United States v. an Article of Drug Neo-Terramycin, 540 F. Supp. 363 (N.D. Tex. 1982)

*364 *365 *366 *367 *368 Richard E. Geyer, Assoc. Chief Counsel for Veterinary Medicine, Rockville, Md., Paula Mastropieri-Billingsley, Asst. U. S. Atty., Dallas, Tex., for plaintiff.

Robert Becker, Kleinfeld, Kaplan & Becker, Washington, D. C., Lee Simpson, Akin, Gump, Strauss, Hauer & Feld, Dallas, Tex., for Pfizer, Inc.

ROBERT M. HILL, District Judge.

Came on for consideration before the Court the motion for disposition under 21 CFR § 558.15 filed by Pfizer, Inc. (Pfizer), claimant, and the motion for summary judgment filed by the United States, plaintiff. Having considered the motions, briefs and oral argument of the parties, as well as the evidence submitted in support of the motions, the Court is of the opinion that the motion for disposition under 21 CFR § 558.15 should be denied and that the motion for summary judgment should be granted.

The procedural posture of this action is set out in the Court's Order of December 14, 1981, 529 F. Supp. 230, denying Pfizer's motion to dismiss. To summarize, the United States initiated this action in rem under the Federal Food, Drug and Cosmetic Act (the Act), 21 U.S.C. § 334 (1972), seeking seizure and condemnation of a specified quantity of the animal drug Neo-Terramycin Soluable Powder Concentrate (Neo-Terramycin). Neo-Terramycin is a combination drug composed of the antibiotics neomycin and oxytetracycline. Following a trial to a jury, the jury found that Neo-Terramycin "is generally recognized" as safe and effective, contrary to the allegations of the United States. Subsequently, the United States filed a motion for a new trial, which the Court granted on May 8, 1981.

Pfizer contends that the marketing of Neo-Terramycin is subject to the jurisdiction of the Food and Drug Administration *369 (FDA) under 21 CFR § 558.15 and that its status is governed by the interim marketing provision of that regulation. The government opposes this motion on the ground that Neo-Terramycin is not currently covered under § 558.15. The government also urges that this action can most expeditiously be disposed of on the merits and that returning the action to the FDA would simply result in the indefinite delay of the ultimate disposition of the action.

Section 558.15 was designed to encompass low-level (subtherapeutic), continuous (or long term) use of antibiotics and other anti-bacterials in feed for food-producing animals. The regulation resulted from an accumulation of evidence that long-term, low-level use of antibiotics in animal feed might result in the development of antibiotic-resistant bacteria, increasing the risk of bacterial disease in animals or humans. Upon the recommendation of an Antibiotic Task Force, the FDA, after notice and opportunity for comment, published a final regulation, 21 CFR § 135.109, the precursor of § 558.15, in the Federal Register of April 20, 1973. 38 Fed.Reg. 9811.^[1] At the time 21 CFR § 135.109 was proposed and issued, the FDA considered water soluble products to be included within the definition of "animal feed" and thus within the scope of the regulation. Neo-Terramycin is a water soluble drug and therefore was initially covered by the regulation.

The FDA originally interpreted "animal feed" in its regulation to include drinking water. "Animal feed" is defined in the Act as

... an article which is intended for use for food for animals other than man and which is intended for use as a substantial source of nutrients in the diet of the animal, and is not limited to a mixture intended to be the sole ration of the animal.

21 U.S.C. § 321(x) (1972). Under this definition, the FDA was not required to include drinking water and non-nutrient additives within the definition of animal feed, because neither water nor antibacterial additives provide a substantial source of nutrients in the diet of the animal.

Section 135.109 required the submission of new animal drug applications (NADA's) for drugs not previously covered by NADA's, and certain data, by specified dates. Those drugs for which NADA's were submitted in response to the regulation were accorded "interim marketing" privileges, i.e., the drugs were allowed to remain on the market even though data was unavailable conclusively establishing their safety and effectiveness. The interim marketing privilege was provided to allow the continued marketing of drugs which had been marketed prior to the adoption of the regulation while testing was done that would conclusively demonstrate whether a real danger of the development of an antibiotic resistant bacteria existed as a result of long-term, subtherapeutic use of the drug in animal feed. It is undisputed that Pfizer submitted a NADA for Neo-Terramycin on July 16, 1973, in compliance with § 135.109.

On August 6, 1974, the FDA published in the Federal Register a list of all of the *370 drugs that were to be covered by § 135.109. 39 Fed.Reg. 28382. No water soluble products were included in the list. See 39 Fed. Reg. 28394 et seq. On March 27, 1975, the FDA redesignated § 135.109 as § 558.15 as part of a general series of recodifications of FDA regulations. 40 Fed.Reg. 13802. The preamble to this recodification provided that the changes being made were nonsubstantive and therefore notice and public procedure was not required. Id. at 13803. A final order was issued by the FDA on February 25, 1976, listing the drugs covered by § 558.15. 41 Fed.Reg. 8282. Again, water soluble products were excluded from the list of drugs covered by the regulation.

When wetted down, the issues presented by Pfizer's motion are whether the FDA changed its interpretation of § 558.15 to exclude water soluble products and, if so, did it do so in a proper manner. The Court is of the opinion that the FDA properly did change its interpretation so that Neo-Terramycin is not covered by the interim marketing provision.

