United States v. Generix Drug Corporation, a Corporation, Lewis Michael Orlove, Gary R. Dubin, and Ofelia Perez, Individual

JAMES C. HILL, Circuit Judge: Before a “new drug” may be introduced into • interstate commerce it must be approved by the Food and Drug Administration’s “new drug application” (NDA) process. 21 U.S.C. § 355(a). Only drugs that are “new drugs” under 21 U.S.C. § 321(p) require an NDA as a condition of marketing. The NDA process is often expensive and time consuming. In the district court, *1115 498 F.Supp. 288, the plaintiff-appellee, the United States, alleged that Generix was distributing “new drugs” without approved new drug applications. The United States sought to enjoin this distribution pursuant to 21 U.S.C. § 332. These appeals arise from an order granting in part and denying in part the United States’ motion for a preliminary injunction and denying Generix’s motion for rehearing and motion to amend order on preliminary injunction. 28 U.S.C. § 1292(a)(1). The issue in these appeals is whether the definition of “new drug” in the Federal Food, Drug and Cosmetic Act, 21 U.S.C. § 321(p) (West 1972) applies to the entire drug product or whether it applies only to a drug’s active ingredients. We hold that the language and intent of Congress limits the definition to a drug’s active ingredients. Accordingly, we vacate the preliminary injunction issued by the district court. 1 I. Definitions are important in this case. Appellant Generix distributes generic drugs. A generic drug is a copy of the active ingredient of another manufacturer’s drug product which as has been shown to be safe and effective. 2 A drug product is composed of an active ingredient, or incipient, which represents up to 10% of the product, and excipients, such as binders, coating, and capsules, which account for the remainder. Because manufacturing techniques vary, generic drug products often combine an active ingredient that is generally recognized as safe and effective and excipients that are unique to the individual manufacturer. The active ingredients of Generix’s drug products were not generally in issue in the district court proceedings. Rather, the government’s experts testified that the variances in excipients due to different manufacturing processes can directly affect the bioavailability of the active ingredient. Bioavailability is a measure of the time it takes for a given drug product to deliver the active ingredient to the particular organ or area. Additionally, the differences in excipients can affect bioequivalence, a measure based on the comparison of one drug to another. Drug products which are bioequivalent can be used interchangeably for the treatment of the same illness. Both government experts stressed that the differences in bioavailability and bioequivalence between the product originally prepared using the proved safe and effective “pioneer drug” and the product later produced using the same drug with, perhaps, different binders, coatings, etc. may effect the safety and efficacy of the generic drug product. This is especially true where the drug product involved is a sustained release drug. Differing rates of solubility due to differences in excipients may cause “dumping” of the active ingredient into the bloodstream all at once. In drugs where there is high toxicity, this may lead to an overdose. In essence, the government argues that any difference between excipients renders a drug product a “new drug” within the meaning intended by the statute. According to the government, whether the active ingredient is generally recognized as safe and effective is irrelevant to the discussion. Therefore, the government concludes, Generix’s products are “new drugs” due to differing excipients. Generix argues that if the active ingredient is generally recognized as safe and effective, the new drug product *1116 is not a “new drug” despite differences in excipients. After reviewing the statute and the case law the district court settled on the following legal standard: . . . [t]he manufacturer of the questioned product is entitled to a declaration that its product is not a “new drug” within the meaning of 21 U.S.C. § 321(p), only if the evidence has shown no reasonable possibility that the differences between the excipients in the recognized and questioned products will make the questioned products less safe and effective than the recognized product. District Court Opinion at 292, quoting Premo Pharmaceutical Laboratories, Inc. v. United States, 475 F.Supp. 52, 55 (S.D.N.Y. 1979), reversed 629 F.2d 795 (2d Cir. 1980). The district court then concluded that the government’s experts had raised a reasonable possibility that the safety and effectiveness of certain Generix drug products were suspect because of their excipients. The court further concluded that Generix had not refuted the government’s proof. Accordingly, the district court enjoined Generix from distributing their products which contained allopurinal, spironolactone, furosemide, chlorothiazide with reserpine, amitriptyline, and diethylpropion hydrochloride. District Court Order at 294. The district court refused to enjoin the distribution of Generix products containing prochloreperazine meleate and chlorthalidone. Id. II. In determining whether or not the term “new drug” refers only to the active ingredient of a drug product, as distinguished from excipients, we turn first to the relevant statutes. Title 21 U.S.C. § 321(g)(1) defines a “drug” as follows: (g)(1) The term ‘drug’ means (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 of 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. 