Cystic Fibrosis: Tests, Treatments Improve Survival
by Ricki Lewis, Ph.D.

     Alex Deford had been ill almost from the moment of her
birth on Oct. 30, 1971. Her frequent colds and ear infections
coupled with her small size, despite a healthy appetite,
prompted doctors to vaguely diagnose "failure to thrive." When
Alex developed double pneumonia at 4 months, it was clear that
something was very wrong.
     That something turned out to be cystic fibrosis, the most
common inherited illness among white people of Northern and
Western European ancestry, although it is seen in all ethnic
groups. Symptoms include thick, sticky mucus clogging the
lungs, impairing breathing and attracting infection; a blocked
pancreas that cannot release digestive enzymes, causing pain
after eating; stubbed fingers from poor circulation;
infertility; salty sweat; and other problems. Patients may
have any or all of these symptoms--Alex had quite a list.
     When she was diagnosed at Boston Children's Hospital
early in 1972, Alex was so ill that she was expected to live
only days. She survived eight years, but not easily.
     Alex began each day by inhaling a decongestant. Then her
parents took turns providing "postural drainage," a 30- to 60-
minute pounding and pressing on each of 11 segments of the
lungs, to loosen the mucus, which she coughed up. Alex would
then take drugs--antibiotics to prevent lung infection and
powdered digestive enzymes mixed into applesauce.
     Despite this daily regimen, Alex died in January 1980.
Her father, sportswriter and commentator Frank Deford, tells
her story in his book, Alex, the Life of a Child.
     Cystic fibrosis (CF) is inherited and affects 30,000
Americans. In 1989, scientists discovered the gene that causes
cystic fibrosis (see accompanying article.) This discovery is
enabling researchers to develop new diagnostic tests that will
help identify those who can benefit from traditional as well
as several new treatment approaches being evaluated by FDA.
How CF Is Inherited
     CF is typically passed from parents who each carry the
gene, to children of either sex. Carriers have one faulty copy
of the gene, which is responsible for the illness, plus one
normal copy, which prevents symptoms. Each child of carrier
parents has a 1 in 4 chance of inheriting CF; a 1 in 4 chance
of being completely free of the mutant gene; and a chance of
1 in 2 of being a carrier, like the parents.
     Couples usually learn that they carry CF when they have
an affected child. By 1985, individuals who had a sibling with
CF could find out if they carried the gene by taking a
"genetic marker" (linkage analysis) test that spots a
particular family's CF-carrying chromosome, but not the gene
itself. Finding the CF gene makes it possible to detect most
carriers, even if there are no affected relatives.
     The Office of Technology Assessment estimates that 100
million to 200 million people in the United States might want
to take a CF carrier test. About 8 million people in the
United  States, or 1 in 25 whites, may be carriers.

Diagnosing CF
     The same gene discovery that has led to development of
carrier tests is expected to help to more quickly diagnose CF,
whose symptoms resemble those of other illnesses.
     The most widely used and best-known CF test is the
electrolyte sweat test. It detects the excess sodium,
potassium and chloride (charged chemicals called electrolytes)
found on the skin of many people with CF. A physician would
perform a sweat test in a child with unexplained failure to
gain weight, or with very frequent respiratory infections.
     The sweat test evolved from the observations made by a
physician, Dr. Paul di Sant'Agnese, during a 1953 heat wave in
New York City. He was curious why so many children with CF
were being brought to Babies and Children's Hospital, where he
worked, with heat prostration. The youngsters were unable to
cope with the heat because too much salt exited their bodies
in sweat. The fact that the sweat of a person with CF contains
two to six times as much salt as normal sweat gave him the
idea for the sweat test.
     The sweat test became widely used by the mid-1950s, and
is the only CF test cleared by FDA for marketing. (A
forerunner of the sweat test was the observation that a
child's brow was salty when kissed. At the turn of the
century, this is how midwives identified babies with cystic
fibrosis.)
     Although the sweat test is a critical part of a CF
diagnostic work-up, salty sweat can indicate any of several
disorders. Other tests help focus the diagnosis. Some of these
tests are based on methodologies developed by reference
laboratories, which perform medical tests and send the results
to physicians. According to Freda Yoder of FDA's Center for
Devices and Radiological Health, methodologies developed in-
house have not traditionally been regulated by the agency.
     Explains Tom Tsakeris, director of the division of
clinical laboratory devices at FDA, "FDA regulates products,
not laboratories. As long as they are not marketing the test
itself, we do not regulate the lab." However, he adds, the
Clinical Laboratory Improvement Act, signed into law in 1988
but not yet fully implemented, will regulate reference
laboratories.
     One test developed by reference labs measures the amount
of the protein trypsinogen in a newborn's blood. Trypsinogen
is manufactured by the pancreas and sent to the intestine,
where it is snipped to a shorter form, trypsin, which helps
digest proteins. If the pancreas is clogged by the sticky
mucus of CF, trypsinogen levels are elevated, because the
longer protein cannot be cut down to size.
     In one study conducted by researchers at the University
of Colorado School of Medicine and Children's Hospital in
Denver, the trypsinogen test identified 95.2 percent of
infants with CF who did not have the earliest sign, a greenish
discharge called meconium ileus indicating intestinal
blockage. But in the study there were many false positives--of
96 infants who tested high for trypsinogen on two tests, only
31 had CF. So, although the trypsinogen test alone is not
perfect, combined with a sweat  test and observing symptoms,
it can begin to paint a portrait of CF.
     Another test detects the level of certain fetal
intestinal enzymes in the amniotic fluid (the liquid
surrounding the fetus). Amniotic fluid is collected for
testing by a procedure called amniocentesis (see "Genetic
Screening: Fetal Signposts on a Journey of Discovery" in the
December 1990 FDA Consumer). In a fetus with CF, these enzymes
are decreased. Again, however, other disorders besides CF can
produce this finding, and therefore it is not a specific
disease marker. Researchers have turned to the genetic
material to develop a definitive CF test.