Neither party specifically addressed the proper scope of review to be conducted by the Court. In all cases, agency action must be set aside if the action was arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law. Citizens to Preserve Overton Park v. Volpe, 401 U.S. 402, 414, 91 S. Ct. 814, 822, 28 L. Ed. 2d 136 (1971); 5 U.S.C. § 706(2) (A) (1976). In certain narrow, limited circumstances, agency action is to be set aside if not supported by "substantial evidence." 5 U.S.C. § 706(2) (E) (1976). Review under the substantial evidence test is authorized only when the agency action is taken pursuant to the Administrative Procedure Act (APA), 5 U.S.C. § 553 (1976), or when the agency action is based on a public adjudicatory hearing. Overton Park, 401 U.S. at 414, 91 S. Ct. at 822. In other equally narrow circumstances, the reviewing court is to engage in a de novo review of the action and set it aside if it was "unwarranted by the facts." 5 U.S.C. § 706(2) (F) (1976). De novo review is authorized when the agency action is adjudicatory in nature and the agency fact-finding procedures are inadequate, or when issues that were not before the agency are raised in a proceeding to enforce a nonadjudicatory agency action. Overton Park, 401 U.S. at 415, 91 S. Ct. at 823.

De novo review is clearly inapplicable in this action. The Court does not decide whether the proper scope of review is the "arbitrary and capricious" standard or the "substantial evidence" standard, because the FDA's action in denying interim marketing approval to water soluble products meets the more stringent substantial-evidence test.^[2]

A court should accord substantial deference to the interpretation of a regulation given by the agency charged with its administration. United States v. Rutherford, 442 U.S. 544, 553-54, 99 S. Ct. 2470, 2475-76, 61 L. Ed. 2d 68 (1979); Udall v. Tallman, 380 U.S. 1, 16, 85 S. Ct. 792, 801, 13 L. Ed. 2d 616 (1965). Such deference is especially prudent in areas involving scientific and medical controversies. Public Citizen v. Foreman, 631 F.2d 969, 975 (D.C.Cir. 1980). The ultimate standard of review is a narrow one; the court is not empowered to *371 substitute its judgment for that of the agency. Overton Park, 401 U.S. at 416, 91 S. Ct. at 823. Even if the action involves a new interpretation of a regulation, the traditional standard of deference applies. Montana Power Co. v. Environmental Protection Agency, 608 F.2d 334, 347 (9th Cir. 1979).

An agency, when faced with new facts, or in light of reconsideration of the relevant facts, may alter past interpretations or overturn past administrative rulings and practice. American Trucking Assoc. v. Atchison, Topeka & Santa Fe Ry., 387 U.S. 397, 416, 87 S. Ct. 1608, 1618, 18 L. Ed. 2d 847 (1967); Spartan Radio Broadcasting Co. v. FCC, 619 F.2d 314, 322 (4th Cir. 1980). This principle, applied where a regulation was being amended in American Trucking and Spartan Radio, should be equally applicable in a situation where an agency is making a change not subject to the informal rule-making procedures of the APA. New interpretations may contradict the result obtained under the previous interpretations. Montana Power, 608 F.2d at 347. Generally, an agency's interpretation of the regulation should stand if it is one of several reasonable alternatives, even though it may not appear as reasonable as some other alternative. Expedient Service v. Weaver, 614 F.2d 56, 57 n.1 (5th Cir. 1980); Floyd S. Pike Electrical Contractor, Inc. v. OSHA, 576 F.2d 72, 75 (5th Cir. 1978).

Although the ultimate standard of review is a narrow one, a reviewing court must engage in a searching and careful review of the facts. Overton Park, 401 U.S. at 415-16, 91 S. Ct. at 823. This substantial inquiry into the facts must be made regardless of whether the agency action must meet the substantial evidence test or the less onerous arbitrary and capricious test. Id. at 415, 91 S. Ct. at 823.

The record in this action establishes that the FDA's interpretation of § 558.15 is supported by substantial evidence. When the FDA published its 1974 proposal in the Federal Register, it listed all the drugs that were to be covered by the regulation. This list did not include water soluble products, but was instead limited to feed premixes, i.e., drugs that were to be added to dry feed. The listings in § 558.15, and its predecessor § 135.109, refer to "use levels" and designate dosage in terms of "grams per ton." In contrast, the labeling for Neo-Terramycin is in terms of "milligrams per gallon."

The FDA changed its interpretation after studying the substance of the report of the Antibiotic Task Force. Following this evaluation of the report, the FDA concluded that the Task Force had really spoken to the use of antibiotics in dry feed. (Tr. 1141). The Task Force was specifically concerned with antibiotic drug resistance, which is most often a problem when there is subtherapeutic, long-term use of a product. (Tr. 1141-42). Water soluble products generally are not used very long in an animal, nor are they generally used at less than a therapeutic level. (Tr. 1142).^[3] As admitted by one of Pfizer's experts, § 558.15 was directed towards subtherapeutic use of antibiotics. (Tr. 1109-10). Therapeutic claims are made in the Neo-Terramycin labels.

When the FDA published its final list in 1976, it responded in comment 10 to a question regarding the absence from § 558.15 of a provision for a combination drug containing procine penicillin-steptomycin administered in drinking water. Comment 10, in part, stated:

The Commissioner advises that the deliberations and conclusions of the FDA Task Force on the Use of Antibiotics in Animal Feeds were directed at the use of antibacterial agents administered to animals in feeds. For this reason, the Commissioner has not included in § 558.15 drugs administered in drinking water. The safety and effectiveness of drugs administered in these dosage forms are being independently considered. *372 While Dr. Gerald B. Guest, the Acting Director of the FDA's Bureau of Veterinary Medicine, admitted that in the context of the definition used in the Task Force criteria that the above statement was incorrect (because the definition of feed included drinking water), he testified that in the context in which it was used by the FDA, the statement was correct. (Tr. 1168).

