that is what we can see with the engines and the negative chemical ionization.

So, it is not too terribly bad, and yes, it is only a single peak for the analyte, only a single peak for the internal standards.

Again, for chuckles and grins, I thought I would show you what it looks like when we go down to that 20 picogram per ml limit of detection, limit of quantitation.

Internal standard is doing just fine. This is not the prettiest chromatogram that you would like to see, but you certainly get separation from baseline with the peak there.

So, again, we are pressing the performance envelope of the assay. I have two minutes left and I think that is my last slide, so I actually beat the limit. Thanks very much.

(Applause.)

DR. HILDERBRAND: There was a comment made earlier that the result of not having a long and illustrious career was having a short bio.

There can be an alternative explanation and that is that a person is a little more modest in what they present to be introduced with. I am certain that is the case here with Dr. Sal Salamone.

I have a short bio but I don't think that this represents everything that he has accomplished.

He has a bachelors degree in chemistry and biology from Villanova, MS and PhD in chemistry from Rutgers, post-doctoral fellow in bioorganic chemistry from Oxford University.

He has over 90 scientific publications and abstracts and is an adjunct professor at Fordham University, New Jersey University of Dentistry and Medicine and the New Jersey Institute of Technology.

He has 14 years in research and development at Roche Diagnostic Systems and is currently responsible for the United States R&D and manufacturing support operations for Roche Diagnostic Systems.

This responsibility includes centralized and decentralized testing for both therapeutic drugs and drugs of abuse.

Dr. Salamone will speak on the confirmation and on-site testing.

DR. SALAMONE: I don't have a great deal to say about confirmation testing, since the testing is urine based testing, and the antibodies that we are using are the same antibodies, or they have similar characteristics to antibodies that you use in the instrument based test.

In addition, the cut off values, if you look at all these on site tests, the cut off values are identical in most cases.

So, what we are recommending is that you take presumptive positives and then you follow the proper chain of custody as you would in HHS testing, send it to an HHS lab and perform GC/MS confirmation according to HHS procedures.

On my first slide I just -- it is as simple as that. The thing that becomes more critical with these on-site tests is the proficiency testing.

There are a couple of critical characteristics of on-site testing that are different than laboratory based testing.

One is, in laboratory based testing, you use a small sample size and you dilute it in the reagents. Perhaps only five or 10 percent of the total reaction mixture is urine.

Well, with most on-site tests, you are using the urine neat.

What you are doing, you have lyophilized reagents or reagents in some sample pad.

What you are doing is you are putting in the urine and the urine is the reaction medium in which the reaction takes place.

So, these types of tests are more prone to matrices. So, when one is setting up a proficiency testing program for urine based onsite tests, you really have to evaluate these tests in terms of how they react with different urine matrices. That is a very important aspect.

Another important aspect is the cut off of these assays.

When you are performing an instrument based test, what you have is every day you can run a calibrator and base the assays on that calibrator.

Well, these on-site tests are fixed cut offs. They don't vary or you cannot put in a new calibrator and recalibrate the test.

What Dr. Caplan was saying earlier, the test valid and test complete doesn't really indicate the performance of the test, but it indicates that the reagents are working properly.

I have been at this game long enough, and I have evaluated enough on-site tests to tell you that FDA approval or FDA certification doesn't mean that the test is going to perform according to package insert criteria.

Any proficiency program that you put together really has to have samples that challenge the cut off, not necessarily challenge the cut off plus or minus 25 percent of the cut off, but challenge it in some way that you can feel good that the cut off that the package insert states is indeed the cut off. That is about all that I had to say.

(Applause.)

DR. HILDERBRAND: This is good. We actually had a fair amount of time for questions anyway. The next session starts tomorrow morning at 8:15, but this gives a few additional minutes.

If there are questions from the Drug Testing Advisory Board, I am sure the panel members would be pleased to take those.

DR. KWONG: I have two quick questions. The first one is for Dr. Armbruster. In your amphetamine assay, the 5000 nanograms per ml of ephedrine and pseudoephedrine, is that what you would expect in sweat for someone who abuses amphetamine or has taken a high dose of ephedrine.