Enter Genetic Testing
     Developing a test to detect the gene that causes CF would
provide a definitive diagnosis, because this mutant gene is
the only cause of the disorder. The first step was to find out
where the gene behind CF lies among the 23 pairs of
chromosomes.
     By 1985, several research teams had narrowed the search
to a part of chromosome 7 (the seventh largest chromosome).
Until the CF gene itself was isolated and characterized in
1989, relatives of patients could take an indirect test that
uses linkage analysis. Because of the complexity of test
interpretation, these tests are primarily performed at
academic centers.
     A genetic linkage test tracks a known DNA sequence (a
genetic marker) that, within a family, always occurs in people
with CF, and never in those who do not have the illness. A
genetic marker and the gene responsible for the disorder
behave like two inseparable friends. If you see one at a
party, you know the other is nearby. Genetic linkage testing
is based on the observation that genes carried close together
on the same chromosome tend to be inherited together.
     Ray White at the Howard Hughes Medical Institute at the
University of Utah in Salt Lake City and Robert Williamson of
St. Mary's Hospital Medical School in London each found a
marker, one on either side of the CF gene. Using these two
markers, a couple who already had a child with CF could have
fetal chromosomes tested in a subsequent pregnancy. If the two
markers on the two chromosome 7's in the fetus matched those
of the affected child, then it, too, has likely inherited the
disease.
     A major limitation of linkage tests is that they only
work on families known to have CF. Because people can carry CF
without having symptoms, a disease-causing gene can be in a
family without anyone in recent memory being ill. Finding the
CF gene itself, however, may make possible a test useful on
anyone, so that carriers could be detected in families where
no one has CF.
     Like other genetic tests, CF tests can be performed on
any type of tissue, because all human cells (except red blood
cells) contain two copies of all of the genes, and sperm and
egg have one copy of each. The first CF tests used white blood
cells. Then Williamson's group in London came up with a
pleasanter alternative--a mouthwash! After swishing a
saltwater solution in the mouth, the person spits into a
bottle. The CF gene can be  spotted in cells dislodged from
the inside of the cheek.
     Taking a cue from London, Genzyme Corp. (Cambridge,
Mass.) developed a cheekbrush test for CF, which is
investigational. A patient swabs cheek cells onto a brush, and
the physician sends the sample to Genzyme. The presence of
both normal and mutant CF genes indicates carrier status. If
only mutant genes are there, CF is indicated.
     
To Test or Not To Test?
     A carrier test provides information to couples who are
not ill but whose children are at high risk of inheriting the
condition.
     Many experts predict that the day of universal CF
screening is approaching, with several companies developing CF
tests that simultaneously screen for several CF mutations.
     Two factors contribute to the sensitivity of a CF carrier
test. The first is the number of mutations that can be
detected. The more mutations tested for, the more carriers
will be spotted.
     Ethnic background is the other important factor, says
Marisa Ladoulis, a genetic counselor at Collaborative
Diagnostic Services in Waltham, Mass. For example, a 12-
mutation test that spots 84 percent of whites with a Northern
or Western European background will detect 92 to 95 percent of
Ashkenazi Jews, and the 16-mutation test finds 96 to 98
percent of them.