Finally, Pfizer places great emphasis on inconsistent letters written by the FDA after publication of the final list in 1976, in which Neo-Terramycin was referred to in the context of § 558.15, as indicating that the FDA did not change its interpretation of the regulation to exclude water soluble products. Erroneous letters do not necessarily bind an agency. See Pollock v. General Finance Corp., 552 F.2d 1142, 1144 (5th Cir. 1977); Bentex Pharmaceuticals, Inc. v. Richardson, 463 F.2d 363, 368 n. 17 (4th Cir. 1972), rev'd on other grounds, 412 U.S. 645, 93 S. Ct. 2488, 37 L. Ed. 2d 235 (1973); Maxwell Co. v. NLRB, 414 F.2d 477, 479 (6th Cir. 1969). The last inconsistent letter was written in February 1977. On June 19, 1978, the FDA warned Pfizer that Neo-Terramycin was not subject to the interim marketing exception of § 558.15. This warning was reiterated in letters to Pfizer dated March 15, 1979, and March 26, 1979. In addition, since the final list was published, the FDA has administratively withdrawn approval of water soluble products, which clearly indicates an interpretation of the regulation excluding water soluble products from the interim marketing provision. See 41 Fed.Reg. 28340 (July 9, 1976); 43 Fed.Reg. 7040 (Feb. 17, 1978).

There is substantial evidence, therefore, that the FDA changed its interpretation of its regulation to exclude water soluble products from the interim marketing privilege afforded under § 558.15. Similarly, this construction is supported by substantial evidence.

A final inquiry must be directed toward whether the FDA's action followed the necessary procedural requirements. Overton Park, 401 U.S. at 417, 91 S. Ct. at 824. See also note 2, supra. The Government contends that because it was merely defining a term in the regulation, the APA does not apply and notice and comment rulemaking was not required. Pfizer argues that notice and comment rulemaking was required, and that these procedures were not followed. The Court is of the view that the stricter requirements of informal rulemaking were complied with, and, as with the proper scope of judicial review, a determination as to what the proper procedure actually was need not be made.

The APA requires notice of proposed rulemaking and an opportunity for interested persons to participate in the rulemaking through the submission of written data, views, or arguments. 5 U.S.C. § 553(b) & (c). The notice must contain "either the terms or substance of the proposed rule or a description of the subjects and issues involved." 5 U.S.C. § 553(b) (3). The FDA complied with this notice requirement when it listed all of the drug products covered by the interim marketing provision. The proposed regulation listed the dosages only in grams per ton, indicative of an intent to exclude water soluble products. In addition, because the proposed list included all of the drugs that would be afforded the interim marketing privilege, each sponsor was obligated to review the proposed rule to determine whether its product was listed.

The APA requires general notice of proposed rulemaking unless the persons subject thereto have actual notice. In this case, all of the entities affected by the new interpretation were put on notice that their products were going to be denied the interim marketing privilege. Thus, they had an opportunity to comment and to inquire specifically why their product was not being afforded the interim marketing privilege.

Pfizer's final argument rests on due process grounds. Pfizer contends that since it initially was covered under the interim marketing provision, the FDA could not revoke interim marketing approval without complying with the regulations and *373 affording some type of hearing. As mentioned earlier, no rule of law forbids an agency from changing its regulations. Kelly v. United States Dept. of Interior, 339 F. Supp. 1095, 1100 (E.D.Cal.1972). Agencies need sufficient latitude to adjust their rules to reflect actual experience and may even reverse their thinking if necessary. Id. While these changes may not be done arbitrarily, a change in regulations that results in a loss of pre-existing benefits does not in itself show that the agency acted arbitrarily. Id.; General Telephone Co. of Southwest v. United States, 449 F.2d 846, 863 (5th Cir. 1971). Where a rule has retroactive effects, it may still be sustained despite such retroactivity if it is reasonable. Id.; Public Service Co. of Colorado v. Andrus, 433 F. Supp. 144, 154 (D.Col.1977).

It should be noted that an interpretative rule does not primarily have future effect and, therefore, is not a "rule" within the meaning of 5 U.S.C. § 551(4). Energy Consumers and Producers Ass'n. v. Dept. of Energy, 632 F.2d 129, 139-40 (Em. App.), cert. denied, 449 U.S. 832, 101 S. Ct. 102, 66 L. Ed. 2d 38 (1980). An interpretative rule is expressly exempt from the notice comment procedures of the APA. Id. at 142. Thus, if the effect of the changed interpretation of the rule in the case sub judice was to delete water soluble products from the scope of the interim marketing provision retroactive to the date when the provision was first effective, a hearing would not be required because there is no prospective effect.

Assuming that some form of due process must be afforded in order to revoke existing interim marketing approval, a hearing is still not necessarily required. The FDA established the interim marketing privilege for a class of sponsors through notice and comment rulemaking. Continued marketing of drugs while testing was done to establish their safety and effectiveness was a privilege afforded these sponsors by the FDA because the drugs had been marketed prior to the adoption of the regulation, and was not mandated by statute. The FDA revoked interim marketing approval for sponsors of water soluble products in the same way in which it was first given, by notice and comment rulemaking, when it listed all of the products covered by the interim marketing provision.

