This is a petition by Pfizer, Inc., a prominent pharmaceutical manufacturer, pursuant to § 701(f) of the Federal Food, Drug and Cosmetic Act, 21 U.S.C. § 371(f), here applicable by virtue of the last sentence of § 507(f), to review a final order of the Food and Drug Administration (FDA) issued July 14, 1970.1 The order, made pursuant to § 507(f) of the Act, see fn. 3 infra, repealed regulations, first issued during the 1950's, authorizing the distribution of six antibiotic drugs manufactured by Pfizer and revoked all certificates of safety and effectiveness thereunder. The FDA took this action without granting Pfizer an evidentiary hearing upon its objections. It sought to justify this summary procedure on the basis of a finding that Pfizer had failed to state "reasonable grounds" for its objections within the meaning of § 507(f). Judge Anderson granted a stay of the order pending argument of the petition on an expedited schedule; at argument we extended the stay until the filing of an opinion.

All six of the drugs are to be taken orally. Three are called Signemycin and three Sigmamycin. All are "combination" drugs, containing two parts of tetracycline (or tetracycline hydrochloride), long and still approved antibiotics, and one part of triacetyloleandomycin (or oleandomycin phosphate), antibiotics also long and still approved. Since there is no contention that the case with respect to any of one of the preparations differs from the others, we shall follow the parties in referring to all six drugs as Signemycin.2

It will be useful to begin with some history. Section 505 of the Federal Food, Drug and Cosmetic Act of 1938, 52 Stat. 1052-53, prohibited the introduction into interstate commerce of any drug unless an application therefor had become effective. Safety was the primary criterion for passing on such applications, § 505(d). Section 505(e) provided that once an application had become effective, it could be suspended, after due notice and opportunity for hearing to the applicant, if the Secretary, then of Agriculture, found that clinical experience, tests by new methods, or tests by methods not deemed reasonably applicable when the application became effective showed that the drug was unsafe for use, or that the application contained any untrue statement of a material fact.

Seven years later Congress amended the Act to deal with the first of the antibiotics, penicillin, by enacting an entirely new provision, § 507, 59 Stat. 463-64. This empowered the Federal Security Administrator, to whom administration of the Food, Drug and Cosmetic Act had then been confided, to provide for the certification of batches of drugs composed of penicillin or its derivatives. He was to do this "if such drug has such characteristics of identity and such batch has such characteristics of strength, quality, and purity, as the Secretary prescribes in such regulations as necessary to adequately insure safety and efficacy of use." Section 507(b) directed him to issue regulations providing for such certification. Section 507(f) stated that any interested person could apply for the issuance, amendment, or repeal of any such regulation, and any interested person could file objections to such action. The section went on to say, substantially as it still does, that if reasonable grounds for such objections had been stated, the Administrator was to hold a hearing thereon. No drug subject to § 507 was to be deemed subject to § 505. The teeth of the statute were an addition to § 502 of a new paragraph (l) making a penicillin drug not complying with § 507 a misbranded article. Subsequent amendments added four other named antibiotics and their derivatives, including in 1949 chlortetracycline, 63 Stat. 409-10. The net of all this was that while the standard antibiotics were not subject to the elaborate "new drug" procedures of § 505 as such, they were under even more stringent regulation in two respects. Certification of batches of antibiotic drugs was required, and the drugs had to meet a standard of efficacy as well as of safety.

In 1962 Congress substantially amended the Food, Drug and Cosmetic Act. Section 505 was greatly stiffened. The Secretary, now of Health, Education and Welfare, was authorized by § 505(d) to deny approval of a new drug not only if "he has insufficient information to determine whether such drug is safe for use * * *" but also if "there is a lack of substantial evidence 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." The subsection provided that:

As used in this subsection and subsection (e) of this section, 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 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.

Section 505(e) authorized withdrawal of approval of an application not only for falsity or for considerations relating to safety, stated more broadly than in the 1938 Act, but also

on the basis of new information before him with respect to such drug, evaluated together with the evidence available to him when the application was approved, that there is a lack of substantial evidence 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 thereof.

However, withdrawal could be effected only "after due notice and opportunity for hearing to the applicant" except when the drug was found to be an immediate danger to public health, in which event the Secretary might suspend approval immediately and give the applicant an expedited hearing.

