UNITED STATES OF AMERICA, PETITIONER V. GENERIX DRUG CORPORATION, ET AL. No. 81-1222 In the Supreme Court of the United States October Term, 1981 On Writ of Certiorari to the United States Court of Appeals for the Eleventh Circuit /1/ Brief for the United States TABLE OF CONTENTS Opinions below Jurisdiction Statutes involved Statement Summary of argument Argument: I. The explicit language of the Act establishes that the Act's "new drug" provisions apply to generic drug products A. The statutory definitions of the terms "drug" and "new drug" encompass generic drug products B. The Act's new drug approval provisions require that FDA review new drug products II. The legislative history of the Act confirms that generic drug products are not exempt from the requirement of premarketing approval A. The 1938 Act was intended by Congress to prohibit the marketing of any potentially unsafe drug product B. The 1962 Amendments did not exempt generic drug products from the new drug approval provisions of the Act III. Premarketing approval of generic drug products is necessary to effectuate the fundamental purpose of the Act A. Premarketing review is the only means by which FDA can protect the public from unsafe or ineffective generic drug products B. Premarketing approval of generic drug products promotes rather than impairs competition in the marketing of safe and effective generic drugs Conclusion OPINIONS BELOW The opinion of the court of appeals (Pet. App. 1a-15a) /2/ is reported at 654 F.2d 1114. The decision of the district court (Pet. App. 18a-31a) is reported at 498 F. Supp. 288. JURISDICTION The judgment of the court of appeals (Pet. App. 16a-17a) was entered on September 4, 1981. On November 24, 1981, Justice White extended the time within which to file a petition for a writ of certiorari to and including January 2, 1982. The petition was filed on December 30, 1981, and was granted on March 1, 1982 (J.A. 66). The jurisdiction of this Court rests on 28 U.S.C. 1254(1). STATUTES INVOLVED The relevant statutory provisions are set forth at Pet. App. 33a-44a. QUESTION PRESENTED Whether a prescription drug product -- even though potentially unsafe or ineffective -- is exempt from the requirement of premarketing approval by the Food and Drug Administration merely because it contains the same active ingredient as another drug that has received FDA approval or that is generally recognized as safe and effective. /*/ STATEMENT 1. Section 505(a) of the Federal Food, Drug, and Cosmetic Act ("the Act"), 21 U.S.C. 355(a), prohibits any person from introducing into interstate commerce any "new drug" that has not been approved by the Food and Drug Administration ("FDA"). /3/ The term "new drug" is defined in Section 201(p) of the Act, 21 U.S.C. 321(p), as any drug that "is not generally recognized, among experts qualified by scientific training and experience to evaluate the safety and effectiveness of drugs, as safe and effective for use under the conditions prescribed, recommended, or suggested in the labeling thereof," /4/ or that has not been used to a material extent and for a material time under other than experimental conditions. /5/ Generix Drug Corporation ("Generix") is a distributor of generic drugs, i.e., drugs that purport to be copies of brand name or "pioneer" drugs manufactured by other companies. While the "active" ingredients of these generic drugs purport to be identical to those contained in the pioneer drugs, the "inactive" ingredients (e.g., coatings, binders) may differ. Many of the generic drugs Generix sold in interstate commerce had not been approved by the FDA and, in the government's view, were "new drugs" within the meaning of the Act. After a series of warnings, respondents continued to distribute a variety of unapproved prescription drugs (J.A. 11-15, 17). Accordingly, on November 29, 1979, the United States filed an action in the United States District Court for the Southern District of Florida against Generix and several of its officers and employees. The complaint requested both preliminary and permanent injunctive relief. /6/ Generix did not deny that the particular drug products referred to in the complaint had not been approved by FDA. /7/ The hearing on the government's request for a preliminary injunction focused, therefore, on the question whether the generic drugs distributed by Generix were "new drugs" within the meaning of 21 U.S.C. 321(p). In arguing that they were not, Generix relied on the fact that FDA had approved other drug products containing the same active ingredients as Generix's drugs. Such prior FDA approval, Generix argued, was evidence that the active ingredients were generally recognized as safe and effective. From this premise, Generix asserted that its products were not "new drugs" even if their inactive ingredients and methods of compounding were not identical to those used in the products previously approved by FDA. /8/ Accordingly, Generix contended that the prescription drugs it distributed were subject to no FDA clearance whatever. The government argued that FDA's prior approval of prescription drugs containing the same active ingredients as the drugs marketed by Generix was irrelevant because the Act requires FDA approval of every drug product unless that product is generally recognized as safe and effective and has been used to a material extent and for a material time. The government's evidence established that Generix's unapproved drug products were not generally recognized by qualified experts as safe and effective (Tr. 49-52, 93-94, 96-97). In addition, the government presented evidence showing why distribution of unapproved generic drugs poses a danger to the public health. As expert witnesses explained, prescription drugs generally contain one or more active ingredients, which may comprise 10% or less of the total compound. They also contain inactive ingredients, known as excipients, whose purposes are to coat or bind the active ingredients into a convenient form and to control their solubility (Tr. 83-84). Because of differences in inactive ingredients or in methods of compounding, drugs containing the same active ingredients can differ from each other in the amounts of the active ingredient that enter the bloodstream and the rate at which this occurs. /9/ Since the quantity and speed of delivery of active ingredients are critical factors that determine the effect of a drug on the patient, it cannot be assumed that a copy of an approved new drug will achieve the same therapeutic effect as the pioneer product (Tr. 39-40, 46-67). Often it does not. /10/ The unapproved drugs distributed by Generix include drugs prescribed for the treatment of serious and potentially life-threatening conditions such as high blood pressure (Tr. 47-48). The government's experts demonstrated that differences in formulation could affect the safety or effectiveness of such drugs (Tr. 94-98). Moreover, because the labeling of these generic products is virtually identical to that of the brand name products they copy, and because Generix markets these unapproved drugs in a way that would lead physicians to believe that they are interchangeable with approved drugs (Tr. 45-47), they create serious dangers to patients. /11/ Finally, the evidence showed that doctors and pharmacologists generally regard premarketing testing and approval as necessary to insure the safety and effectiveness of drugs. Because of the serious dangers that may be created by differences in bioavailability, physicians would not knowingly prescribe an unapproved generic drug (Tr. 55-47). /12/ Their willingness to prescribe generic drugs at all is based on their assumption that generic drugs have received FDA approval. 2. On June 12, 1980, the district court issued an order granting in part and denying in part the government's motion for a preliminary injunction (Pet. App. 18a-31a). Rejecting both the government's and Generix's interpretations of the Act, the court held that the government had the burden of showing a potential hazard associated with Generix's drugs. /13/ The court then found that the government had presented evidence that differences in bioavailability could affect the safety and effectiveness of drugs containing any one of six active ingredients or combainations of ingredients /14/ and enjoined distribution of those drugs. Because such specific evidence was not presented as to Generix's other unapproved generic drugs, /15/ however, the court denied the government's motion for a preliminary injunction against distribution of those other products (id. at 31a). /16/ The preliminary injunction was stayed pending Generix's appeal (id. at 32a). 3. The court of appeals reversed that portion of the district court's order granting a preliminary injunction and remanded the case to the district court with instructions to dismiss the complaint. The court of appeals held that the term "new drug" applies only to active ingredients and not drug products. Therefore, the court concluded, a generic drug product containing the same active ingredient as an approved pioneer drug /17/ is not a "new drug" and need not comply with the Act's requirement that all new drugs be approved by FDA before they are marketed. The court expressly declined to follow the recent decision in Premo Pharmaceutical Laboratories, Inc. v. United States, 629 F.2d 795 (1980), in which the Second Circuit held that the Act's new drug definition applies to drug products, not to active ingredients alone, and that FDA approval of a brand name drug does not exempt other new drugs that contain the same active ingredient from the requirement of premarketing approval. The court of appeals acknowledged that there may be differences in therapeutic effect between a pioneer drug product that has been approved by FDA, and unapproved generic drugs having the same active ingredient but different inactive ingredients (Pet. App. 3a). It also noted the evidence that these differences "may effect (sic) the safety and efficacy of the generic drug product" (ibid.). But the court of appeals construed the Act not to protect the public from these dangers. The court concluded: "the term 'new drug' as used in (21 U.S.C.) Section 321(p) applies only to the active ingredient of a drug product" (id. at 15a). The