21 U.S.C. § 321(g)(1). The record shows that the drugs listed in the sources described in subsection (A) of § 321(g)(1) are “drug substances.” not drug products. More importantly, Title 21 U.S.C. § 321(p) defines “new drug” as follows: (1) Any drug . . . the composition of which is such that such drug 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 . . .; or (2) Any drug . . . the composition of which is such that such drug, as a result of investigations to determine its safety and effectiveness for use under such conditions, has become so recognized, but which has not, otherwise than in such investigations, been used to a material extent or for a material time under such conditions. 21 U.S.C. § 321(p). Under subsection (2), after a period of time has elapsed and marketing experience has been acquired, a “new drug” substance will no longer be considered a “new drug” requiring an approved NDA to be marketed. As the Supreme Court has noted: Under subsection (2) [21 U.S.C. §’321(p)(2), a drug cannot transcend “new drug” status until it has been used “to a material extent or for a material time.” Yet, a drug cannot be marketed lawfully before an NDA has been approved by the Commissioner on the basis of “substantial evidence.” As the Solicitor General argues, “the Act is designed so that drugs on the market, unless exempt, will have *1117 mustered the requisite 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., 412 U.S. 609, 631, 93 S.Ct. 2469, 2484, 37 L.Ed.2d 207 (1973). If “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.’ ” This is so because the factors that bear on the safety and effectiveness of a specific product— manufacturing techniques, quality control, and the precise manner of formulating the finished dosage — are not part of the information made known to the medical community. Indeed, as we are informed by counsel, a significant part of such information is a closely guarded trade secret that may not be revealed by the F.D.A. See 21 U.S.C. § 331(j). Specifically, FDA regulations prohibit public disclosure of certain information in a new drug application, including manufacturing methods and processes, quality control procedures and quantitative and semiquantitative formulas. 21 C.F.R. § 314.14(g). In addition, only the identity of inactive ingredients previously disclosed may be released. See 21 C.F.R. § 321(p)(2). Accordingly, the “general recognition” test called for by § 321(p) must apply to a drug product’s active ingredient. Otherwise, each new drug product would be considered a “new drug” and § 321(p)(2) would be superfluous. 3 III. In effect, the FDA’s interpretation of the Act would create a product-by-product licensing system. 4 In addition to rendering the statutory language useless, such an interpretation is plainly contrary to the Con *1118 gressional intent. 5 Congress did not deal extensively with the role of generic drugs when the Food, Drug and Cosmetic Act was initially enacted in 1938. However, the House Report did state that “this [section 505, U.S.C. § 355] is not a license provision, but is intended merely to prevent the premature marketing of new drugs not properly tested for safety.” H.R.Rep. No. 2139, 75th Cong., 3d Sess. 9 (1938). The 1938 Legislation was amended in 1962 by the “Drug Industry Act of 1962.” See Senate Rep. No. 1744, 87th Cong., 2d Sess. 1962 U.S. Code Cong. & Ad. News 2884. Generic drugs played a central role in this legislation. The legislation originated as the result of an investigation by the Senate Subcommittee on Antitrust and Monopoly. See S.Rep. No. 1744, 87th Cong., 2d Sess., 1962 U.S. Code Cong. & Ad. News 2887. That Subcommittee was concerned with the large profit and tight market control of a few large companies in the drug industry. The Subcommittee found that “one of the principal factors underlying this market control was the success of major drug companies in persuading physicians to prescribe drugs by the companies’ trade names despite the existence of less expensive, chemically identical generic-named drugs.” Note, Drug Amendments of 1962-Generic Name Prescribing: Drug Price Panacea, 16 Stan.L. Rev. 649, 651-52 (1963). Accordingly, the 1962 amendments were designed, in part, to encourage generic-name prescribing and thereby lower prices. See Views of Senator Estes Kefauver et al., reprinted in 1962 U.S. Code Cong. & Ad. News 2898. The reasoning of Congress, which we set out in full, follows: Most drugs have what is often called a generic name; that is, a nonproprietary name recognized in the official compendia or by common usage. This legislation uses the term “official name,” and permits the Secretary to designate the official name in certain circumstances. Section 9(a) of the bill, as amended, would require that the label of every drug bear the official name with the conspicuousness required by section 502(c) of the Food, Drug, and Cosmetic Act; that is, so as to make it likely to be read and understood by the ordinary individual under customary conditions of purchase and use. Many