In determining if a rule revoking a privilege or expectation is reasonable, an important consideration is whether the challenger's conduct would have differed if the rule in issue had been applied from the start, and the challenger is prejudiced by the change. Adams Nursing Home of Williamstown, Inc. v. Mathews, 548 F.2d 1077, 1081 (1st Cir. 1977). See also Springdale Convalescent Center v. Mathews, 545 F.2d 943, 956 (5th Cir. 1977). Had the rule been applied from the beginning, Pfizer would not have had the benefit of interim marketing at all. As it stands, Pfizer has been able to market Neo-Terramycin several years longer than it otherwise might had the interpretation been applied from the start. Pfizer also has had an opportunity to present its views to the agency that water soluble drugs should be covered by the interim marketing provision. It is difficult to see how Pfizer has been prejudiced.

Based on the above analysis, the motion for disposition under 21 CFR § 558.15 should be denied.

The government's motion for summary judgment is predicated on its contention that Neo-Terramycin is a "new animal drug" within the meaning of 21 U.S.C. § 321(w).^[4] The government has limited its argument for purposes of this motion to the *374 contention that Neo-Terramycin is not generally recognized as effective for the indications on the labeling.

The standards to be applied by a court in ruling on a motion for summary judgment are well established. The trial court has no duty to decide factual issues, only whether there is an issue of fact to be tried. Jones v. Western Geophysical Co. of America, 669 F.2d 280, 283 (5th Cir. 1982). The party seeking summary judgment has the burden of demonstrating that there exists no genuine issue as to any material fact. United States v. An Article of Food Consisting of 345/50-Pound Bags, 622 F.2d 768, 771 (5th Cir. 1980). In reviewing the pleadings, depositions, answers to interrogatories, admissions, and affidavits to determine whether a genuine issue of material facts exists, all reasonable doubt must be resolved in favor of the party opposing the motion for summary judgment. Id. The fact that it appears that the nonmover is unlikely to prevail at trial or that the mover's facts are more plausible are not reasons to grant summary judgment. Id. at 773; Jones, 669 F.2d at 283.

In order to succeed on its motion for summary judgment, the government has the burden of demonstrating that no material issue of material fact exists as to either of two issues: the lack of a general reputation for the product's efficacy among the appropriate experts or the lack of an adequate foundation upon which to base such a general reputation. See United States v. Articles of Food & Drug Consisting of Coli-trol 80, 518 F.2d 743, 746 (5th Cir. 1975); 345/50-Pound Bags, 622 F.2d at 771. The government contends that the record establishes that there is a genuine dispute among experts; therefore, there can be no general reputation for effectiveness. The government also argues that there is inadequate scientific support for a conclusion that Neo-Terramycin is effective.

The general rule is that where there is a genuine difference of opinion among experts properly qualified by training and experience, then a drug cannot be generally recognized as effective. United States v. An Article of Drug ... Furestrol ..., 294 F. Supp. 1307, 1311 (N.D.Ga.), aff'd, per curiam, 415 F.2d 390 (5th Cir. 1968). General recognition does not require unanimity; instead, the adverb "generally" is defined to mean "In General; extensively, though not universally; most frequently, but not without exception; ..." United States v. 7 Cartons ... Ferro-Lac, 293 F. Supp. 660, 662-63 (S.D.Ill.1968), modified on other grounds, 424 F.2d 1364 (7th Cir. 1970) (quoting Webster's Dictionary); United States v. An Article of Drug ... Bentex Ulcerine, 469 F.2d 875, 877 (5th Cir. 1972) (per curiam). Existence of a "severe conflict" in the expert testimony would preclude a finding that a drug is generally recognized as effective. United States v. X-Otag Plus Tablets, 441 F. Supp. 105, 113-14 (D.Col.1977), aff'd, 602 F.2d 1387 (10th Cir. 1979). Thus, upon a showing that there is a disagreement among properly qualified experts on the issue of general recognition of effectiveness, it must be concluded that the drug is a new animal drug under the Act. United States v. Articles of Drug ... Beuthanasia D Regular, Food Drug Cosmetic Law Rep. (CCH) ¶ 38,265, (D.Neb.1979).

The government submitted affidavits by eminently qualified experts attesting to the lack of general recognition of effectiveness of Neo-Terramycin. From this, the government concludes that there can be no showing of general recognition of effectiveness. See Id. The government argues that in the context of general recognition for effectiveness, the existence of a genuine dispute among experts establishes that there is not a genuine material issue of fact for trial.

Unfortunately, this issue is not so easily resolved. The question presented is a factual one as to whether there is a genuine difference of opinion among experts qualified by scientific training and experience. Furestrol, 294 F. Supp. at 1311. It is not enough to show merely a conflict in the evidence of general recognition. Colitrol *375 80, 518 F.2d at 746. At the same time, in ruling on a motion for summary judgment, the Court should not assess the probative value of evidence presented. 345/50-Pound Bags, 622 F.2d at 773. In order to determine whether a genuine dispute exists between experts in this action, the Court would normally be required to assign some probative value to the government's affidavits where there is a direct conflict between experts. Because this is forbidden on a motion for summary judgment, a trial on the merits would generally be required on the issue of whether a genuine dispute between experts exists.^[5]

The standard of general recognition for effectiveness requires "substantial evidence"^[6] that the drug is effective. United States v. An Article of Drug ... Entrol-C Medicated, 513 F.2d 1127, 1128-29 (9th Cir. 1975). The government contends that no substantial evidence exists from which an expert could reasonably conclude that Neo-Terramycin is generally recognized as effective. In response, Pfizer argues that the studies cited by its experts in fact constitute substantial evidence of effectiveness, or at least create a material issue of fact which precludes summary judgment.