bases for this conclusion were the court's assumptions that "the drugs listed in the sources described in subsection (A) of (21 U.S.C.) Section 321(g)(1) are 'drug substances' not drug products" and that "(i)f 'new drug' refers to the drug product as opposed to the active ingredient, a drug could never 'muster the requisite scientifically reliable evidence of effectiveness' in order to 'drop out of active regulation by ceasing to be a new drug'" as is contemplated by the Act (id. at 5a, 7a). /18/ The court of appeals considered the new drug approval procedures of 21 U.S.C. 355 unnecessary in the case of generic drugs because, in the court's view the Act gave the FDA other means to regulate bioavailability and bioequivalence. The court, in addition, concluded that premarketing review of generic drugs by FDA would be contrary to Congressional intent "to encourage generic-name prescribing and thereby lower prices" (Pet. App. 10a). Because the court of appeals concluded that in no circumstances is prior FDA approval required for a drug containing the same active ingredient as a previously approved new drug, it vacated the preliminary injunction -- even as to those drugs for which there was evidence that lack of bioequivalence could be dangerous to the patient-- and ordered the complaint dismissed (Pet. App. 15a). SUMMARY OF ARGUMENT The court of appeals' opinion ignored the plain meaning of the Federal Food, Drug, and Cosmetic Act and, consequently, reached a result at odds with clear Congressional intent and sound public policy. As a result, the court of appeals has sanctioned the distribution of unapproved -- and, as far as the record reveals, untested -- prescription drugs. Unless the decision below is reversed, its inevitable result will be the ingestion of such unapproved drugs by patients suffering from life-threatening ailments, patients who are in need of the assurances of safety and effectiveness the Act mandates. The Federal Food, Drug, and Cosmetic Act prohibits the marketing of any "new drug" without prior FDA approval. 21 U.S.C. 355(a). A new drug is any drug that is not generally recognized by experts as safe and effective or that has not been used to a material extent and for a material time. 21 U.S.C. 321(p). The statutory language expressly provides that a generic drug product is a drug and that the Act's prohibition on marketing of unapproved new drugs applies to generic drug products. That construction of the Act is supported by the Act's legislative history and is necessary to effectuate the Act's underlying purpose -- protection of the public health. 1. The Act defines "drug" to include "articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals." 21 U.S.C. 321(g)(1)(B). Generic drug products clearly fit within this definition. Since generic drug products are drugs, 21 U.S.C. 355 prohibits the marketing of any generic drug product without FDA approval unless "such drug," i.e., the generic product itself, is "generally recognized, among experts * * * as safe and effective" and has "been used to a material extent" and "for a material time." 21 U.S.C. 321(p)(1), (2). The product-specific statutory requirements for new drug applications (21 U.S.C. 355(b)) further support the conclusion that general recognition of a pioneer drug does not exempt other products containing the same active ingredient from the requirement of premarketing approval. Since evidence about the safety and effectiveness of a pioneer drug product would not be sufficient for approval of a new drug application for a generic copy of the pioneer product, such evidence is not a sufficient basis for an expert consensus that a generic copy is safe and effective. Weinberger v. Hynson, Westcott & Dunning, Inc., 412 U.S. 609, 632 (1973). 2. The legislative history of the new drug provisions of the Act confirms that each drug product that has not itself achieved general recognition among experts must be approved by FDA before it is marketed. The direct stimulus for the enactment of premarketing approval requirements in 1938 was the fact that numerous deaths had been caused by a new product that, although it contained an active ingredient previously used in other products, was rendered toxic by the use of an untested inactive ingredient. The 1962 amendments were intended to strengthen the protection afforded the public from potentially unsafe or ineffective drugs, and nothing in those amendments exempts generic drug products from premarketing approval. 3. Premarketing approval of generic drug products is necessary to effectuate the Act's fundamental purpose -- protection of the public health. Unapproved generic drugs may be unsafe or ineffective, and measures available to remove them from the market after they have caused harm simply do not provide the protection Congress intended. Moreover, a requirement of premarketing approval will not hamper competition in the drug industry. To the contrary, competition will be promoted by assuring that all prescription drugs on the market have satisfied the statutory standards of safety and efficacy. In the absence of the assurance provided by premarketing approval, careful doctors and pharmacists would be deterred from prescribing or dispensing generic drug products, despite lower prices. ARGUMENT I. THE EXPLICIT LANGUAGE OF THE ACT ESTABLISHES THAT THE ACT'S 'NEW DRUG" PROVISIONS APPLY TO GENERIC DRUG PRODUCTS This case presents a narrow question of statutory construction with broad consequences for the public health and the future of competition in the distribution of safe and effective prescription drugs. Since it is undisputed that respondents distributed generic prescription drugs without FDA premarketing approval, the issue before this Court is whether such drugs are exempt from the statutory preclearance scheme. Respondents' primary contention is that their products are not "new drugs" within the meaning of the Act. In support of this position, respondents have at various stages of this litigation relied on two principal arguments. First, they assert that their drugs are not "new drugs" because they contain the same active ingredients as other drugs that have received FDA approval, even though the inactive ingredients may differ. Thus, it is respondents' position that their drugs should be presumed to be safe and effective and, consequently, exempt from FDA review, even if a difference in ingredients results in dissimilar effects on the human body. Second, respondents contend that the active ingredients in the pioneer drugs their products purport to copy have received general expert recognition as safe and effective and have been used to a material extent and for a material time, and that therefore the purported copies are not "new drugs," even though, again, the copies may contain different inactive ingredients, may have dissimilar effects on the human body, have themselves received no expert recognition, and have not been used to a material extent and for a material time. /19/ Respondents' positions fly in the face of the clear statutory language. /20/ Accordingly, we turn first to the express language of the Act to demonstrate that the prescription drugs at issue were, in the absence of premarketing approval, distributed unlawfully, as the Second Circuit held in Premo Pharmaceutical Laboratories, Inc. v. United States, supra. /21/ A. The Statutory Definitions of the Terms "Drug" and "New Drug" Encompass Generic Drug Products Section 505(a) of the Act, 21 U.S.C. 355(a), prohibits the marketing of "any new drug, unless an approval of an application filed pursuant to (21 U.S.C. 355(b)) is effective with respect to such drug." The Act defines a "new drug," i.e., a drug that cannot be marketed without prior FDA approval, as: "Any drug * * * the composition of which is such that such drug is not generally recognized, among experts * * * as safe and effective" or "(a)ny drug * * * which has not * * * been used to a material extent or for a material time." 21 U.S.C. 321(p)(1), (2) (emphasis added). 1. The term "drug," as used in the definition of "new drug," is broadly defined in the Act (21 U.S.C. 321(g)(1)) to include: (A) articles recognized in the official United States Pharmacopoeia, official Homoeopathic Pharmacopoeia of the United States, or official National Formulary, or any supplement to any of them; and (B) articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals; and (C) articles (other than food) intended to affect the structure or any function of the body of man or other animals; and (D) articles intended for use as a component of any article specified in clauses (A), (B), or (C) of this paragraph; but does not include devices or their components, parts, or accessories. Generic drug products are within the literal language of at least one of four subsections of this statutory definition, which this Court has held must be interpreted to provide coverage "as broad as its literal language indicates." United States v. An Article of Drug . . . Bacto-Unidisk, 394 U.S. 784, 798 (1969). As the Second Circuit held in Premo Pharmaceutical Laboratories, Inc. v. United States, supra, generic drug products are "drugs" because they are "articles intended for use in the diagnosis, cure, mitigation, treatment, or prevention of disease in man or other animals." 