drugs are marketed under brand names which serve to identify a particular product coming from a particular source. It provides the incentive for the manufacturer to maintain quality surpassing the minimum. At the same time, the brand name does not sufficiently identify the chemical structure of the drug. The use of the generic or official name is important so that practitioners and pharmacists can turn to the official compendia and other literature to ascertain the qualities and specifications of the product, and the competing products. The committee does not feel that the statute should specify the relative prominence of trade names and official names. It does feel, however, that the statute should specify that the official name be shown with sufficient conspicuousness so that practitioners and pharmacists can easily see it. Thus, competition would be fostered in two ways. The trade name would continue to be prominent, with its assurance of identity of source so that manufacturers could continue to have the incentives to strive for the widest possible distribution and sales. At the same time, to the extent that other manufacturers can market a competing product meeting the same specifications, the conspicuous showing of the official name will lend opportunities for such competition to flourish. The choice will then be left to physicians as to whether to prescribe by the trade name upon which they have come to rely or to prescribe the so-called *1119 generic equivalent by using the official name or by authorizing the pharmacist to select a product bearing the official name. 5. Rep. No. 1744, 87th Cong., 2d Sess.; 1962 U.S. Code Cong, and Ad. News 2893-2894. Excipients are unique to each individual drug manufacturer. Hence, if they are included within the definition of “new” drug no drug could ever obtain an “official” or “generic” name through general recognition as is contemplated by the above Congressional language. 6 IY. In the 1962 Amendments to the Act, Congress provided the means to regulate the excipients in a drug product containing an approved generic drug without subjecting the drug product to the new drug approval process. Specifically, the FDA may invoke the drug provisions of 21 U.S.C. § 351 to insure that a drug is not “adulterated.” The FDA may find that a drug is adulterated (1) if there have been inadequate controls in its manufacture, § 351(a); (2) if the strength, quality or purity differ from the official compendium, § 351(b); if the strength of a drug is misrepresented, § 351(c); or if a drug’s mixture with another substance reduces “its quality or strength.” § 351(d). Presumably the FDA could regulate bioavailability and bioequivalence through the provisions of § 351. As the Senate Report stated: Section 5 of the bill, as amended, is intended to give the Food and Drug Administration additional authority to require that sound methods, facilities, and controls be used in all phases of drug manufacture and distribution. By an amendment of section 501(a)(2) of the Food, Drug, and Cosmetic Act, a drug would be deemed to be “adulterated” if such methods, facilities or controls were found not to conform to current good manufacturing practice to assure safety, identity, strength, quality and purity of the product. This section, together with the registration and factory inspection provisions of sections 3 and 4 of this amended bill, would greatly increase the Food and Drug Administration’s authority over drug manufacture and distribution. These provisions should prove helpful in reaching persons who are responsible for putting drugs of questionable or unknown quality on the market. S.Rep. No. 1744, 87th Cong., 2d Sess.; 1962 U.S. Code Cong. & Ad. News 2884, 2890. V. The gist of the government’s argument is that if drug products are not subject to the NDA process their safety and effectiveness can not be properly regulated as required by 21 U.S.C. § 321(p). It may well be good public policy to regulate the bioavailability and bioequivalence through the NDA or *1120 some equivalent process. However, we have found that the statutory language and history of the Food, Drug and Cosmetic Act does not require such approval. The judicial branch may not legislate to fill perceived voids in public policy. Our Constitution leaves that job to Congress. See Wilson v. First Houston Investment Corp., 566 F.2d 1235, 1243-45 (5th Cir. 1978) (Hill, J., dissenting) vacated in light of Transamerica Mortgage Advisors, Inc. v. Lewis, 444 U.S. 11, 100 S.Ct. 242, 62 L.Ed.2d 146 (1979). For example, scientific studies since 1962 might provide a strong argument that ex-cipients should be included in the definition of “new drug.” Such a decision necessarily implicates complex chemical and pharmacological considerations. The judicial branch is not equipped properly to assess such considerations. In sum, we hold that Congress legislated that the term “new drug” as used in § 321(p) applies only to the active ingredient of a drug product. Accordingly, we vacate the preliminary injunction issued by the district court. VACATED and REMANDED with instructions to DISMISS. Footnotes [1] Our resolution of this case was greatly aided by the excellent briefs and arguments of the parties and the thorough opinion of the district court. [2] “Safe and effective” is a term of art. See 