In Weinberger v. Hynson, Westcott & Dunning, 412 U.S. 609, 629-30, 93 S. Ct. 2469, 2483, 37 L. Ed. 2d 207 (1973), the Supreme Court established the general rule that before a drug can be generally recognized as effective, substantial evidence of effectiveness at least equivalent to that required for approval of a NADA must be presented.^[7] Specifically, there must be adequate and well-controlled investigations supporting the drug's efficacy. Id., Coli-trol 80, 518 F.2d at 746. In a companion case decided the same day, the Supreme Court applied this rule in stating that

"[w]hether a particular drug is a "new drug" depends in part on the expert knowledge and experience of scientists based on controlled clinical experimentation and backed by support in scientific literature."

Weinberger v. Bentex Pharmaceuticals, Inc., 412 U.S. 645, 652, 93 S. Ct. 2488, 2493, 37 L. Ed. 2d 235 (1973).

Three charges are leveled against the studies Pfizer cites by the government in support of its contention that as a matter of law substantial evidence of Neo-Terramycin's effectiveness has not been produced:

(1) None of the studies that Pfizer contends are adequate and well-controlled were done with the drug itself, i.e., no studies were done with Neo-Terramycin administered in drinking water;

(2) None of the studies were conducted showing the efficacy of the combination of neomycin and oxytetracycline sufficient to meet the requirements of the FDA's combination policy; and

(3) There are no studies of any kind for some of the label claims.

The Court agrees that as a matter of law Pfizer has not controverted the government's showing that there is no substantial *376 evidence of effectiveness established by adequate and well-controlled investigations.

As a starting point, the Court notes that the test of substantial evidence is rigorous, reflecting the conclusion that clinical impressions of practicing physicians and poorly controlled experiments do not constitute an adequate basis for establishing efficacy. Hynson, 412 U.S. at 630, 93 S. Ct. at 2483. In addition, the studies must be widely published in recognized scholarly journals in order for there to be general recognition of effectiveness. Coli-Trol 80, 518 F.2d at 747; Bentex Ulcerine, 469 F.2d at 880; United States v. 41 Cases, More or Less, 420 F.2d 1126, 1130 (5th Cir. 1970); United States v. An Article of Drug ... Mykocert, 345 F. Supp. 571, 575 (N.D.Ill. 1972).

In this case, the government has shown that there are no studies of the effectiveness of Neo-Terramycin administered in drinking water, no reports of in vivos tests involving Neo-Terramycin, and that many of the studies appeared only in Pfizer publications rather than in recognized scientific journals.

FDA regulations provide that a new method of administering a drug may cause a drug to be a new animal drug even though when administered by another method it is not a new animal drug. 21 CFR § 510.3(i).^[8] Other courts have held that general recognition of effectiveness of a drug in another form, dosage, or method of administration does not constitute general recognition of the drug in its exact form, dosage or method of administration. United States v. Articles of Drug ... Colchicine, 442 F. Supp. 1236, 1243 (S.D.N.Y.1978); X-Otag Plus Tablets, 441 F. Supp. at 111; Mykocert, 345 F. Supp. at 575. See Hoffman v. Sterling Drug, Inc., 485 F.2d 132, 137 (3d Cir. 1973).

Therefore, because there are not studies showing the effectiveness of Neo-Terramycin administered in drinking water, there can be no general recognition unless each of the component parts are generally recognized "with the critical caveat that the combination of these parts does not in any way create a new drug...." Mykocert, 345 F. Supp. at 575;^[9]Entrol-C Medicated, 513 F.2d at 1129. This requires that there must be general recognition that such a combination, although not in the exact form dosage or application of the recognized drugs, will yield a result no different in terms of effectiveness than that of the recognized drugs; merely showing that there is a general recognition for each component part is insufficient. Id. at 1129-30.

The rationale for special treatment of combination drugs is that a single component of a drug may react with other components, enhancing or reducing the total effect of the drug, or producing unexpected side effects. Coli-Trol 80, 518 F.2d at 746; X-Otag Plus Tablets, 441 F. Supp. at 111. The FDA regulations also recognize that newness of an animal drug may arise by the combination of two or more substances, none of which is itself a new animal drug. 21 CFR 510.3(i) (2). There must be at least one adequate, well-controlled, completed test of the combinations being challenged. Coli-Trol 80, 518 F.2d at 747.

A further requirement is that it must be shown that there is some benefit to the fixed combination in terms of greater effectiveness. X-Otag Plus Tablets, 441 F. Supp. at 111. This requirement is implicit in the FDA's combination policy, codified at 21 CFR § 514.1(b) (8) (v):

*377 Each ingredient designated as active in any new animal drug combination must make a contribution to the effect in the manner claimed or suggested in the labeling, and, if in the absence of express labeling claims of advantages for the combination such a product purports to be better than either component alone, it must be established that the new animal drug has that purported effectiveness.

As administered by the FDA, the combination policy requires studies designed to show whether each ingredient makes a contribution by comparing the effectiveness of the combination with the effectiveness of the active ingredients alone and with a control group. See Masti-Kure Products Co. v. Califano, 587 F.2d 1099, 1101-02 (D.C. Cir.1978). The policy thus requires a showing that the combination is more effective than either of its components alone. Id. at 1104.

As previously mentioned, none of the studies relied on by Pfizer were done with Neo-Terramycin itself. In addition, none of the cited studies were factorial design 4 × 4, which compares four test groups: one treated with the combination, one treated with one of the active ingredients separately, one treated with the other active ingredient separately, and a control group. Finally, Pfizer offered only one study purporting to show that the Neo-Terramycin as a combination has a synergistic effect, i.e., that the combination itself is more effective than its individual components. The study, however, was done only in vitro and not in the animal. In vivos tests would appear to be required. See Coli-Trol 80, 518 F.2d at 747.