DR. ARMBRUSTER: That is why I qualified, at the very bottom of the slide, we haven't seen any amphetamines above 1000 nanograms per ml.

That is a good question. We are still in the infancy of sweat testing. I can't honestly say that we have the data to show that if a person was taking a certain dosage of ephedrine, pseudoephedrine or some other amphetamine compound, over the counter type preparation, what we would really expect to see. That is, at this time, I guess our best guesstimate.

The second question I have is for Dr. Rohrig. I am still nagged by this concept of nanogram per device. The question I have is, just like Dr. Armbruster saying that the sample having to be eluted from the patch, in saliva you have to collect it in a pad as a liquid or it is in a pad and you have to elute it.

Don't you have an aliquoting staff where you can quantitatively aliquot saliva so that your results can be in a nanograms per ml?

DR. ROHRIG: Yes. Most of the devices that are in use today have buffer in the pad. In the normal process for the initial screening, you will centrifuge the whole device and theoretically force the liquid out of the pad, and you can aliquot that buffer.

The problem is that you are having several assumptions, that the pads collect an equivalent volume of oral fluid.

That is one of the problems, for instance, with the Epitope OraSure device. The requirement is to keep it in the mouth for two minutes.

Two minutes, to some people collecting it, is an inordinate long period of time and other times two minutes is very quickly.

I could say, well, there are so many nanograms per ml of buffer. To me that is just as useful information as saying there are so many nanograms per whole device.

The purpose that we are utilizing, at least in our industry, is to identify a cocaine user. We are not making any predictions such as impairments or that sort of TDM type of work.

You could call it per ml but then you would have the same discussion as you would with the sweat patch, where you are eluting it off.

I just felt that saying nanograms per total device, test kit, more accurately reflects what we found analytically, instead of just making an arbitrary determination of so many nanograms per ml of buffer.

DR. KWONG: If you were to use saliva in work place drug testing, would you feel comfortable using that concept, nanograms per device?

DR. ROHRIG: It depends on how it is collected. If you are using it in work place testing and you have a protocol as far as a stimulated sample and you spit in a cup, if you will, then yes, you could get by with that.

Again, coming from our industry, there is other testing going on. We are not doing drug testing in a vacuum. We have to adjust what we are doing to fit the needs of the industry, and accurately reflect what we are doing so that the data is not misused.

DR. JONES: One quick question, especially to the hair and sweat and saliva device testing. Are there compounds, or are there analytes that we should be aware of, that you are aware of, that should be included in proficiency testing programs as challenges to analytical procedures that are unique to either of the three of you?

It follows up from Tai's question a little bit, but in the urine testing program we see the sympathomymedica means put into the PT program.

I know we may be getting into this a little bit tomorrow also, but is there anything that you have identified in your experiences that we should be aware of?

DR. ARMBRUSTER: when our assays were validated, we did interference studies using the typical interferants that one would test for in urine.

Obviously, in terms of proficiency testing, for the amphetamines and for PCP it would be the same as for urine testing. For cocaine, we really would have to be challenged probably

by using cocaine, and for THC actually THC and not the metabolite carboxylic acid.

In opiates, heroin and 6-MAM, which we don't even look for heroin in urine, but that is the major drug we look for in the sweat patch.

DR. CAIRNS: I think in the hair analysis all the analytes I showed you, the three opiates, the cocaine trio, carboxy THC, but more important we are talking extremely low levels.

Remember, carboxy THC, in hair the cut off is half a picogram. The question I am often asked is how do we equate the amounts in hair and cut off levels relative to the other matrices.

You know, it is very difficult to compare a ml of urine to 10 milligrams of hair, considering that the roots that the drug gets into hair are entirely different from those for urine.

When you consider it and say parts per trillion and parts per billion, that half picogram cut off for carboxy THC is certainly three or four orders of magnitude below what urine testing is involved in.

I think when you are talking about proficiency testing, the challenge is not in the analyte itself. The challenge is in being able to do that concentration level per 10 milligrams of hair.

DR. ROHRIG: as far as saliva, you have all the challenges that you have of the other matrices that we need to be aware of.