All CF Mutations Are Not Equal
     Checking for an errant CF gene may be easy, but
interpreting the results may not be. Researchers are finding
that different CF mutations cause different degrees of
sickness. Alex Deford probably had two copies of delta F508,
the most common and one of the more serious mutations that can
cause CF. But a researcher in the laboratory of Francis
Collins, the co-discoverer of the CF gene, has a milder case
of CF because he inherited the delta F508 mutation as well as
a different one.
     This young man must perform postural drainage on himself
and take antibiotics and digestive enzymes, but he also plays
the trumpet, bikes, and sings. Still, a respiratory infection
can send him to the hospital for a week or longer. Clinicians
are finding that some people who have frequent bouts of
pneumonia and other respiratory infections actually have CF.
     Some people with CF may not even have lung or digestive
symptoms. Aubrey Milunsky, D.Sc., Director of the Center for
Human Genetics at the Boston University School of Medicine,
found that some men who were referred to him because they were
having difficulty fathering a child actually had CF. In
examining x-rays that had been taken as part of a standard
fertility work-up, Milunsky noticed the men lacked the vas
deferens, the paired tubes that deliver sperm from the body.
Knowing this is a symptom in 90 percent of men with CF,
Milunsky tested their genes and found they had inherited CF.
     "Cystic fibrosis is not a simple single mutation to look
for," says Margaret Wallace, Ph.D., assistant professor in the
division of genetics in the department of pediatrics at the
University of Florida in Gainesville. "There will be a lot of 
problems in doing the diagnosis and giving an idea of what it
means," she adds.

Treating CF
     CF symptoms are controlled with a number of drugs.
Antibiotic drugs combat infections to which CF patients are
prone, including Pseudomonas aeruginosa bacteria, a type of
microbe that is attracted to the sticky mucus in the lungs.
The combination of animal enzymes, called Viokase, that Alex
Deford took regularly is still used today by CF patients. It
is approved as a prescription digestive aid for CF patients
and others with pancreatic insufficiencies. Combined with a
high-calorie diet, this enzyme preparation aids digestion,
helping the patient to maintain weight.
     Many patients also take anti-inflammatory prescription
drugs, such as ibuprofen (Motrin and others), prednisone
(Deltasone, Winpred, Orason, and others), and naproxen
(Anaprox, Naprosyn and others).
     The drug amiloride (Midamor, Moduretic), introduced in
1967 and approved as an adjunct to treatment with some
diuretic drugs, is now being tested as a treatment for CF.
Scientists believe amiloride thins lung secretions by blocking
sodium uptake by lung cells. Clinical studies are under way to
assess amiloride as a CF treatment alone, and in combination
with the biological products adenosine triphosphate (ATP) and
uridine triphosphate (UTP). (ATP and UTP are components of the
nucleic acids DNA and RNA.)
     Other investigational products are aimed at tempering the
body's immune response to lung infection, which can be
excessive. One such product is deoxyribonuclease. The March
19, 1992, New England Journal of Medicine reported that in a
pilot study, this protein biologic given in an aerosol helped
clear the lungs of 16 adult CF patients. It is being tested in
900 CF patients at 50 medical centers in the United States.

Gene Therapy
     FDA has designated recombinant cystic fibrosis
transmembrane conductance regulator (the gene's protein
product, abbreviated CFTR) as well as gene therapy as orphan
products. This gives their sponsors special incentives because
they are developing products for a condition affecting
relatively few people.
     The first human gene therapy study of CF got under way
last April 17 at the National Heart, Lung, and Blood Institute
after FDA gave the go-ahead the previous day. An engineered
cold virus (adenovirus) was introduced into the cells lining
the nose and airways of a 23-year-old man with CF. The virus
was altered to carry the normal CFTR gene and lacks the genes
to cause a cold and to replicate.
     The research was the first use of gene therapy for a
common genetic disorder and the first use of a cold virus to
transport genes. The study includes 10 patients age 21 or
older who have mild to moderate CF symptoms.
     Previous experiments in rats indicated that replacing the
CF genes in just 10 percent of the lung lining cells improves 
lung function. However, because the genes go to the patients'
lungs but not their sex cells, CF can still be passed to the
patients' children.
     New knowledge of CF is coming so fast that the goals of
carrier screening may change even before the tests are cleared
for marketing.
     Soon, detecting the gene for CF may be a way of finding
who needs treatment, as early as possible, just as is
presently done for high blood pressure and elevated blood
cholesterol. Says Wallace, "CF research is moving so quickly,
with a lot of hope for treatment in the near future. It will
be treatable, and possibly easily." 

Ricki Lewis is a genetic counselor and is the author of
textbooks on biology and human genetics.