Section 507 was also amended. Section 507(a) was broadened to include all antibiotic drugs, which were defined. An entirely new section, 507(h), established a special procedure governing antibiotics previously approved under § 505, i. e., those antibiotics not specifically named in the pre-1962 version of § 507, and brought within the latter section for the first time under the 1962 revision. The initial issuance of regulations providing for their certification or exemption under § 507 was not to "be conditioned upon an affirmative finding of the efficacy of such drug." Any subsequent amendment or repeal of a regulation on the ground of a lack of efficacy could be effected

only if (1) such amendment or repeal is made in accordance with the procedure specified in subsection (f) of this section (except that such amendment or repeal may be initiated either by a proposal of the Secretary or by a petition of any interested person) and (2) the Secretary finds, on the basis of new information with respect to such drug evaluated together with the information before him when the application under section 505 became effective or was approved, that there is a lack of substantial evidence (as defined in section 505(d)) that the drug has the effect it purports or is represented to have under such conditions of use.

Section 507(f), which controls here, was left relatively unchanged; we quote it in full in the margin.3

The 1962 amendments presented the FDA with an enormous task of reviewing the efficacy of all new drugs that had been cleared, for safety only, between 1938 and October 10, 1962. On July 9, 1966, it announced, 31 F.R. 9426, an agreement to enlist the National Academy of Sciences-National Research Council (NAS-NRC) in this enterprise. The agreement also contemplated a survey of those antibiotic drugs like Signemycin, which had been certified or exempted from certification under § 507 prior to October 10, 1962. To implement this agreement with respect to antibiotics and to facilitate the task of the Council in determining whether any certification or release should be rescinded or any regulation under § 507 should be amended or repealed, the FDA, three months later, issued an order requiring each manufacturer to furnish specified information about the antibiotics in question. This included a "[l]ist of literature references most pertinent to an evaluation of the effectiveness of the drug for the purposes for which it is offered in the label, package insert, or brochure" accompanying its sale and "[u]npublished articles or other data pertinent to an evaluation of the claims." Pfizer responded with respect to the Signemycin products.

In the winter of 1969, a panel of six physicians, chaired by Dr. William Kirby, reported on its study of the Signemycin drugs. The conclusion was that they were "ineffective as a fixed combination." We quote as an example a portion of its report on the most basic claim with respect to Signemycin Syrup:

I. Signemycin is indicated in the therapy of acute severe infections caused by susceptible organisms and primarily by bacteria more sensitive to the combination than to either component alone.

EVALUATION: Ineffective as a fixed combination.

COMMENTS: Although either tetracycline or oleandomycin is effective in the treatment of some infections, there are serious objections to the use of the combination as formulated. Oleandomycin is present in insufficient amounts for effective treatment. If sufficient oleandomycin is administered for effective treatment, tetracycline overdosage results. The panel is not aware of infections caused by bacteria more sensitive to this combination than to either of its components.

Establishing efficacy would involve demonstrating that the clinical response to the combination is greater than to either agent used alone. No such evidence is available. In vitro studies of this combination have given variable results. The effects may be antagonistic, additive, or synergistic, and are unpredictable for any particular organism.

Thus, there is no evidence that each of the active ingredients contributes to the effect as claimed.

Four references were cited in support of this conclusion. Other adverse conclusions were stated to be based on the "[i]nformed judgment of the Panel." In response to this report, the FDA issued a notice on April 2, 1969, 34 F.R. 6008, expressing concurrence with the view that there was a "lack of substantial evidence that each ingredient in the combination drugs listed above contributes to the total effects which the drugs are purported or represented to have under the conditions of use prescribed, recommended, or suggested in their labeling" and an intention to initiate proceedings to amend the antibiotic drug regulations to discontinue certification of the products containing these combinations. Interested parties who might be adversely affected were given 30 days to submit pertinent data with respect to this proposal.

Pfizer responded by letter on April 30. It cited studies of animals and in vitro which were alleged to demonstrate the greater effect of the combination than of either component and successful clinical experience, and appended numerous references to the literature. The letter also recognized the development of new laboratory and diagnostic procedures permitting more accurate evaluation than in the past. Accordingly Pfizer proposed three sets of new studies: One would be an in vitro study, taking three to four months, wherein 500 "clinical isolates" would be separately tested with tetracycline, oleandomycin, and a 2-1 combination. A second study, of similar duration, would cover four groups of 20 monkeys each, who had been inoculated to produce staphylococcal septicemia; one group would be orally treated with tetracycline, another with triacetylo-leandomycin, a third with the 2-1 combination and the fourth with saline solution. Most important, Pfizer proposed a study of 600 patients having "a Gram + infection" divided into three groups of 200, one to be treated with tetracycline, another with oleandomycin, and the third with the combination. The patient's "cultures and sensitivities" would be examined and a wide variety of other tests would be made. It estimated that this study would take six to nine months.