21 U.S.C. § 355(b). A “new drug” must be proved “safe and effective” by substantial evidence. The term ‘substantial evidence’ means 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 he 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. Section 505(d), 21 U.S.C. § 355(d). [3] In reaching our decision we have carefully considered the opinions of the Third Circuit, United States v. Articles of Drug (Lannett), 585 F.2d 575 (3d Cir. 1978) and the Second Circuit, Premo Pharmaceutical Laboratories, Inc. v. United States, 629 F.2d 795 (2d Cir. 1980.) Our primary reliance, however, has been on the language and intent of Congress. Although we reach a different result, we agree with the Second Circuit that the FDA “is usually better equipped by reason of its expertise to make the determination” of whether a drug product is safe and effective. 629 F.2d at 804. However, simply because the FDA is better suited to •assess the safety and effectiveness of drug products does not mean that Congress authorized the broad grant of jurisdiction that the government argues for in this case. [4] The government argues that the FDA’s interpretation of the Act’s new drug provisions is entitled to considerable deference since the FDA is the agency charged with the responsibility of administering the Act. United States v. Rutherford, 442 U.S. 544, 99 S.Ct. 2470, 61 L.Ed.2d 68 (1979). As a general proposition, of course, this is correct. However, this principle does not mandate that we ignore the plain language and history of the Act. Espinoza v. Farah Mfg. Co., 414 U.S. 86, 94-95, 94 S.Ct. 334, 339-340, 38 L.Ed.2d 287 (1973). Furthermore, as Generix points out, the FDA’s position on the definition of “new drug” has changed several times since 1938. For example, the FDA exempted drug products from the NDA process by issuing “not new drug” letters until 1968. 40 Fed.Reg. at 26, 143. Even when that policy was discontinued the FDA still maintained an interpretation of “new drug” which supports Generix’ position. [T]he agency fully realized that it would not be feasible from an administrative standpoint, or necessary for protection of the public health, or consistent with the definition of new drug in section 201(p) of the act to require either a full or an abbreviated NDA for all human prescription drugs. 40 Fed.Reg. at 26, 144 (emphasis added). Around 1968, the FDA commenced the practice of approving what is known as an “abbreviated NDA” or ANDA, to regulate drugs whose safety and effectiveness are well-established. The legal validity of this regulatory hybrid is questionable. See Hoffmann-LaRoche, Inc. v. Weinberger, 425 F.Supp. 890 (D.C.C. 1975). It is interesting to note the district court’s comment that FDA’s laxness in enforcing the NDA system stemmed in part from FDA’s fear “that it would have a difficult time in court contending that a specific version [of an established drug] is a new drug within the meaning of 21 U.S.C. § 321(p) . ... ” Id. at 892-93. In sum, the FDA’s shifting interpretation of “new drug” provides no basis for affording the FDA interpretation great deference. [5] We note that there are isolated references which support the government’s position. For example, the sponsors of the 1962 Drug Amendments stated: every brand new product, and every claim for an existing product, would be subject to the tests and procedures established in Section 505 of the act [21 U.S.C. 355] 108 Cong.Rec. 17366 (August 23, 1962) (emphasis added). However, isolated references in the legislative history can not overcome the plain language and override logic of the Act. [6] Generix urges that because the FDA has sought to establish the result they seek in this case through Congressional action and that Congress has declined to so act, we may infer that the government’s position in this case is invalid. We reject this argument and agree with the government that the failure of Congress to enact legislation that would clarify or modify an agency’s authority in a particular area does not limit the agency’s existing power under its current statute. As the Supreme Court has noted, “[pjublic policy requires that agencies feel free to ask [Congress for] legislation which will terminate or avoid adverse contentions and litigations.” Wong Yang Sung v. McGrath, 339 U.S. 33, 47, 70 S.Ct. 445, 453, 94 L.Ed. 616 (1950). The Court has also noted: [t]he advocacy of legislation by an administrative agency — and even the assertion of the need for it to accomplish a desired result is an unusual and unreliable, and not a highly desirable, guide to statutory construction. The possibility of its use to prove more than its means may, but should not, deter administrative agencies from seeking helpful clarification of authority or a fresh and specific Congressional mandate. American Trucking Ass’n v. Atchison T. & Santa Fe Ry. Co., 387 U.S. 397, 418, 87 S.Ct. 1608, 1619, 18 L.Ed.2d 847 (1967). Furthermore, we agree with the government that any attempt to rely on proposed legislation would be even more inappropriate in this case because of the number and variety of the bills involved. In the 96th Congress alone there were approximately nine bills (including H.R. 54, 2217, H.R. 4248, S. 772, S. 774, S. 780, S. 1045, S. 1075, S. 1138) that would have amended in different ways, the new drug provisions of the Act.