Pfizer attacks the government's reliance on the combination policy on several grounds. It contends that by the terms of the regulation itself, the combination policy comes into play only if there are express label claims of advantages for the combination or the product purports to be better than the ingredients alone; according to Pfizer, the government has not shown any such label claims or otherwise demonstrated how the product purports to be better. Additionally, Pfizer argues that the combination policy does not apply in "general recognition" cases, and that there is no rationale for the combination policy as applied to animal drugs.

While there are not express label claims of the advantages of Neo-Terramycin as a combination, it is apparent that Neo-Terramycin purports to be better than its component parts. Oxytetracycline and neomycin have different spectra of antibiotic activity, acting against different organisms, although there is some overlap. The two drugs are also active in different areas, neomycin primarily in the gut and oxytetracycline primarily systemically. As attested by Pfizer's experts, this combination should be more effective than the individual components alone.^[10] The government's experts attested to the fact that the dosage levels for oxytetracycline must be substantially higher than those indicated on the labeling for Neo-Terramycin in order to have therapeutic effect if oxytetracycline was being administered alone. This was uncontroverted by Pfizer's experts. Clearly, the inference implied by the label is that Neo-Terramycin purports to be better than its individual components.

With respect to Pfizer's contention that the combination policy applies only to agency approval of NADA's and not to general recognition of effectiveness, it is enough to note that for a drug to be generally recognized it must have the same quantity and quality of data as is required for FDA approval. Hynson, 412 U.S. at 629, 93 S. Ct. at 2483. A combination drug must meet the requirements of the combination policy before it can be approved, and there is no reason why a more lenient standard *378 should be applied in the area of general recognition.^[11]

The reasons justifying the combination policy are sound. New combinations of even well-known drugs constitute new drugs for purposes of the Act exactly because the effects of the drugs in combinations are not the sum of its parts. Coli-Trol 80, 518 F.2d at 746. In addition, there should be scientific evidence to support the claims for any drug expressly or by implication purporting to be better than the sum of its parts. Pfizer suggests that while the combination policy is justified when it is applied to humans, since the policy is based on a desire to avoid unnecessary uses of medication in order to avoid unnecessary adverse side effects, it has no application to feed producing animals, because the focus should be on the economics of production. This involves immediate treatment on a wide spectrum in order to insure survival of the maximum number of animals, and the possibility that an animal may receive more medication than absolutely necessary should not be of overriding concern. This ignores other substantive considerations behind the policy, including assuring that the drug meets its claims of effectiveness. The combination policy can also be interpreted as defining adequate and well-controlled investigations sufficient to provide substantial evidence of a combination drug's effectiveness.

As a final ground for its motion, the government contends that there are no studies for about one half of the poultry label claims. Pfizer, in response, argues that it is possible to extrapolate that Neo-Terramycin is effective for these label claims based on the studies that were done with respect to the other poultry claims. Pfizer's experts stated that if one or more clinical investigations show that an antibiotic is effective against a particular disease-causing organism, it is possible to conclude fairly and reasonably that the same drug will also be effective against other sensitive organisms in that system, after making any necessary adjustments for dosage and other applicable variables, with the caveat that it is necessary to have a variety of additional information available as part of one's training and experience to reach such conclusions.

Given the large number of claims for which no specific studies have been done, substantial evidence may not be present to support these label claims. It should be noted that in some contexts the FDA has permitted extrapolations to a general indication or a list of indications on the basis of studies in a few indications. This is not, however, the FDA's policy with respect to antibiotics. No extrapolations are permitted, at least with antibiotics for human use, and a sponsor must demonstrate effectiveness of the drug in humans against each bacterium he wishes to claim and in each organ for which he wishes to claim effectiveness of the drug in eradicating infection. Also, of the fourteen studies cited by Pfizer's experts as substantial evidence of effectiveness of Neo-Terramycin in poultry, nine did not include neomycin. Four of the remaining five studies were reported only in Pfizer publications, rather than recognized scientific journals, and of these four, two reported studies that measured the effect of Neo-Terramycin in feed in increasing weight gains and improving feed efficiency, which are non-disease indications not appearing on the Neo-Terramycin label. The remaining study cited is the in vitro study indicating synergism between neomycin and oxytetracycline.

*379 The labeling for Neo-Terramycin lists a total of seventeen different drug claims, considering the separate claims for prevention, treatment, and control. It is fundamental that a drug must be generally recognized for all of the conditions for which it is labeled to avoid new drug status. See, e.g., United States v. An Article of Drug ... Quinaglute, 268 F. Supp. 245 (E.D. Mo.1967); Merritt Corp. v. Folsom, 165 F. Supp. 418 (D.C.1950); 21 U.S.C. § 321(w). Thus, a finding that a drug is not generally recognized as effective for one or more of the label claims would result in a determination that the product is a new drug, even if it is assumed that it is generally recognized as effective for the remaining label claims.