Our experience has been mainly cocaine, but if you are looking at amphetamines, if it has been established -- I don't have the answer to this but if there are high levels of pseudoephedrine or something that appears in the oral fluid, then obviously your techniques should be able to, if you are just gearing in on methamphetamine or amphetamine, differentiate and remove that interference.

You have all the issues if you use a CB derivative and things like that, as far as the artificial formation. So, you have the additional challenges.

I think that is one thing that separates oral fluid and some of the newer matrices, is information on drug disposition.

Every time I have a question on that, I try to find one of Ed's subjects and see if they got that sample from them, and look at some of his data to see what I am dealing with.

DR. CONE: I have a question for Dr. Cairns. I am always worrying about whether people are going to get accused falsely or not.

I believe it was in your presentation that you said that you are testing for the carboxy acid metabolite of marijuana.

Did I hear you say also that that does not occur in marijuana smoke? If you did say that, have you actually produced data that would support that contention?

DR. CAIRNS: I am being logical here, Dr. Cone, in that carboxy THC is a two step metabolic process in the body. How in god's name would I create that in smoke. As a scientist I don't see any rational or logical explanation how I would produce it in smoke. You agree with me on that one, I hope.

DR. CONE: I don't agree with you. I don't know what occurs in a pyrolysis product at 1,200 degrees.

DR. CAIRNS: I hardly think a carboxylic acid, such a complex structure, would survive pyrolysis. It would dehydrate right away.

DR. CONE: It is an extremely simple thing to test, though, is what I am asking. Have you tested it.

DR. CAIRNS: It is a good point. If I were to try and convince a skeptic like yourself yes, I did an experiment, but you have to agree with me, it is highly improbable.

DR. CONE: I agree with you.

I have more of a lighter question for the panel. This is kind of a dumb thing to say, but just for want of discussion, I would like to -- this is something we should probably bring up in the next conference. I assume we will have another one in nine years. But I want to ask it in this conference.

Let me make a statement first. What I think this country needs is a good 25 cent on-site test for saliva, sweat and hair and identification of drugs of course.

Do you guys see us going to on-site testing of alternate matrices, and what do you think about the idea?

DR. CAIRNS: You give me two million samples and I will certainly see the price go down.

DR. ARMBRUSTER: I think it would be a challenge for sweat considering the low concentrations of the drug. If somebody was willing to pour money into the research I imagine you could come up with such a test.

With urine I think it would be a lot easier to do on-site testing than with sweat. What I kind of think may happen is that we will take the approach I have always favored in the clinical laboratory, multivariant analysis.

You take several analytes and you get a bigger answer to what is going on when you look at it all put together instead of just a single analyte.

Here we have had several specimens giving us different bits and pieces of the puzzle. If you put it all together in a holistic approach and we get more information about what is going on.

DR. SALAMONE: The antibodies that you would need would be somewhat different, of course, for saliva. Any time we have looked into what the market potential of that was, it just -- it wasn't that great, to put a very big R&D effort into it. There is not a great deal of work going on.

DR. CAIRNS: Let me ask the question. How would you feel about conformation that did not conform to rigorous forensic standards as we know them today?

Is there a bartering to reduce those? We have pretty high forensic standards right now, number of ions, ratios, et cetera. Are you indicating you would relax those a little bit to reduce costs in onsite testing?

DR. CONE: I will respond to that. I really didn't come up to respond to questions but no, I am not advocating lowering the standard.

I am advocating that, as technology involves, we will have sensitivity in those different areas that will allow you to detect drugs at lower and lower levels.

I never thought we would be testing at femtogram levels before. So, I can see where we may be heading in the future toward much more sensitive assays than we have now, that could be implemented in onsite testing scenarios.

DR. SMITH: Ed, I just wanted to say that I can't answer

your question, but I thought that I might give an anecdote.

When the on-site testing methods and procedures were not as good as they are now, we used them actively in Europe.

One of the jobs of the drug testing laboratory was kind of as a QC organization. We would try to go to different sites and see which people had positive field screen tests. When they turned up negative in the lab to a great proportion, somebody would go investigate.