21 U.S.C. 321(g)(1)(B). In holding that only active ingredients and not drug products are drugs, the court of appeals in this case ignored the literal language of the statutory definition, disregarding 21 U.S.C. 321(g)(1)(B). Indeed, the court entirely omitted subsection (B) of 21 U.S.C. 321(g)(1) from its discussion, and failed to explain why it thought that a generic drug product would not fit within this definition. Rather, the court considered only another subsection of the definition of "drug" (Pet. App. 5a), 21 U.S.C. 321(g)(1)(A), and even that subsection does not support its conclusion. Thus, the court stated that the "record shows that the drugs listed in the sources described in subsection (A) * * * are 'drug substances,' not drug products" (Pet. App. 5a); but there is nothing to that effect in the record. On the contrary, the pharmacopoeia mentioned in 21 U.S.C. 321(g)(1)(A) contain monographs for drug products as well as active ingredients. /22/ More importantly, however, an article need fall within only one of the subsections of 21 U.S.C. 321(g)(1) in order to be a "drug." /23/ Therefore, regardless of the scope of 21 U.S.C. 321(g)(1)(A), generic drug products are drugs as defined in 21 U.S.C. 321(g)(1)(B). /24/ 2. Since a generic drug product is a drug, the language of 21 U.S.C. 321(p) compels the conclusion that a generic drug product is a "new drug" if "such drug," i.e., the generic drug product, is not generally recognized by experts as safe and effective or has not been used to a material extent and for a material time. There is nothing in the statutory language to support the court of appeals' conclusion that a generic drug product can escape "new drug" status by virtue of general recognition of the safety and effectiveness of a pioneer drug product containing the same active ingredient or of the active ingredient itself. B. The Act's New Drug Approval Provisions Require That FDA Review New Drug Products The conclusion that a generic drug product cannot escape "new drug" status unless the generic product is itself recognized by experts as safe and effective and has been used to a material extent and for a material time is further supported by the need to harmonize the definition of "new drug" with the Act's requirements for obtaining approval of a new drug application ("NDA"). 21 U.S.C. 355(b)-(e). See Weinberger v. Hynson, Westcott & Dunning, Inc., supra, 412 U.S. at 631-632. 1. The statutory provisions governing NDAs (21 U.S.C. 355(b)) show that the FDA's premarketing review of a "new drug" is a review of the safety and effectiveness of a drug product, i.e., of the combined effect of specific active and inactive ingredients. The Act requires the applicant to submit information about the components, the methods of manufacturing, and the labeling of the drug product for which approval is sought. /25/ As the record in this case demonstrates, this information will be different for each new drug product, even if it has the same active ingredient as a previously approved drug product. Thus, one of the government's expert witnesses testified in this case that (Tr. 102-103): When you say it's "safe and effective," that depends entirely upon the drug delivery system you put it in. It's not safe if you give it in the wrong dose. Until you define the drug delivery system * * * you're dealing with, one can't say that allopurinol is safe and effective in the absence of the drug delivery system. You can say that allopurinal is a pharmacologically active drug which, in the proper use and circumstances, it can be safe and effective, but until you define those characteristics in a situation in which it is going to be used, then the terms are meaningless in may opinion. * * * * * You can't say that something's safe and effective, again, a drug substance, until you define how you are going to deliver it to the body. Is it safe and effective, given intravenously? * * * Is it safe and effective by a drug delivery system that isn't absorbed? Until you define the total performance of a product, you can't make judgments on safety and efficacy of the way it will be used in humans. In enacting the statutory approval process, Congress provided for consideration of these factors. Thus, on the basis of the information included in the application, the FDA must determine whether "adequate tests by all methods reasonably applicable" show that "such drug," i.e., the drug product described in the application, "is safe for use under the conditions prescribed, recommended, or suggested in the proposed labeling thereof" and whether "there is * * * substantial evidence /26/ that the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the proposed labeling thereof." 21 U.S.C. 355(d). /27/ Since inactive ingredients and methods of compounding can affect the safety and effectiveness of a drug product, FDA's determination that the requisite evidence of safety and effectiveness has been presented for a particular drug product cannot be viewed as a determination as to other products containing the same active ingredient. This obvious conclusion is supported by the testimony in this case, which reveals that 90-95% of those generic drugs that are not bioequivalent to the pioneers lack bioequivalence precisely because of their inactive ingredients or methods of manufacture (Tr. 83-84). As this Court has recognized, therefore, each NDA is applicable only to a specific drug product. USV Pharmaceutical Corp. v. Weinberger, 412 U.S. 655, 664 (1973). Accordingly, approval of a pioneer drug product does not permit the marketing of an unapproved generic drug containing the same active ingredient. 2. The exclusion from the "new drug" definition of drugs that are generally recognized by experts as safe and effective and have been used to a material extent and for a material time cannot be construed "to deprive FDA of jurisdiction over a drug which, if subject to FDA regulation, could not be marketed" because of the absence of sufficient evidence of its safety or effectiveness. Weinberger v. Hynson, Westcott & Dunning, Inc., supra, 412 U.S. at 631. Such a construction of the statute would erroneously "impute to Congress a purpose to paralyze with one hand what it sought to promote with the other." Ibid. Therefore, a drug cannot be generally recognized as safe and effective unless there is an "expert consensus" that it is safe and effective and that consensus is based on tests showing safety and effectiveness that would be sufficient to warrant approval of an NDA for that product and that are available to experts in the scientific literature. 412 U.S. at 629-634; Weinberger v. Bentex Pharmaceuticals, Inc., 412 U.S. 645, 652 (1973). /28/ Under the Act, however, evidence of the safety and effectiveness of a pioneer drug product would not be sufficient to warrant approval of a generic drug containing the same active ingredient (see pages 19-22, supra). Therefore, even if the required expert consensus and supporting evidence in the scientific literature exist for a pioneer drug product, those facts do not provide a sufficient basis for a finding that a generic drug with the same active ingredient is not a "new drug." /29/ Expert consensus supported by published evidence about the generic product itself is required. In this case it was not disputed that there are no published studies from which experts could have reached a consensus that respondents' generic prescription drugs were safe and effective. /30/ Therefore, these products are "new drugs" under the plain language of the Act, and it is a violation of the Act to market them without prior FDA approval. 3. The court of appeals incorrectly assumed that FDA approval of a pioneer drug is equivalent to an agency finding that the active ingredient of that pioneer drug is "generally recognized" as safe and effective. In deciding whether to approve a drug product, FDA does not -- and indeed could not -- focus solely on the safety and effectiveness of the active ingredient. Rather, as required by the Act, it evaluates the drug product as a whole, including its inactive ingredients and methods of manufacture. 21 U.S.C. 355(b). See pages 19-21, supra. Moreover, FDA's approval of a new drug application is not a finding that the drug is generally recognized as safe and effective, and FDA approval does not mean that the approved drug is no longer a new drug. This Court has held that: the Act is designed so that drugs on the market, unless exempt, will have mustered the prerequisite scientifically reliable evidence of effectiveness long before they are in a position to drop out of active regulation by ceasing to be a 'new drug.' Weinberger v. Hynson, Westcott & Dunning, Inc., supra, 412 U.S. at 631 (emphasis added). /31/ 4. The court of appeals apparently believed that if general recognition attached to products rather than active ingredients, no drug product could ever achieve "not new" status and the statutory distinction between a "new drug" and other drugs therefore would be "superfluous" (Pet. App. 7a). To the contrary, as this Court clearly contemplated in Hynson, it is possible for a drug product to achieve expert general recognition and pass out of "new drug" status at some point after being marketed for a period of time under an NDA. 412 U.S. at 631. The court of appeals' conclusion that a drug product could never become recognized as safe and effective was based on its incorrect assumption that recognition of a drug product's safety and effectiveness would require general expert knowledge of the ingredients and manufacturing processes of the product, which usually are trade secrets and would not become known to experts generally (Pet. App. 7a). In fact, published test results showing safety and effectiveness, on which expert general recognition is based, are tests of products -- not active ingredients. Thus, experts can and do form opinions about a drug product on the basis of these tests without knowing how a drug was produced or what ingredients it contains. /32/ II. THE LEGISLATIVE HISTORY OF THE ACT CONFIRMS THAT THE GENERIC DRUG PRODUCTS ARE NOT EXEMPT FROM THE REQUIREMENT OF PREMARKETING APPROVAL The legislative history of the new drug provisions of the 1938 Act, and of the 1962 amendments to those provisions, confirms that each drug product that has not itself achieved expert general recognition must be approved by the FDA before it is marketed. A. The 1938 Act Was Intended by Congress to Prohibit the Marketing of Any Potentially Unsafe Drug Product Prior to the passage of the Act in 1938, drugs could be marketed without any testing or governmental review. Following an incident in which numerous deaths were caused by the drug Elixir Sulfanilamide, however, Congress determined that removal of drugs from the market after they had caused harm simply was not an adequate means of protecting the public. It therefore enacted a system of premarketing review by FDA "to prevent the premature marketing of new drugs not properly tested for safety." H.R. Rep. No. 2139, 75th Cong., 3d Sess. 9 (1938). Elixir Sulfanilamide was itself an untested drug composed of a widely used active ingredient and a new inactive ingredient; the inactive ingredient was responsible for the deaths caused by the drug. /33/ Set against this historical background, it would be contrary to clear congressional intent for the requirement of premarketing review to be eliminated for generic drug products merely because they purport to contain the same active ingredient as other drugs that have been proved safe and effective. A statute with such an exemption would not have prevented the Elixir Sulfanilamide tragedy that first caused Congress to recognize the need for premarketing review. /34/ Similarly, it would not protect the public from generic drug products that may be unsafe or ineffective because of their inactive ingredients. Therefore, generic drugs must be subject to the new drug provisions on a product-by-product basis to implement the Congressional determination that removal of unsafe drugs from the market after they have caused harm simply is not adequate to protect the public. See, e.g., United States v. Premo Pharmaceutical Laboratories, Inc., supra, 511 F. Supp. at 988-989. B. The 1962 Amendments Did Not Exempt Generic Drug Products From The New Drug Approval Provisions Of The Act 1. The 1962 amendments to the Act were adopted in the wake of great public concern over thalidomide, a drug that had caused numerous cases of severe birth defects in European children whose mothers had taken that drug during pregnancy. Congress recognized that it was only because of the Act's premarketing clearance procedures that thalidomide was kept off the market in the United States. /35/ There is nothing in the language or the legislative history of the 1962 amendments that suggests that Congress meant to weaken the new drug provisions by exempting generic drugs from the premarketing approval requirements of 21 U.S.C. 355. 2. To protect the public from ineffective drug products, Congress amended the Act to prohibit the marketing of drugs whose effectiveness has not been established. /36/ Thus, Congress added to the statutory requirement of adequate testing to show safety the further requirement that substantial evidence of effectiveness must be demonstrated. A corresponding change was made in the definition of "new drug"; drugs cannot excape "new drug" status unless their effectiveness as well as their safety is generally recognized by qualified experts. The amendment of the Act and Congress' desire to promote the use of generic drug products did not, as the court of appeals thought, mean that Congress intended to exempt generic products from the approval process. Senator Kefauver himself, in first proposing the amendment of the new drug provisions to require that FDA review products for effectiveness as well as safety, premised his proposal on his understanding that the review would be on a product-by-product basis. Only if FDA reviewed the effectiveness of each product, he argued, could generic brands compete effectively with pioneer products: Despite the fact that different brands are identical generically, the contention was made at the hearings that identical products which met the safety requirements of the law, nevertheless, were for various reasons not equal in therapeutic quality or efficacy. I am satisfied that much of the high cost of drug products has come about as a result of this claim, whether it is fact or fiction. Hence, the requirement of this bill that the Food and Drug Administration pass on the efficacy as well as the safety of new drugs should make it more difficult to contend successfully that there is a difference between two brands which are identical generically and both cleared as to safety and efficacy. Thus, promotion costs of new drugs attributable to this contention could be saved and the price of new drugs lowered accordingly. 106 Cong. Rec. 12665 (1960). Congress also increased the protection afforded the public by eliminating the procedure whereby an NDA became effective automatically in the absence of FDA action to disapprove it. Under the Act as amended, "a new drug application is not effective unless it has affirmative approval both for safety and effectiveness." H.R. Rep. No. 2464, 87th Cong., 2d Sess. 10 (1962). See also S. Rep. No. 1744 (Pt. 2), 87th Cong., 2d Sess. 4-5 (1962). 3. Contrary to the court of appeals' interpretation (Pet. App. 10a-12a), the addition of the "established name" labeling requirements (21 U.S.C. 352(e)) /37/ does not indicate any Congressional intent to permit the marketing of potentially unsafe or ineffective generic drug products. Rather, Congress sought to facilitate substitution only of safe and effective generic drugs, /38/ thereby encouraging cost savings and competition in the drug industry. /39/ Congress assumed that no drug -- pioneer or generic -- would reach the market unless its safety and effectiveness were established through general recognition or an approved NDA. /40/ There is nothing in the legisative history to support the view that Congress intended to subject the public to the risk that generic drug products, while they might be lower priced, would not be safe and effective. III. PREMARKETING APPROVAL OF GENERIC DRUG PRODUCTS IS NECESSARY TO EFFECTUATE THE FUNDAMENTAL PURPOSE OF THE ACT A. Premarketing Review Is The Only Means By Which FDA Can Protect The Public From Unsafe Or Ineffective Generic Drug Products Narrowing the definition of "drug" to exempt generic drug products from the new drug approval requirements, as the court of appeals has done, is contrary not only to the specific statutory language and legislative history discussed above but also to the fundamental principle -- protection of the public health -- that has guided the courts in interpreting the Act. As this Court noted in United States v. Dotterweich, 320 U.S. 277, 280 (1943): The purposes of (the Act) * * * touch phases of the lives and health of people which, in the circumstances of modern industrialism, are largely beyond self-protection. Regard for these purposes should infuse construction of the legislation if it is to be treated as a working instrument of government and not merely a collection of English words. See also Kordel v. United States, 335 U.S. 345, 349 (1948); United States v. Sullivan, 332 U.S. 689, 696 (1948). /41/ Moreover, as this Court has held: "Congress fully intended that the Act's coverage be as broad as its literal language indicates -- and * * * broader than any strict medical definition might otherwise allow." United States v. An Article of Drug . . . Bacto-Unidisk, supra, 394 U.S. at 798. See also United States v. Rutherford, 442 U.S. 544 (1979). To exempt generic drug products from premarketing review would be contrary to the Act's basic intent because it would remove the only effective means of protecting the public from potentially unsafe or ineffective generic drugs. 1. As wer have shown (pages 19-22, supra), because of differences in their ingredients, methods of compounding, and therapeutic effect, FDA review of pioneer drugs does not insure that generic drugs are safe and effective. Nor could FDA's power to institute post-distribution enforcement actions prevent the marketing of unsafe or ineffective generic drugs if premarketing approval were not required. Indeed, if FDA had to institute an enforcement action and demonstrate that an unapproved generic drug product was in fact unsafe or ineffective, the burden of proving safety and effectiveness that Congress placed squarely on manufacturers would be reversed, and unsafe or ineffective products could be marketed to the detriment of the public until FDA gathered the requisite evidence and instituted enforcement actions. Generix's argument (Br. in Opp. 27) that the government should be required to establish that specific unapproved drugs are harmful harks back to the days prior to the 1938 passage of the new drug provisions, when potentially harmful drugs could be placed on the market with impunity until the government was able to prove them unsafe. Under current law, however, it is clear that when the "new drug" status of a drug is at issue, the government has no burden to show that the drug is in fact unsafe or ineffective. /42/ For this reason, the actual safety and effectiveness of the drug is irrelevant to the question whether the drug is "generally recognized" as safe and effective. See, e.g., Premo Pharmaceutical Laboratories, Inc. v. United States, supra, 629 F.2d at 803-804; United States v. 1,048,000 Capsules . . . Afrodex, 494 F.2d 1158, 1160 (5th Cir. 1974); Tyler Pharmacal Distributors, Inc. v. United States Department of Health, Education & Welfare, 408 F.2d 95, 99 (7th Cir. 1969). 2. Contrary to the court of appeals' assumption (Pet. App. 13a-14a), the adulteration provisions of the Act, 21 U.S.C. 351, as amended in 1962, also are inadequate to prevent the marketing of unsafe or ineffective generic drugs "without subjecting the drug product to the new drug approval process" (Pet. App. 13a). The new drug approval requirement and the requirements of 21 U.S.C. 351 address two different problems, and both are necessary to protect the public health. /43/ By requiring the applicant to describe the components of a new drug and the process by which it is manufactured, and to present evidence that that drug is safe and effective, 21 U.S.C. 355(b) assures that FDA will receive the information it needs to determine whether the product described in an NDA is safe and effective. The complementary purpose of the adulteration provision, 21 U.S.C. 351, is to assure that every batch of the drug product as manufactured is the product whose safety and effectiveness have been demonstrated in the NDA or by expert recognition. Thus, 21 U.S.C. 351 provides that a drug is "adulterated" and may not be marketed if it does not have the safety, identity, strength, quality and purity characteristics the product purports or is represented to possess. /44/ Under the 1938 Act, FDA had only limited power to protect the public from drugs that were adulterated because of deficiencies in strength, quality or purity. The enforcement tools available to the government -- seizure, injunction, and criminal prosecution -- could not be used until offending products had been marketed; even then the government had to prove that the product was deficient. In 1962, therefore, Congress expanded the adulteration provisions to prohibit the marketing of drugs by manufacturers who did not adhere to good manufacturing practices, i.e., the procedures necessary to assure that the manufacturer follows its "recipe" in producing a drug product. 