The next step was the FDA's publication in the Federal Register on September 19, 1969, 34 F.R. 14596, of what it described as "Procedural and Interpretative Regulations" governing, among other things, the repeal of antibiotic drug regulations.4 This announced that before initiating proceedings for such repeal based on adverse reports of the National Academy of Science — National Research Council, Drug Efficacy Group, the agency would afford interested persons "an opportunity to present any available evidence that would satisfy the requirements of law as defined by the term `substantial evidence.'" In the absence of such evidence a formal proceeding would be initiated. Persons who would be adversely affected were entitled to request a hearing but "[b]efore any evidentiary hearing will be ordered, it must appear affirmatively that there is a genuine and substantial issue of fact requiring such a hearing." Unless the manufacturer "can show a reasonable likelihood that he is prepared to produce `substantial evidence' derived from adequate and well-controlled clinical investigations in support of his promotional claims, there is no basis for a hearing to receive evidence that would not in any event satisfy the legal requirement as to proof of effectiveness." The Regulations spelled all this out, and included a detailed description of what would be regarded as "adequate and well-controlled clinical investigations."

Acting on the basis of these Regulations, the Commissioner on January 29, 1970, issued an order repealing the regulations covering the six drugs and revoking all certificates issued thereunder. 35 F.R. 1234-35. It reviewed the history and characterized Pfizer's response of April 30 as containing "no new clinical data" and not providing "substantial evidence of the effectiveness" of the drugs. The order allowed 30 days for the filing of objections, stating reasonable grounds and requesting a hearing thereon. The statement of reasonable grounds was required to "identify the claimed errors in the NAS-NRC evaluation and the Administration's conclusions as to lack of `substantial evidence' of the effectiveness of the combination drugs" and was to "identify and provide a well-organized and full factual analysis of any adequate and well-controlled investigations the objector is prepared to prove in support of the objections as a basis on which it could reasonably be concluded that the combination drugs would have the effectiveness claimed for their intended uses." The order was to become effective within 40 days unless stayed by the Commissioner.

However, the September 19 Regulations had meanwhile been declared invalid in Pharmaceutical Manufacturers Ass'n v. Finch, 307 F.Supp. 858 (D.Del. 1970), for failure by the FDA to give the notice of proposed rule-making and opportunity for comment required by § 4 of the Administrative Procedure Act (APA), 5 U.S.C. § 553. Evidently considering discretion the better part of valor, the FDA on February 17, 1970, republished the September 19 Regulations as a proposal and requested comment thereon,5 35 F.R. 3073, meanwhile staying the effectiveness of the order that had been made with respect to Signemycin prior to the court decision.6 On May 8 the FDA promulgated the Regulations without substantial change,7 35 F.R. 7250. In a preliminary statement it dealt with the comments of the pharmaceutical manufacturers; we shall later have occasion to mention some parts of this.

On June 5, 1970, Pfizer filed its objections and request for a hearing. It attacked the NAS-NRC report on various grounds. One was that the report ignored a 1959 Italian study by Arachi and Therardi showing that the oleandomycin tetracycline combination had cured 34 out of 37 patients with chronic nonspecific urethritis which had resisted other antibiotics, including, in eleven cases, tetracycline. The objections cited three new clinical studies, allegedly well-controlled. One, by Drs. Thomas and Durchell, concerned a number of healthy hospital personnel shown by culture of the nose and throat to be carriers of staphylococcus strains. Of 42 patients evaluated, 21 were "cleared" of the strains by tetracycline. Thirteen of the failures were treated with triacetyloleandomycin, and eight were cleared. Two of the remaining five were treated with Signemycin and both were cleared. A second, by Drs. Cullen and Isenberg, also related to nasal carriers of staphylococci, in this instance prisoners, but divided the carriers into groups instead of treating them seriatim. The group treated with Signemycin showed better results than either group treated with a component alone. A third study, by Dr. Ente, compared the results of Signemycin and Ampicillin on two groups having serious throat infections; the two did about equally well.