I don't really have any hard statistics, but my sense was, from all the problems we had, almost all of them were operator related. When they went to the site and they trained the operator,

then most of the problems with field testing went away.So, we can have all the good, sensitive and well thought out procedures that we want, but to get back to what Yale said, if the people are not properly trained and that is your principal problem, then that is the problem that we really need to address.

DR. CONE: You may have Bubba looking at you, too.DR. HILDERBRAND: For the record, I would like to identify

the speakers. I don't know if you need that. Okay, any other questions?

DR. WILKINS: I just have one quick question for Dr.

Armbruster and Dr. Cairns, and it is a related question. Are calibrators and controls in the confirmation assays prepared in the appropriate matrices in both of those?

For Dr. Armbruster, for example, are controls prepared by spiking them to the patch and taking them through the elution process? For Dr. Cairns, obviously the same type of question. Are

controls and calibrators made in a hair matrix by fortifying with known concentrations of drugs.

Then take that one step further, each of you, if you would.

What are the CVs for replicate samples in real samples, not fortified samples.

My experience has been they are usually higher than in fortified hair samples. I am just curious, in routine specimen analysis, what kind of CVs you typically get.

DR. ARMBRUSTER: The answer for sweat testing is yes, we find it necessary to duplicate reality to the spike patches. Sometimes we see matrices as well, which we have to duplicate by spiking actual pads.

Is it possible to defer to somebody else like Bob Vogelson, who would have a better idea of what we typically run for CVs.

DR. CAIRNS: In hair analysis, you are quite right. We identify some negative hair and prepare what we call a pool negative hair digest. That is the matrix that we spike, say, for carboxylic THC.

We also include in the batch, apart from that part up front that shows linearity as well as capability, we also have what are known as quality control samples, which are hair from users, which we characterized -- it is like an internal proficiency test in its own way.

We have characterized it by multiple assays, taken the mean, and we put that into the batch on a regular basis.

The CV for those samples on a regular basis is well within 10 or 12 percent.

DR. WILKINS: Those proficiency tests were those that were alluded to -- I forget which of the speakers mentioned it. I think it may have been Dr. Baumgartner earlier, who mentioned that some of the previously confirmed positive hair specimens were used as quality control. Is that the same material or is that something different?

DR. CAIRNS: This is bulk hair that we acquire, and homogenize it by cutting it to very small ends, mixing so that we have a fairly homogenous sample.

Then that is digested as a normal 10 milligram sample and it is like replicate analysis. We have got hundreds of replicates from that batch of hair from a user.

The CV on that is running between 10 and 12.

DR. WILKINS: So, that is different from the blind QC.

DR. CAIRNS: That is different from the blind.

DR. HUESTIS: I have a couple of questions. First, for on-site tests, would you recommend a percentage of the negative specimens that result from on-site testing that should be sent to confirmation to test the reliability of the on-site testing process.

DR. SALAMONE: That would have to be determined, but I think it would be a valid way of going, yes, to send a certain percentage of the negative tests.

DR. HUESTIS: Do you have any recommendation for what that percentage might be at this point?

DR. SALAMONE: No, I don't have a feel for that.

DR. HUESTIS: Also, does the fact that you are using a much higher concentration of the urine specimen itself in the on-site device make this technology more susceptible to adulteration.

DR. SALAMONE: With adulteration, yes, it does. With adulteration what generally happens is that you don't get the test valid sign showing up, or test complete.

It completely changes the reaction so that you don't -- the reaction just doesn't complete.

DR. HUESTIS: I have questions for each of the alternative matrix speakers. If you would address the approximate percentage of positive initial tests that are confirmed and how this percentage has driven your selection of confirmation cut offs.

DR. ROHRIG: As I stated earlier, one of the driving forces for confirmation cut off is to match our urine. That is really one of the major factors as far as hitting it.

Then you can play statistics as far as screen positives versus confirmed positives. One of the problems we run into with our population -- I was talking to Dr. Peat and he experiences similar things, which doesn't surprise me -- many of our screen positives are also HIV positive.

That means that a lot of the specimen is already consumed even before it goes to cocaine confirmation. So, we can run anywhere from 10 to 20 percent of these are Q & S. That is, there is just nothing left there to be tested.