21 U.S.C. 351(a)(2)(B). /45/ So that FDA could enforce the good manufacturing practices requirement, Congress increased FDA's authority to inspect prescription drug manufacturers, 21 U.S.C. 374(a), and required manufacturers periodically to register uith FDA, 21 U.S.C. 360. /46/ The adulteration prohibitions, even with the addition of the good manufacturing practices provision, however, provide no mechanism for assuring that a generic drug product's effect will be the same as that of a pioneer drug product containing the same active ingredient. Nothing in 21 U.S.C. 351 or in the related FDA regulations (21 C.F.R. Parts 210-211) prohibits differences in inactive ingredients or methods of compounding that may cause generic products to have effects on the patient different from those of the pioneer drug. United States v. Premo Pharmaceutical Laboratories, supra, 511 F. Supp. at 980. /47/ If differences between the "recipe" for a generic drug and that for the pioneer result in a lack of bioequivalence between the two drugs, that problem obviously is not cured by the generic drug manufacturer's careful adherence to the generic drug recipe. Thus, the good manufacturing practices requirements simply are not a substitute for premarketing review of safety and effectiveness of each individual generic drug product. B. Premarketing Approval of Generic Drug Products Promotes Rather Than Imapirs Competition in the Marketing of Safe and Effective Generic Drugs The court of appeals (Pet. App. 9a-11a) also was incorrect in its assumption that exempting generic drug products from the Act's "new drug" premarketing approval requirements is necessary to effectuate the Congressional purpose of improving competition in the drug industry. To the contrary, generic prescription drugs would be less likely to compete effectively with pioneer drugs if the public were deprived of the safeguards afforded by FDA premarketing review. 1. It is not necessary to exempt generic drugs from the new drug provisions of the Act in order to allow competitive marketing of safe and effective generic drugs. Indeed FDA has sought to minimize the burden of filing and obtaining approval of generic drug NDAs to the extent consistent with insuring that the public is not subjected to unsafe or ineffective drugs. /48/ For many generic prescription drugs, FDA has determined that complete safety and effectiveness studies need not be submitted by each manufacturer in order to satisfy the requirements of 21 U.S.C. 355(d). Instead, manufacturers of certain generic drugs may obtain FDA approval b) submitting an abbreviated new drug application ("ANDA"). See 21 C.F.R. 314.1(a) and (f). /49/ Under this procedure, the manufacturer may establish that the generic prescription drug is safe and effective by showing that it contains the same active ingredient as an approved new drug, and also that the effect of the generic drug on the patient is equivalent to that of the approved drug. In effect, therefore, FDA relies in part on the body of knowledge developed for the pioneer drug product in evaluating the ANDA. If the manufacturer of the generic drug shows that there are no differences between the approved drug and the "me-too" drug that will result in differences in safety or effectiveness, the generic drug is deemed to have the requisite evidence of safety and effectiveness. If the other requirements are also satisfied, the ANDA is approved. /50/ 2. If generic drugs were not subject to premarketing approval, however, a physician or pharmacist would have no assurance that generic products were as safe and effective as the brand name drugs for which they purport to be substitutes. Therefore, as the government's experts testified in this case (Tr. 55-57), careful doctors and pharmacists, if they became aware that generic drugs were not subject to premarketing review, would be deterred from prescribing generic drugs regardless of the cost savings that might be obtained. /51/ Since current law prohibits the labeling of any drug with a notice that it has (or has not) been approved by FDA, 21 U.S.C. 331(l), the exemption advocated by respondents and adopted by the court below would have a negative impact on sales of generic drugs -- including those that have been shown to be safe and effective and have been approved by FDA. Unless the decision below is reversed, the only way to be assured of using a safe and effective product would be for doctors to specify and pharmacists to dispense only the pioneer. Such a result would be contrary to the very congressional policy of increased competition that the court of appeals thought supported its decision. Moreover, it would hamper efforts by the FDA, other federal government agencies, and state governments to foster the use of safe and effective substitutes for more costly prescription drugs. /52/ CONCLUSION The decision of the court of appeals should be reversed. Respectfully submitted. REX E. LEE Solicitor General WILLIAM F. BAXTER Assistant Attorney General LOUIS F. CLAIBORNE Deputy Solicitor General JERROLD J. GANZFRIED Assistant to the Solicitor General JOHN J. POWERS, III NANCY C. GARRISON Attorneys JEFFREY B. SPRINGER Deputy Chief Counsel DONALD O. BEERS Associate Chief Counsel for Enforcement JACQUELINE H. EAGLE Assistant Chief Counsel for Enforcement Food and Drug Administration Rockville, Md. 20857 APRIL 1982 /*/ In addition to respondent Generix Drug Corporation, the defendants in the district court were Lewis Michael Orlove, Gary R. Dubin, and Ofelia Perez, all of whom were officers or employees of Generix. /1/ This case was decided by Unit B of the Fifth Circuit which, on October 1, 1981, became the Eleventh Circuit. /2/ "Pet. App." refers to the Appendix to the Petition for Certiorari filed in this case. "J.A." refers to the joint appendix filed in this Court. "Tr." refers to the transcript of the hearing in the district court on the government's motion for preliminary injunction. "R." refers to the district court record as transmitted to the court of appeals. /3/ The authority to approve new drugs, granted by statute to the Secretary of Health and Human Services, has been delegated to the Commissioner of Food and Drugs. 21 C.F.R. 5.1(a)(1); see 46 Fed. Reg. 26052 (1981). The introduction or delivery for introduction into interstate commerce of a new drug for which no approved new drug application is in effect is prohibited under 21 U.S.C. 331(d). Violations of 21 U.S.C. 331 may be enjoined under 21 U.S.C. 332. Persons who violate Section 331 are also subject to criminal penalties (21 U.S.C 333), and drugs that fail to comply with 21 U.S.C. 355 may be seized and condemned (21 U.S.C. 334). In any of these proceedings, the actual safety or effectiveness of the unapproved drug is not an issue for the court and will not provide a defense. See Weinberger v. Hynson, Westcott & Dunning, Inc., 412 U.S. 609, 612-613 (1973); Premo Pharmaceutical Laboratories, Inc. v. United States, 629 F.2d 795, 803 (2d Cir. 1980). /4/ A drug can be "generally recognized" within the meaning of the Act only when there is an expert consensus founded upon the type of evidence necessary for approval of a new drug application (see page 22, infra). Weinberger v. Hynson, Westcott & Dunning, Inc., supra, 412 U.S. at 632. /5/ A drug also may be excluded from the new drug approval requirements if it comes within one of the grandfather provisions of 21 U.S.C. 321(p), not relevant here. /6/ The unapproved new drugs distributed by respondents included "Goldline" products containing the following active ingredients: allopurinol, spironolactone, spironolactone with hydrochlorothiazide, furosemide, chlorothiazide with reserpine, amitriptyline with perphenazine, diethylpropion hydrochloride, prochlorperazine maleate, and chlorthalidone (R. 28 n.3). The request for injunctive relief to prevent distribution of unapproved new drugs, however, was not limited to these particular drugs (ibid.). The complaint also alleged (J.A. 8-11, 15-17) that respondents were violating the Act by holding for sale and distributing in interstate commerce drugs that were adulterated within the meaning of 21 U.S.C. 351(a)(2)(B) because of failure to conform to current good manufacturing practices as set forth in 21 C.F.R. Part 211. Those allegations were not directly related to the "new drug" allegations and were resolved without trial. Thus there is no issue before this Court concerning the good manufacturing practices allegations. In separate proceedings, commenced on August 15 and 19, 1980, the United States brought seizure actions pursuant to 21 U.S.C. 334, to condemn more than five million dosage units of unapproved generic prescription drugs in Generix's possession. Although Generix moved (without opposition) to consolidate the seizure actions with the instant injunction action, the district court did not consolidate the cases. Following the decision by the court of appeals in the injunction action, the district court granted Generix's motion for summary judgment in the seizure actions and ordered the United States to return the seized drugs. On January 18, 1982, this Court stayed the district court's order pending appeal. United States v. Undetermined Quantities of Articles of Drugs, No. A-617. /7/ R. 79-81; see also R. 28 n.3. /8/ R. 106-149, 197-207, 212-224; Tr. 26-27, 250-253. /9/ The quantity and delivery speed of active ingredients to the bloodstream and thus to the site of action in the body is referred to as "bioavailability" (Tr. 39-42; see also 21 C.F.R. 320.1(a)). Drugs with similar bioavailability are considered "bioequivalent." Because differences in bioavailability can be caused by differences in either active or inactive ingredients, or by differences in methods of manufacture, drugs containing the same active ingredient may not be bioequivalent. One of the government's expert witnesses testified as follows about common causes of lack of bioequivalence (Tr. 83-84): The primary reason for this is that the solubility of the drug is altered, either (sic) because of the chemical nature of the drug, itself, which may be modified. For