Box:
For more information, contact The Cystic Fibrosis Foundation,
6000 Executive Blvd., Suite 510, Rockville, MD 20852;
telephone (1-800) FIGHTCS.
Advances and Stumbling Blocks
     The symptoms of CF were first described in medical
journals in 1938. The malady was attributed to a defect in the
channels leading from certain glands--a remarkably accurate
description, it would turn out. But the disorder was
recognized before it was given a name, as illustrated by the
17th century English saying, "A child that is salty to taste
will die shortly after birth."
     In 1960, a CF patient rarely lived past the age of 12. By
1970, only half lived to see their 18th birthdays. In the
1970s, when postural drainage began to be implemented and FDA
approved enzyme replacement and antibiotic therapy, the
average lifespan began to creep upwards. Today, it is 29
years, according to the Cystic Fibrosis Foundation. New, more
targeted therapies may raise survival age higher.
     Cystic fibrosis researchers marked a medical milestone on
Oct. 8, 1989, when Science magazine published a report by
Francis Collins and his co-workers at the University of
Michigan at Ann Arbor and Lap-chee Tsui at the Hospital for
Sick Children in Toronto on precisely how a specific gene
disrupts a certain protein to cause CF.
     The researchers named the protein the "cystic fibrosis
transmembrane conductance regulator," or CFTR for short. CFTR
is normally manufactured inside cells lining glands in the
respiratory passages, small intestine, pancreas, and sweat
glands. The protein travels to the cell's surface, where it
controls the flow of salt in and out of the cell like a
gateway in the cell membrane.
     In the disorder, CFTR protein is abnormal in a way that
prevents it from reaching the cell's surface. Without the
gateway in the membrane, salt is trapped inside cells.
Following a natural chemical tendency to try to dilute the
salty interiors of cells, moisture is drawn inside them
through other gateways. This dries out the surrounding
secretions, causing symptoms. In most people with CF, the
protein is missing just one amino acid building block out of
1,480--a tiny, but devastating, glitch.
     Almost as soon as Collins and Tsui described the mutation
that causes CF, dubbed delta F508, a difficulty arose. Delta
F508 was not the only way that the gene could be altered. (A
gene consists of sequences of four types of building blocks.
Just as a sentence can have an error in any of its letters, a
gene can be altered in many ways. A person with CF inherits
two abnormal forms.)
     But within days of the publication of the Science report,
several biotechnology firms were already devising carrier
tests for delta F508. A test for the disease-causing gene
variant became available on an investigational basis by
November 1989. But on Feb. 1, 1990, Collins, Tsui, and several
others reported in The New England Journal of Medicine that
only 75.9 percent of white CF patients of Northern and Western
European backgrounds had the delta F508 variant. How useful
would a test for delta F508 be, researchers worried, if this
wasn't the only variant responsible for CF? At current count,
more than 200 variants of the gene are known.
     The multiple guises of the CF gene meant that a test to
spot delta F508 would miss about 24 percent of Northern or
Western  European descended whites in the United States who do
carry a CF gene. This, in turn, meant that the test would find
only about half the couples in the United States who risk
passing CF to a child (this figure is derived by multiplying
the chances of each parent having delta F508). But it would be
too costly to develop a test for more than 200 different
mutations, when only a few of them are common.
     Adding to the complexity is that different populations
have different proportions of the CF gene variants. For
example, delta F508 occurs in only 35 percent of African-
Americans and Jews of Central and Eastern European ancestry
(called Ashkenazi) who carry CF, making the test for this
mutation even less valuable than it is for non-Jewish whites.
For Hispanics and Italians, the frequency of delta F508 is 50
percent.
     The potential powder keg of a carrier test for a common
genetic disease that would, at best, only work three-quarters
of the time set off a flurry of statements by professional
medical organizations. On Nov. 13, 1989, the American Society
of Human Genetics urged caution in carrier testing until a
greater percentage of the CF-carrying population could be
identified, calling for pilot programs to test the tests.
Meanwhile, they suggested the test only for those with a close
affected relative.
     In early March 1990, a panel of physicians, geneticists,
genetic counselors, and attorneys met at the National
Institutes of Health in Bethesda, Md., to develop guidelines
for CF carrier testing. This group echoed the earlier call for
pilot programs, adding that widespread testing should wait
until tests could detect 90 to 95 percent of carriers.
     In December 1992, the American Society of Human Genetics
reevaluated their 1989 statement, in light of the ability to
detect many CF mutations. Their advice remains unchanged--for
now, CF testing should be offered only to those with a
relative who has the disorder. The organization also calls for
informed consent and genetic counseling, confidentiality of
results, and quality control of the laboratory performing the
test. 
--R.L.
                            ####