Two of the claims made on the label of Neo-Terramycin are that it is effective for the prevention and control of bluecomb in chickens and turkeys. The government's expert stated that the primary cause of bluecomb is a virus, although there is a secondary bacterial component. Antibiotics such as Neo-Terramycin are not effective against viruses. Pfizer's experts dispute that the cause of the disease has been established to be a virus, and go on to contend that there is clearly a secondary bacterial component against which antibiotics are effective.^[12] If bluecomb is in fact caused by a virus, it is equally clear that Neo-Terramycin cannot prevent bluecomb, whatever benefits it may have against the secondary bacterial infection with respect to helping the bird recover faster or preventing death loss in the flock from the secondary bacterial organisms associated with the disease. Although Dr. Vezey's declaration may create an issue of fact with respect to the causes of bluecomb, see note 12, supra, he previously stated at the trial of this action that he believed bluecomb is probably caused by a viral agent, and that antibiotics will have no effect on the virus. (Tr. 863).

Even assuming that an issue of fact is created regarding the causes of bluecomb, there are still no adequate and well-controlled studies from which it can reasonably be concluded that Neo-Terramycin is effective against bluecomb. The study upon which Dr. Vezey relied, claimants exhibit 71 at the trial, did not include neomycin as one of the antibiotics studied. In order to reach the conclusion that Neo-Terramycin is generally effective against bluecomb, a significant amount of extrapolation would have to be done, including reaching the conclusions that a change in the method of administration and dosage does not affect the effectiveness of the drug, and that the addition of neomycin does not change the effectiveness of the drug. In addition, if the combination policy is applicable, it must also be assumed that neomycin makes a contribution to the total effectiveness of the drug. Therefore, the Court is of the opinion that even if the study on which Dr. Vezey relied is enough for general recognition of effectiveness of oxytetracycline against bluecomb, there are no studies on which to base the conclusion that Neo-Terramycin is generally recognized as effective against bluecomb. See Coli-Trol 80, 518 F.2d at 747.

As a final ground for denial of the government's motion, Pfizer argues that general recognition of effectiveness may be based on the informed judgment of experts relying upon experience and careful observation over long periods in numerous herds and flocks of animals. Support for the proposition, it is contended, can be found in the limited exception to the Hynson rule requiring adequate and well-controlled studies *380 propounded in Bentex, 412 U.S. 645, 93 S. Ct. 2488, 37 L. Ed. 2d 235. In Bentex, 412 U.S. at 652, 93 S. Ct. at 2493, the Supreme Court stated in dicta that:

It may, of course, be true that in some cases general recognition that a drug is efficacious might be made without the kind of scientific support necessary to obtain approval of an NDA. But as we indicate in Hynson, supra [412 U.S.] at 631 [93 S. Ct. at 2484], 37 L. Ed. 2d at 224, the reach of scientific inquiry under [the section of the Act providing for approval of NDA's and under the section defining new drugs] is precisely the same.

Pfizer contends that long-term use of Neo-Terramycin in hundreds or thousands of herds or flocks, coupled with its apparent effectiveness, makes adequate and well-controlled studies scientifically unnecessary for general recognition of effectiveness among experts.

The exception enunciated in Bentex has been described as "limited but indefinite." Coli-Trol 80, 518 F.2d at 746. In Hynson, 412 U.S. at 630, 93 S. Ct. at 2483, the Supreme Court stated that the substantial evidence requirement reflected the conclusion of Congress that clinical impressions of practicing physicians and poorly controlled experiments do not constitute an adequate basis for establishing efficacy. The Supreme Court then rejected the contention that general recognition of effectiveness could be based "merely on expert testimony and reports with respect to investigations and clinical observation regardless of the controls used." Id. at 631, 93 S. Ct. at 2484. At the same time, the view that long-term use of a drug constitutes substantial evidence under the theory that the drug had passed tests of safety and efficacy by trial and error has also been specifically rejected. Coli-Trol 80, 518 F.2d at 746.

Pfizer's experts attested to their belief that long-term observation of hundreds or thousands of animals who have been treated with Neo-Terramycin constitutes substantial evidence that the drug is effective and that the positive results were not due to chance. It is precisely this kind of anecdotal evidence, however, that was rejected by the Supreme Court in Hynson. It may, of course, be used as corroborative support for effectiveness established by adequate and well-controlled investigations. But the Court is of the view that because there is an absence of adequate and well-controlled investigations sufficient to support a finding of general recognition of effectiveness, field observation of the effectiveness of Neo-Terramycin cannot constitute substantial evidence of effectiveness.

There have been no adequate and well-controlled studies performed with Neo-Terramycin in its soluble form. Similarly, no studies have been conducted showing the efficacy of the combination of neomycin and oxytetracyline, and there are no studies specifically directed at determining the effectiveness of Neo-Terramycin for several of the label claims. Specifically, there are no studies from which it can be concluded that Neo-Terramycin is generally effective against bluecomb. Alone, any of these deficiencies may be sufficient to support the conclusion that there is not an adequate foundation from which a qualified expert could reasonably conclude that Neo-Terramycin is generally recognized as effective for all of its label claims. In combination, the lack of adequate and well-controlled studies at the various levels compels the conclusion that Neo-Terramycin is not generally recognized as effective for all of its label claims. Pfizer has not demonstrated that a material issue of fact is presented which would preclude reaching this result on summary judgment. Therefore, the government's motion for summary judgment should be granted. The government is directed to submit a proposed judgment within 10 days of the file date of this Order. Pfizer will have 5 days from the date the proposed judgment is filed to raise any objections.

Accordingly it is ORDERED that Pfizer's motion for disposition under 21 CFR § 558.15 be denied and the motion for summary judgment filed by the United States be granted.