example, different kinds of crystalline forms dissolve in different ways. You can also alter the way that the drug dissolves by the excipients. For example, some excipients were added to the tablet to help the tableting process, a lubricant to help the tablet come out easier, are insoluble and can coat the product, itself, act as a raincoat, so the water can't get to the particles and doesn't dissolve. All drugs must dissolve in the gastrointestinal tract before they are absorbed through the membrane. So anything that alters the rate of solubility, that is the dissolution rate, may influence whether the drug is absorbed. Probably 90, 95 percent of the cases of inequivalence is because of the excipients or the way the drug is manufactured, altering the solubility. The way that the drug is manufactured also includes things such as how hard you press down on the tablet to form the tablet. Its been shown in scientific studies that simply the force of the tableting may alter the dissolution characteristics of that particle or that tablet. The tablet must first break apart in small particles. That's altered by both the ingredients and the processes of the manufacturer. Then, these small particles must break down into further tiny particles which then go into solution. And all of these manufacturing variables influence this sequence of steps. /10/ See also United States v. Premo Pharmaceutical Laboratories, Inc., 511 F. Supp. 958, 962-965, 980-989 (D.N.J. 1981), in which the district court made detailed factual findings on the dangers created by differences between an approved prescription drug and generic copies of that drug. /11/ For example, if the products used to treat high blood pressure do not perform properly, the patient is in serious danger either from recurrence of the high blood pressure, if absorption of the drug is inadequate, or from toxicity if the drug is overabsorbed (Tr. 47-49). /12/ See United States v. Premo Pharmaceutical Laboratories, Inc., supra, 511 F. Supp. at 963; Pharmadyne Laboratories, Inc. v. Kennedy, 466 F. Supp. 100, 106 (D.N.J.), aff'd on other grounds, 596 F.2d 568 (3d Cir. 1979). /13/ In so doing, it relied primarily on the district court decision in Premo Pharmaceutical Laboratories, Inc. v. United States, 475 F. Supp. 52 (S.D.N.Y. 1979); that decision was subsequently reversed by the Second Circuit (629 F.2d 795 (1980)). /14/ Allopurinol, spironolactone, furosemide, chlorothiazide with reserpine, amitriptyline, and diethylpropion hydrochloride (Pet. App. 20a). /15/ Those drugs contained prochlorperazine maleate and chlorthalidone (Pet. App. 20a). /16/ The government's request for an injunction with respect to the good manufacturing practice violations was held in abeyance pursuant to agreement of the parties (Tr. 186-187). The district court also denied the government's request for a recall of Generix's unapproved products (Pet. App. 30a-31a). /17/ The court of appeals apparently assumed that FDA approval of a drug product means that the product's active ingredient is generally recognized as safe and effective. This is incorrect. See pages 24-25 infra. /18/ The court also thought (Pet. App. 10a-12a) that the requirement of Section 502(e) of the Act, 21 U.S.C. 352(e), that drugs be identified by generic name in addition to brand name would be inconsistent with the inclusion of the entire drug product, rather than just the active ingredient, in the definition of "drug." /19/ See Br. in Opp. 5 n.7. The absurdity of respondents' position is graphically illustrated by the affidavit of an expert witness submitted by Generix in opposition to the government's motion for preliminary injunction. J.A. 56-65. After noting that "all of these characteristics (inactive ingredients, manufacturing techniques, bioavailability and bioequivalence) can affect how a drug product performs and can therefore be said to play a role in the safety and effectiveness of the drug" (id. at 64), the affidavit concluded (ibid.): In other words, it would be possible to formulate allopurinol in a product in such a way that the allopurinol would never be released. Under those circumstances, the product would be completely ineffective, but the drug -- allopurinol -- would still be regarded as effective. /20/ The most important evidence of a statute's meaning is the statutory language itself. International Brotherhood of Teamsters v. Daniel, 439 U.S. 551, 558 (1979). "Absent a clearly expressed legislative intention to the contrary, that language must ordinarily be regarded as conclusive." Consumer Product Safety Commission v. GTE Sylvania, 447 U.S. 102, 108 (1980). /21/ Other courts have reached the same conclusion as did the Second Circuit in Premo. For example, in Pharmadyne Laboratories, Inc. v. Kennedy, 466 F. Supp. 100 (D.N.J.), aff'd on other grounds, 596 F.2d 568 (3d Cir. 1979), the court agreed with the FDA that the "new drug" term applies to specific drug products, rejecting dicta in an earlier Third Circuit decision, United States v. Articles of Drug . . . Lannett, 585 F.2d 575 (1978), which had suggested that some or all generic copies of FDA-approved drugs do not themselves have to undergo FDA premarketing clearance. (The Third Circuit in Pharmadyne did not address the generic drug issue, affirming the district court instead on the ground that a court is without power to enjoin the FDA from bringing enforcement actions; but it suggested that differences in inactive ingredients, at least where those differences raise safety issues, would prohibit a generic copy from relying on the not new drug status of the pioneer it copied. See 596 F.2d at 571 n.6.) In two other actions, the FDA has obtained injunctive relief against major manufacturers of unapproved prescription drugs. United States v. Pharmadyne Laboratories, Inc., No. 80-1312 (D.N.J., filed May 8, 1980) (consent decree of permanent injunction); United States v. Premo Pharmaceutical Laboratories, Inc., supra. /22/ For example, the most recent edition of the combined United States Pharmacopoeia-National Formulary has monographs not only for furosemide as an active ingredient but also for "furosemide injection" and "furosemide tablets." The Pharmacopoeia of the United States of America, Twentieth Revision/The National Formulary, Fifteenth Edition 344-345 (1980). A monograph may exist both for an "official substance," defined as "an active drug entity or a pharmaceutical ingredient * * * ," and for a "dosage form," defined as the "finished * * * preparation or product of one or more official substances formulated for use on or for the patient." Id. at 2. /23/ No inconsistency is created by the fact that active ingredients are also "drugs." Indeed, 21 U.S.C. 321(g)(1)(D) expressly includes within the "drug" definition "articles intended for use as a component of any article (referred to in the preceding) clauses." The term "component" refers to active as well as inactive ingredients of a drug (cf. 21 U.S.C. 355(b)(2)). Therefore, a reading of Section 321(g)(1) as a whole shows that the term "drug" as defined in subsections (A), (B), and (C) must refer to drug products and not just active ingredients. /24/ The definition of "drug" in the 1906 Federal Food and Drug Act -- the predecessor to the Federal Food, Drug, and Cosmetic Act of 1938 -- also included both drug substances and drug products. Thus, the 1906 definition of "drug" included, "all medicines and preparations recognized in the United States Pharmacopoeia or National Formulary for internal or external use, and any substance or mixture of substances intended to be used for the cure, mitigation, or prevention of disease of either man or other animals." Ch. 3915, Section 6, 34 Stat. 769 (emphasis added). This definition was changed to the language contained in Section 321(g) when Congress enacted the 1938 Act. The purpose of that change was to expand and not to restrict the definition of the term "drug." See S. Rep. No. 646, 74th Cong., 1st Sess. 1 (1935) ("This expansion of the definition of the term 'drug' and the inclusion of devices are essential if the consumer is to be protected against a multiplicity of abuses not subject to the present law"). /25/ 21 U.S.C. 355(b) requires that an NDA contain (emphasis added): (1) full reports of investigations which have been made to show whether or not such drug is safe for use and whether such drug is effective in use; (2) a full list of the articles used as components of such drug; (3) a full statement of the composition of such drug; (4) a full description of the methods used in, and the facilities and controls used for, the manufacture, processing, and packing of such drug; (5) such samples of such drug and of the articles used as components thereof as the Secretary may require; and (6) specimens of the labeling proposed to be used for such drug. /26/ 21 U.S.C 355(d) defines "substantial evidence" as evidence consisting of adequate and well-controlled investigations, including clinical investigations, by experts qualified by scientific training and experience to evaluate the effectiveness of the drug involved, on the basis of which it could fairly and responsibly be concluded by such experts that the drug will have the effect it purports or is represented to have under the conditions of use prescribed, recommended, or suggested in the labeling or proposed labeling thereof. /27/ If the manufacturer wishes to make any change in the composition or labeling of the approved drug, it must first submit a supplement or amendment to the NDA. See 21 C.F.R. 314.8. If another manufacturer is licensed to produce the same drug, it too must obtain an NDA. If evidence subsequently shows that an approved drug is unsafe or ineffective or that the applicant has failed to maintain required records or to correct misleading labeling, the FDA, after opportunity for hearing, may revoke approval of the NDA. 21 U.S.C. 355(e). /28/ Even if a drug satisfies the general recognition criteria, it remains a "new drug" -- and requires FDA approval -- unless it has been used to a material extent and for a material time under other than experimental conditions. 