I can say that anywhere from like 50 to 80 percent will confirm positive. That is a wide range, but that is highly variable upon HIV status, smoking status.

In our laboratory if they screen positive for cotinine, another analyst will go back and take another sample and re-test for cotinine, again consuming more samples.

I will have to give you just a wide range like 50 up to 80 percent will confirm positive, but that is mainly driven by all the other challenges as far as small specimen size that we have in our particularly unique industry.

DR. HUESTIS: A comment on that. I understand your reasoning for the insurance industry and why they wanted the testing to be similar to urine.

It seems to me that we shouldn't be trying to match cut offs or confirmation rates in these different analytical alternative matrices, because they have different windows of detection and certainly should be looked at in their own right.

DR. ROHRIG: I say it is market driven and we are talking about windows of detection. Again, the actuaries write these complicated mathematical programs determining risk for an individual that is a cocaine user.

The reason we have such lower confirmations is to match urine, but in a very loose general term we are saying that the window of detection, the surveillance window, is similar to that which we see with urine.

That is what they are basing it on. Say we have a casual user and we know we are going to pick him up for X number of days. We want to kind of match using whatever matrices.

The insurance agency could care less if it was saliva, urine, brain tissue, whatever. They want about the same hit rate because that is what all their models are written on.

DR. CAIRNS: First of all, I think the RIA screening assay is what passes samples through to GC/MS for final confirmation, and that is the system that reports a positive.

The RIA assays are a little more sensitive, perhaps, than GC/MS. Our confirmation rate from the RIA assay -- one of the great things is that you can compare the RIA volume with the GC/MS volume, and there should be some correlation between the two, so we can see an absolute correlation between screening and confirmation.

For cocaine it is in the high 90 percent. For the amphetamines it is perhaps 70 percent, for PCP it is 100 percent and for carboxy THC there is a variable there going all the way up to 70 and 80 percentage positive rate, from the RIA screen.

DR. ARMBRUSTER: I don't want to attempt to quote numbers for the positive rates confirmed, but we have much more confidence in the GC/MS confirmatory assays than we do the immunoassays. There is more slop in them, if you will, and they are just not ideal.

Particularly with THC and cocaine, we tend to have screen false positives and put them through GC/MS and they will not confirm. Part of the problem, I think I alluded to earlier, is sweat

testing is one of the new kids on the block and not every manufacturer of immunoassay kits is willing to sink a lot of time, effort and money into developing an optimal screening assay for sweat testing.

I still think we have work to do in increasing the analytical sensitivity of the screen assay. So, we don't crow about fantastic screen positive rates when we get to confirm.

DR. HUESTIS: I think this just emphasizes the importance of controlled clinical studies, again in determining what those confirmation cut off rates should be. Thank you.

DR. HILDERBRAND: I would like to add one comment to that, Marilyn. There were two locations in the navy that had ETSs, and field tests, not hand held devices. Of course, all the positives were sent on.

In addition to that, either five or ten percent of the negatives were sent on. That gave a really good handle on the operation of that particular site. It gave good insight into how they were doing.

I believe with that we will close and I will turn this back over to Dr. Bush. Thank you to the panel.

(Applause.)

DR. BUSH: Well, thank you very much for an outstanding Monday. May I look forward to tomorrow, and remind you that we start at 8:15, and we will likely go to close to 6:00 o'clock.

That is an awful lot of good science we are going to review tomorrow. It is going to be a good day. Let's get together and do it again tomorrow morning.

We will ask you to register again tomorrow morning, because we don't know how many more people are coming in this evening, how many new additions we will have to the audience tomorrow.By rule and regulation, we need to account for who is present from the public to participate in our meeting. So, we will ask you to register again tomorrow. Have your business card out and just ready to put on the table. That is the easy way to do that.On that note, we will see you promptly for start at 8:15.

Coffee will be out probably much earlier than that. Thank you. (Whereupon, at 5:22 p.m., the meeting was recessed, to reconvene at 8:15 a.m. the following day, Tuesday, April 29, 1997.) (Whereupon, at 12:01 p.m. the meeting was recessed, to reconvene at 1:15 p.m., that same day.)