[1] 21 CFR § 135.109(a) provided that:

(a) The Commissioner of Food and Drugs will propose to revoke currently approved subtherapeutic (increased rate of gain, disease prevention, etc.) uses in animal feed of antibiotic and sulfonamide drugs whether granted by approval of new animal drug applications, master files and/or antibiotic or food additive regulations, by no later than 2 years following the effective date of this order, unless data are submitted which resolve conclusively the issues concerning their safety to man and animals and their effectiveness under specific criteria established by the Food and Drug Administration based on the guidelines included in the report of the FDA task force on the use of antibiotics in animal feeds. All persons or firms previously marketing identical, related, or similar products not the subject of an approved new animal drug application must submit a new animal drug application by July 19, 1973, if marketing is to continue during the interim. New animal drug entities with antibacterial activity not previously marketed, now pending approval or submitted for approval prior to, on, or following the effective date of this publication, shall satisfy such criteria prior to approval.

[2] Determination of the proper standard would appear to hinge on whether the FDA's change involved a change in interpretation of § 558.15 or involved an amendment with the concomitant notice and comment rulemaking requirements under the APA. The government argues that the FDA merely changed its definition of animal feed and cites to cases holding that the definition of a term used in a regulation is an interpretative ruling. Garelick Mfg. Co. v. Dillon, 313 F.2d 899 (D.C.Cir.1963); U. S. v. 353 Cases ... Mountain Valley Mineral Water, 247 F.2d 473 (8th Cir. 1957); Gibson Wine Co. v. Snyder, 194 F.2d 329 (D.C.Cir.1952). The APA excludes interpretative rulings from its procedural requirements. 5 U.S.C. § 553(b) (A). Pfizer, in turn, contends that deleting water soluble products from the group of products accorded interim marketing privileges has a substantive impact on its pre-existing rights, requiring notice and comment. See Reynolds Metal Co. v. Rumsfeld, 564 F.2d 663, 669 (4th Cir. 1977), cert. denied, 435 U.S. 955 (1978); Energy Reserves Group, Inc. v. Department of Energy, 589 F.2d 1082, 1102-03 (Em.App.1978) (Judge Christensen concurring). See infra for a further discussion of this issue.

[3] Testimony by administrative officials who participated in a decision which explains their action can be part of the substantial inquiry into the facts and is not a mere "post hoc" rationalization. See Overton Park, 401 U.S. at 419-20, 91 S. Ct. at 825.

[4] 21 U.S.C. § 321(w) provides, in part:

(w) The term "new animal drug" means any drug intended for use for animals other than man, including any drug intended for use in animal feed but not including such animal feed,

....

(1) 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 animal drugs, as safe and effective for use under the conditions prescribed, recommended, or suggested in the labeling thereof; ...

[5] In view of the Court's holding in Part II(B) of this Order, however, it would be possible to base summary judgment on a lack of general recognition for effectiveness. A determination that there is an inadequate foundation to support an expert's opinion that there is a general reputation for effectiveness would leave uncontroverted the opinion of the government's experts that no such general reputation exists.

[6] "Substantial evidence" is defined in the Act to mean "evidence consisting of adequate and well-controlled investigations, including field investigation, 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 reasonably 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." 21 U.S.C. § 360b(d) (3).

[7] Hynson dealt with non-animal drugs under 21 U.S.C. § 321(p) (1). The principles enunciated in Hynson and its progeny, however, are directly applicable to general recognition of effectiveness of animal drugs. See United States v. Naremco, 553 F.2d 1138, 1142 n. 5 (8th Cir. 1977); Coli-trol, 518 F.2d at 746; Entrol-C, 513 F.2d at 1128.

[8] 21 CFR § 510.3(i) provides, in part:

The newness of animal drug may rise by reason of ... (6) the newness of a ... method of administration ... even though such drug ... when used in another ... method ... of administration ... is not a new animal drug.

[9] Of course, it can be argued that as a matter of law general recognition of effectiveness may never be demonstrated from general recognition of effectiveness of the drug's constituent ingredients in those same uses. The Fifth Circuit, when presented with this contention, distinguished Furestrol, 415 F.2d 390, as not supporting the proposition, but otherwise did not rule one way or the other. United States v. ... Capsules ... Afrodex, 494 F.2d 1158, 1161 n. 2 (5th Cir. 1974).

[10] The important point, however, is not whether in theory the combination should be more effective, but whether there are adequate and well-controlled investigations establishing that the combination is more effective. There are no such adequate and well-controlled studies in evidence.

[11] A sponsor of a combination drug purporting to be better than its component parts may be placed in a Catch-22 position if he is attempting to show the drug is not new by showing general recognition of the component parts. First, in order to meet the requirements of the combination policy, there must be adequate and well-controlled studies of the combination itself. The lack of these investigations is precisely why the drug sponsor attempts to show general recognition of the combination through general recognition of the individual components. Second, the combination policy requires a showing of greater effectiveness of the combination. This requirement is contrary to the requirement that a result no different in terms of effectiveness than the recognized drugs be produced by the drugs in the combination.

[12] Pfizer's expert, Dr. Stanley A. Vezey, D.V.M., stated in his declaration:

Bluecomb is the name of a disease that was applied to chickens and turkeys a number of years ago. The cause of the disease at the time it was named was not clearly understood and it is still not understood. Although some experts believe it is caused by a virus, it is clear that it contains a bacterial component. Sieburth, J. M., et al., "Transmissible Entenities of Turkeys (Bluecomb Disease) Preliminary Studies," Poultry Science, 1956, Vol. 36 pp. 256-261 (Claimant's Exhibit 71) describes an adequate and well-controlled study which in my opinion provides a basis upon which experts qualified by scientific training and experience can say that Neo-Terramycin Soluble Powder Concentrate is effective for bluecomb.

Vezey Declaration, ¶ 12 n. *.