The FDA, on July 14, 1970, entered the order repealing the Signemycin regulations here under review, 35 F.R. 11622. Faulting the various studies as failing to demonstrate that Signemycin was more efficacious than either of its components, for reasons discussed below, it held that Pfizer had failed to present "reasonable grounds" for a hearing. The FDA noted that these deficiencies had been discussed at a meeting on June 19 and by a later letter.

Pfizer begins its attack on the denial of an evidentiary hearing with an argument drawn from the new drug section, § 505(e). Under that section, the right to a hearing8 on the proposed withdrawal of approval is alleged to be absolute, save only for the power of the Secretary to suspend for an imminent hazard to public health, in which event an expedited hearing is required. Why then, asks Pfizer, when the FDA proposes withdrawal of approval for an antibiotic long on the market, should the manufacturer have to make any preliminary showing? A sufficient answer is the simple if not altogether satisfying one that Congress said so. When it brought antibiotics under a separate section of the statute in 1945, it chose not to copy the "after due notice and opportunity for hearing to the applicant" language of § 505(e), 52 Stat. 1053 (1938), but to demand as a prerequisite to a hearing that those filing objections to the Secretary's action state "reasonable grounds therefor." 59 Stat. 464. That the Secretary's duty to hold a hearing is conditioned on such a statement emerges clearly from the statute. Only after an interested party has (1) filed his objections to the Secretary's action, (2) specified the changes desired, (3) stated reasonable grounds therefor, and (4) requested a hearing on such objections, does the Secretary's duty "thereupon" to hold a public hearing arise. An objector failing to comply with any of these four requirements has simply failed to meet the statutory criteria on which the right to a hearing is predicated.9

Moreover, there was nothing irrational about this statutory scheme. Congress may well have thought the possibly lesser protection against withdrawal of approval of antibiotics was a fair trade for exemption from some of the long procedures incident to approval as a "new drug." Perhaps experience like that in 1941 and 1943 which is recorded in Atlas Powder Co. v. Ewing, 201 F.2d 347, 351 (3 Cir. 1952), cert. denied sub nom. Glyco Products Co. v. Federal Sec. Adm'r, 345 U.S. 923, 73 S.Ct. 781, 97 L. Ed. 1355 (1953), had caused disillusionment with trial-type hearings on scientific matters. Perhaps there was some other reason now lost in the mists of time, perhaps there was none.10 In any event we must give effect to the restriction on the right to a trial-type hearing in § 507(f), unless it denies due process. We see no reason to think it does. While a controversy over the efficacy of a particular drug raises an issue of "adjudicative facts," see 1 Davis, supra, § 7.02 at 413, we perceive no denial of due process in Congress' directing an agency that it need not accord a trial-type hearing unless the affected party shows in advance that there is something substantial to hear. See NLRB v. Joclin Mfg. Co., 314 F.2d 627, 633 (2 Cir. 1963).11

The serious issue is whether the FDA acted lawfully in holding that Pfizer's submissions did not constitute "reasonable grounds." Whatever meaning that phrase may have in other contexts, our inquiry here has been narrowed by the FDA's promulgation of regulations12 effectively limiting "reasonable grounds" to "adequate and well-controlled investigations."13 The regulations provide:

No regulation providing for the certification of any batch of any drug composed wholly or in part of any kind of penicillin, streptomycin, chlortetracycline, chloramphenicol, bacitracin, or any other antibiotic drug, or any derivative thereof, intended for use by man shall be promulgated and no existing regulation will be continued in effect unless it is established by substantial evidence that the drug will have such characteristics of identity, strength, quality, and purity necessary to adequately insure safety and efficacy of use. "Substantial evidence" has been defined by Congress to mean "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." This definition is made applicable to a number of antibiotic drugs by section 507(h) of the Act, and it is the test of efficacy that will be applied in promulgating, amending, or repealing regulations for the certification of all antibiotics under section 507(a) of the Act as well.

35 F.R. 7252-53. If this provision is valid and Pfizer failed to meet it, the case is ended. Pfizer challenges its validity as applied to the "old" antibiotics governed by § 507(f).