21 U.S.C. 321(p)(2). /29/ This conclusion is fully consistent with this Court's holding in Weinberger v. Hynson, Westcott & Dunning, Inc., supra, that where there was no substantial evidence showing the effectiveness of a pioneer drug or of a generic copy of that drug, the FDA could conclude that both were new drugs. For if there are no adequate effectiveness tests of the pioneer drug, and the manufacturer of the generic product does not produce adequate effectiveness studies in response to a notice of opportunity for hearing, there can be no claim that the generic product has been shown by adequate testing to be effective. On the other hand, because a generic product may differ from the pioneer in bioavailability, the fact that there has been adequate testing to show the pioneer is safe and effective does not, in itself, demonstrate the generic product is either safe or effective. See Pharmadyne Laboratories, Inc. v. Kennedy, supra, 466 F. Supp. at 104 n.7; see also United States v. Premo Pharmaceutical Laboratories, Inc., supra, 511 F. Supp. at 968 n.6a. Generix is incorrect in its assertion (Br. in Opp. 22 n.17) that if the "new drug" definition applied to products, the Court's ruling in Hynson would have been unnecessary. At issue in Hynson was FDA's procedure of, in effect, declaring classes of drug products, i.e., all generic copies related to a specific pioneer, to be "new drugs," thus avoiding case-by-case litigation of the status of each product. Cf. Weinberger v. Bentex Pharmaceuticals, Inc., supra, 412 U.S. at 653. /30/ Generix's witness did not rebut government testimony to that effect (see Tr. 49, 51-52, 92-93, 96-97). Indeed, while Generix's expert witness agreed that generic drugs should not be marketed without testing (Tr. 224-225), he admitted that he had seen a test of only one of the products at issue (Tr. 219), and no evidence of any testing of Generix's drugs was introduced below. The unapproved drugs Generix was distributing included products manufactured by Premo Pharmaceutical Laboratories, Inc., and Pharmadyne Laboratories, Inc. (see United States v. Undetermined Quantities of Articles of Drugs, stay granted by this Court, No. A-617 (Jan. 18, 1982)), companies that subsequently have been enjoined from the manufacture of such products, see page 15 note 21, supra. The court that granted injunctive relief against Premo found specifically that that company placed each of its products on the market without completing testing that might determine whether or not the products were bioequivalent to those drugs for which they would be substituted. United States v. Premo Pharmaceutical Laboratories, Inc., supra, 511 F. Supp. at 990-991. /31/ This Court's understanding that a drug would not cease to be a new drug until long after it had satisfied the requirements for FDA approval is substantiated by the statutory structure itself: 21 U.S.C. 321(p)(2) states that general recognition based on investigations to determine safety and effectiveness (the basis for approval of a new drug application) is not sufficient, absent use to a material extent and for a material time, to exempt a drug from "new drug" status. It is further supported by the legislative history of the 1962 amendments. Thus, a witness before the Senate subcommittee quoted an FDA official as stating that, under the pre-1962 new drug provision: "Although there may be an effective new drug application for a product under which a drug firm is legally marketing it in interstate commerce, until there has been a considerable volume of distribution over a considerable period of time, it is still a new drug(;) * * * there can be no general recognition of safety by experts without appreciable use of the drug." Drug Industry Antitrust Act: Hearings on S. 1552 Before the Subcomm. on Antitrust and Monopoly of the Senate Comm. on the Judiciary, 87th Cong., 1st Sess. 63 (1961). This standard was confirmed when the 1962 amendments were passed. The Committee report suggests that "substantial evidence" of effectiveness, a prerequisite for approval of a new drug application, could exist even when experts generally did not know about the effectiveness studies or would disagree as to whether the drug was effective. See S. Rep. No. 1744 (Pt. 1), 87th Cong., 2d Sess. 16 (1962). The court of appeals' assumption that approval equals non-"new drug" status, and thus exemption from the approval requirement, leads to irrational results inconsistent with the statutory scheme. For example, if a pioneer drug were approved on one day, the generic copy then could escape the approval requirement the next day. The pioneer itself would, on the date of approval, cease to be a "new drug" and the withdrawal of approval provision of 21 U.S.C. 355(e) would be without effect. /32/ A lack of knowledge about inactive ingredients and manufacturing processes does, however, prevent expert general recognition of a generic copy based on testing of the pioneer product, because experts would not have adequate information to conclude the copy would not differ in bioavailability -- and therefore in safety or effectiveness -- from the tested pioneer drug. /33/ See United States v. An Article of Drug . . . Bacto-Unidisk, supra, 394 U.S. at 798; AMP, Inc. v. Gardner, 389 F.2d 825, 829-830 (2d Cir.), cert. denied, 393 U.S. 825 (1968); S. Doc. No. 124, 75th Cong., 2d Sess. (1937); C. Dunn, Federal Food, Drug, and Cosmetic Act, a Statement of Its Legislative Record 1316-1327 (1938). /34/ Generix argues (Br. in Opp. 11) that "(a) lethal chemical is not an inactive ingredient," apparently suggesting that the "new drug" provision as interpreted by the court of appeals would prevent a repetition of the Elixir Sulfanilamide tragedy. This is incorrect. The court of appeals held in this case that there can be no premarketing inquiry whatsoever into properties of a drug product if the identified active ingredient of the product is not a "new drug." In the case of Elixir Sulfanilamide, the manufacturer did not identify diethylene glycol, the lethal component, as an active ingredient in its product. In fact, of the ingredients in the product, only the active ingredient sulfanilamide was listed on its labeling. S. Doc. No. 124, supra, at 2. It follows that the manufacturer of Elixir Sulfanilamide, under the court of appeals' interpretation of the Act, could have marketed its product without preclearance merely because some other sulfanilamide product had been generally recognized as safe and effective. Contrary to Generix's assertion (Br. in Opp. 11) that the legislative history of the 1938 Act never refers to diethylene glycol, the toxic component of Elixir Sulfanilamide, as an inactive ingredient, that history makes clear the congressional understanding that diethylene glycol was in fact an inactive ingredient. See S. Doc. No. 124, supra, at 1 (Elixir Sulfanilamide: Letter from the Secretary of Agriculture) (emphasis added): The lethal effect of the "elixir" was due to its content of diethylene glycol, which was used as a solvent in making a liquid preparation of sulfanilamide, usually administered in tablet or powder form. Sulfanilamide itself is a valuable drug, and was not responsible for the disaster. See generally ibid., passim; C. Dunn, Federal Food, Drug, and Cosmetic Act, A Statement of Its Legislative Record, supra, at 1316-1327. /35/ See, e.g., S. Rep. No. 1744 (Pt. 2) 87th Cong., 2d Sess. 1 (1962); 108 Cong. Rec. 14677-14680, 16316-16317, 16322-16325, 19897-19898, 19909, 22037-22038, 22047-22048 (1962). /36/ See H.R. Rep. No. 2464, 87th Cong., 2d Sess. 1 (1962); S. Rep. No. 1744 (Pt. 1), 87th Cong., 2d Sess. 8 (1962). /37/ 21 U.S.C. 352(e) requires that the common names of the active ingredients and of certain inactive ingredients of a drug be stated on the label of the drug. This information, of course, would not be sufficient to allow a doctor or pharmacist to determine whether a generic drug would have the same effect on the patient as the pioneer. See pages 19-22, supra. Contrary to the court of appeals' assumption that drugs "obtain an 'official' or 'generic' name through general recognition" (Pet. App. 12a), the concept of "genreal recognition" relates only to the definition of "new drug." It does not appear in the labeling provisions of 21 U.S.C. 352(e), and has nothing to do with the assignment of generic names. /38/ Inclusion of the established name for a drug's active ingredient in its labeling assists doctors and pharmacists in identifying generic equivalents and "ascertain(ing) the qualities and specifications of the products and the competing products." S. Rep. No. 1744 (Pt. 1), supra, at 18. Labeling a drug with the generic name as well as the trade name aids physicians in deciding "whether to prescribe by the trade name upon which they have come to rely or to prescribe the so-called generic equivalent by using the official name or by authorizing the pharmacist to select a product bearing the official name" (ibid.). /39/ Moreover, much of the discussion in the legislative history of the need to improve competition related to proposals -- never enacted -- that would have limited the terms of drug patents. Proponents of those proposals recognized that, as reported, the 1962 amendment "will have only a very limited effect on prices * * * . Its major result will be to improve the quality of drugs." S. Rep. No. 1744 (Pt. 1), supra, at 33. /40/ See, e.g., 108 Cong. Rec. 17366 (1962) (emphasis added): (E)very brand new product, and every new claim for an existing product, would be subject to the tests and procedures established in (21 U.S.C. 355). Contrary to the court of appeals' assertion (Pet. App. 9a), this reference is fully consistent with -- rather than contrary to -- the plain language and logic of the Act. /41/ The lower courts have also applied this basic principle in construing the Act. See, e.g., United States v. Naremco, Inc., 553 F.2d 1138, 1141 (8th Cir. 1977); Upjohn Co. v. Finch, 422 F.2d 944, 947 (6th Cir. 1970); Carter-Wallace, Inc. v. Gardner, 417 F.2d 1086, 1094-1095 (4th Cir. 1969), cert. denied, 398 U.S. 938 (1970); United States v. Vitasafe Corp., 345 F.2d 864, 870 (3rd Cir.), cert. denied, 382 U.S. 918 (1965). /42/ There is no doubt, however, that some unapproved generic drugs are dangerous. The Premo court found that some of the drugs Premo made (which Generix distributed), if not bioequivalent to the pioneers they copied, could cause death or irreversible disfigurement in patients treated with those drugs. 511 F. Supp. at 992-993, 994-995. See also Tr. 47-48. /43/ This issue was explicitly explored in United States v. Pre o Pharmaceutical Laboratories, Inc., supra. The court there found that compliance with FDA's good manufacturing practice regulation would not protect the public from drugs that were not bioequivalent. 511 F. Supp. at 989; id. at 976-977. /44/ Nor, contrary to the court of appeals' apparent assumption (Pet. App. 13a), is the requirement of 21 U.S.C. 351(b) an adequate substitute for premarketing approval. That provision requires that a drug that purports to conform to standards set out in an official compendium in fact meet that compendium's standards. Those standards are not designed to assure bioequivalence. Accordingly, compliance with the compendial standards would not assure that a generic copy could be safely substituted for a pioneer. See Tr. 98-100; see also United States v. Premo Pharmaceutical Laboratories, Inc., supra, 511 F. Supp. at 987, 988-989, 977. /45/ A drug is adulterated under 21 U.S.C. 351(a)(2)(B) if the "methods used in, or the facilities or controls used for, its manufacture, processing, packing, or holding do not conform to or are not operated or administered in conformity with current good manufacturing practice to assure that such drug meets the requirements of this chapter as to safety and has the identity and strength, and meets the quality and purity characteristics, which it purports or is represented to possess." See 211 C.F.R. Parts 210-211. /46/ As reported by the subcommittee, the bill that became the 1962 amendments "would have set up a registration system (for drug manufacturers) which * * * would have in effect constituted a licensing system." S. Rep. No. 1744 (Pt. 1), supra, at 12; see S. 1552, 87th Cong., 1st Sess., as introduced by Senator Kefauver on April 12, 1961. The full committee deleted the licensing provision. The congressional rejection of a "licensing" system on which respondents (Br. in Opp. 15) rely, was a rejection of this manufacturer licensing provision; it does not indicate any congressional intent to exempt generic drugs from a product-by-product application of the "new drug" approval requirements. The court of appeals referred to a statement in the legislative history of the 1938 Act that the new drug provision "is not a license provision" (Pet. App. 9a, quoting H.R. Rep. No. 2139 (Pt. 1), 75th Cong., 3d Sess. 9 (1938)). This statement, which is not otherwise explained, may refer also to the rejection of a manufacturer licensing provision that some members of Congress had sought in 1938. See 83 Cong. Rec. app. 2279 (1938) (remarks of Rep. Coffee). /47/ Indeed, as the court of appeals noted (Pet. App. 7a), the formulation of the pioneer drug usually is a trade secret and may not be revealed by FDA. See 21 U.S.C. 331(j). /48/ Generix argues (Br. in Opp. 18-22), and the court of appeals apparently agreed (Pet. App. 8a n.4), that FDA's interpretation of the Act's new drug provisions is not entitled to the deference usually afforded its interpretation of the Act (e.g., United States v. Rutherford, supra) because it has not been consistent. This argument is essentially irrelevant since the plain language of the Act and its legislative history compel the conclusion that the Act's "new drug" requirements apply to generic drug products. At all events, Generix and the court of appeals have seriously distorted the FDA's actions. Contemporaneously with the passage of the 1938 Act, the Secretary of Agriculture (who then had responsibility for enforcement of the Act) promulgated a regulation that made it clear that a drug could be considered a "new drug" if there were any change in, inter alia, the inactive ingredients from which it was formulated (see 3 Fed. Reg. 3161-3162 (1938)). That regulation stating the agency's official interpretation of the "new drug" definition (now 21 C.F.R. 310.3(h)) has remained substantially unchanged to the present day. The FDA's practice, initiated during World War II and terminated in 1968 (33 Fed. Reg. 7758 (1968)), of informing drug manufacturers that they did not need to submit new drug applications for certain drugs because those drugs were not "new drugs" (Pet. App. 8a n.4) was not a statement that generic copies of approved drugs automatically would escape "new drug" status. Indeed, some generic copies of approved drugs were seized and condemned as unapproved new drugs during the time this practice was in effect. See 40 Fed. Reg. 26142, 26143 (1975). Arguably, the so-called "not new drug" letters reflected FDA's understanding that experts at that time would recognize the safety of generic products on the basis of less evidence than experts today would require in light of developing knowledge about the importance of inactive ingredients' effect on bioavailability. Cf. United States v. Western Serum Co., Inc., 666 F.2d 335, 339-341 (9th Cir. 1982) (expert general recognition must be current). Moreover, to the extent "not new drug" letters indicated that generic drug products could be marketed without FDA approval, those letters may be viewed as inconsistent with the Act and with this Court's decision in Weinberger v. Hynson, Westcott & Dunning, Inc., supra, 412 U.S. at 624. See United States v. Articles of Drug . . . Hormonin, 498 F. Supp. 424, 436 (D.N.J. 1980), aff'd without opinion, No. 81-1035 (3d Cir. Sept. 15, 1981). Even if the FDA's past practice is considered to conflict with its present position, therefore, that past practice binds neither it nor this Court. FDA's current abbreviated new drug application procedures (see pages 40-41, infra) simplify the requisite proof of safety and effectiveness for generic drugs, but they do not permit deviation from the statutory requirement of premarketing approval for each drug product, based on evidence that that drug product itself is safe and effective. /49/ The ANDA process currently applies only to copies of drug products for which the FDA has published a notice known as a Drug Efficacy Study Implementation ("DESI") Notice. Most, but not all, of the unapproved drugs Generix was distributing purported to be generic copies of drugs subject to such notices and the manufacturers therefore could have sought approval under this simplified procedure. In addition, the FDA is implementing a so-called "paper NDA" policy for post-1962 generic drugs, 45 Fed. Reg. 82052 (1980); 46 Fed. Reg. 27396 (1981). Under that policy, applicants may rely on the scientific literature, if it provides adequate data, to establish safety and effectiveness of a pioneer drug. Safety and effectiveness of a generic drug is then established by such literature plus the applicant's showing that its generic drug is bioequivalent to the pioneer. 21 C.F.R. 320.22. In an attempt to facilitate further the approval of safe and effective generic drug products, FDA also is exploring the posibility of authorizing use of ANDAs for copies of drugs approved after 1962 and thus not covered by DESI notices. See 47 Fed. Reg. 1765 (1982). /50/ Contrary to Generix's assertion (Br. in Opp. 3 n.5), FDA's procedures for regulating over-the-counter (OTC) drugs are not at issue in this case. The government sought an injunction because of its concern about Generix's distribution of unapproved prescription drugs; the evidence presented to the district court was limited to prescription drugs; and neither the district court nor the court of appeals addressed the separate issue of OTC regulation. Nor would a holding that the Act's new drug approval procedures apply to generic drug products invalidate FDA's current OTC drug regulations (21 C.F.R. Part 330). The OTC regulations themselves deal with drug products, rather than only active ingredients, although the analysis of OTC products is done on the basis of classes of active ingredients. Thus, for example, "(a)n over-the counter antacid product" (emphasis added) that complies with the antacid monograph is regarded by the agency as generally recognized as safe and effective and not misbranded. 21 C.F.R. 331.1. There is no agency finding that any active ingredient itself is generally recognized as safe and effective or is not a new drug. Moreover, to the extent that there are differences between FDA's treatment of prescription and OTC drugs, they are warranted by practical considerations. Lack of bioequivalence between different OTC products containing the same active ingredients presents less danger than a similar lack in prescription drugs because OTC drugs, in contrast to prescription drugs (see 21 U.S.C. 353(b)(1)), are intended for the treatment of minor, and in many cases self-limiting, conditions. Thus the monographs, which limit the amount of OTC drugs' active ingredients to levels that pose no risk of toxicity and impose standards with respect to inactive ingredients, are a reasonable means of coping with the necessity of regulating between 100,000 and 500,000 OTC products. See 37 Fed. Reg. 85 (1972). See also Weinberger v. Bentex Pharmaceuticals, Inc., supra, 412 U.S. at 650-651. In addition, FDA's treatment of OTC drugs is consistent with the statute's definition of "new drug": experts generally require more information about prescription drug products before determining that they are safe and effective than those experts would require of an OTC preparation. /51/ See also Pharmadyne Laboratories, Inc. v. Kennedy, supra, 466 F. Supp. at 106. /52/ Generally, drug substitution programs, both in the states and in the federal system, allow a physician to veto substitution. For example, under the federal "maximum allowable cost" regulations implementing the Medicare and Medicaid reimbursement statutes, a physician may, by so stating on his prescription, veto substitution, 45 C.F.R. 19.3(a)(3). Physicians may be expected to veto substitution where they cannot be assured that all generic copies of the products they are prescribing have been approved by FDA. For a discussion of state substitution programs and the effect of the generic "new drug" issue on those programs, see Pharmadyne Laboratories, Inc. v. Kennedy, supra, 466 F. Supp. at 106. Cf. Note, Consumer Protection and Prescription Drugs: The Generic Drug Substitution Laws, 67 Ky. L.J. 384 (1978-1979).