We start from the proposition that if an administrative agency vested with rule-making powers, such as are possessed by the FDA under §§ 507(a) and 701(a), has promulgated, in a rule-making proceeding pursuant to § 4 of the APA, regulations whereby it will refuse to hear applications not conforming to criteria reasonably related to the regulatory statute, its application of such regulations so as to deny a hearing violates no legal rights. United States v. Storer Broadcasting Co., 351 U.S. 192, 205, 76 S.Ct. 763, 100 L.Ed. 1081 (1956); FPC v. Texaco, Inc., 377 U.S. 33, 39, 84 S.Ct. 1105, 12 L.Ed.2d 112 (1964).14 Pfizer contends that even if this be so, the FDA's insistence on "adequate and well-controlled clinical investigations" as a condition to granting a hearing on the proposed withdrawal of approval of an "old" antibiotic drug does not accord with the statutory scheme. It points to the fact that the definition of "substantial evidence" as meaning "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," was expressly made applicable only to the approval of "new drugs" under § 505(d), the withdrawal of approval of such drugs under § 505(e), and the withdrawal of approval of an antibiotic, not named in the pre-1962 version of § 507, which had originally qualified as a new drug, § 507 (h). It places particular stress on the contrast between the express incorporation of the quoted definition in § 507(h) and the absence of similar cross-reference in § 507(f), and on the provision in § 507(e) that, with an exception not here applicable, no drug subject to § 507 "shall be deemed to be subject to any provision of § 505." It argues that in Upjohn Co. v. Finch, supra, 422 F.2d 944, the Sixth Circuit failed to take this into proper account.

While the contention has force, it does not squarely face the issue before this court — whether the FDA may insist that the "reasonable grounds" in § 507 (f) shall consist of "adequate and well-controlled investigations," not whether it must. We are not at all convinced that the FDA could not have issued regulations whereby nothing short of such investigations would constitute "reasonable grounds," even before the 1962 amendments. Pfizer has furnished us with no persuasive reason why the 1962 Congress should have wished to deny the FDA power to require "adequate and well-controlled investigations" as a condition to continued approval of an "old" antibiotic of questionable efficacy, if it chose to do so, although it granted such power with respect to the thousands of "new drugs" approved since 1938 and the antibiotics approved as new drugs prior to 1962. The closest attempt at showing this is Pfizer's assertion that the "old" antibiotics were of such generally recognized efficacy that Congress saw no need to subject them to "adequate and well-controlled investigations" to establish this. But Congress had recognized from the start that even the named antibiotics might turn out to be unsafe or inefficacious and had authorized the FDA to withdraw approval in that event. In effect Pfizer urges us to hold that because the 1962 Congress did not expressly require the FDA to insist on "adequate and well-controlled investigations" in a proceeding to withdraw approval of an "old" antibiotic, it therefore meant to prevent the agency from adopting a rule that would subject the "old" antibiotics to this test. But there is nothing logically compelling about such an inference and we do not regard Congress' failure to incorporate in § 507(f) something equivalent to the final clause of § 507(h) as suggesting it. For we would hardly be justified in finding, on so scant a basis, a Congressional mandate requiring the FDA to employ a double standard of efficacy in evaluating antibiotics, all of which were now subject to § 507.15

There is much less force in other arguments that Pfizer presents on this score. It says that if the FDA were to require "adequate and well-controlled investigations" of all the "old" antibiotics, there would not be enough investigators for that purpose. But there is nothing to suggest the FDA would do anything of the sort; for example, it has neither taken nor proposed to take any initiative against any of the components of Signemycin. An even worse situation than that envisioned by Pfizer would exist if the FDA were to move against all the "new drugs" approved since 1938, yet Congress plainly authorized the agency to insist on "adequate and well-controlled investigations" to establish the efficacy of any such drug that it had chosen to question. Pfizer says also that adequate and well-controlled investigations are but one of several scientifically valid methods of evaluating a drug's effectiveness, that general clinical experience has its place, and that disapproval resting solely on the lack of such investigations is not supported by substantial evidence. But, apart from the fact that Congress evidently thought otherwise, at least with respect to those drugs subject to § 505 (d) and § 505(h), the point is sufficiently answered by the FDA's observation in promulgating the May 8 Regulations which we quote in the margin.16 Moreover, the FDA did not here rely solely on the lack of adequate and well-controlled investigations but on the references to the literature cited by the NAS-NRC panel and other such references discussed in the FDA's final order.

Pfizer points also to the extreme hardship that would result if the FDA were suddenly to confront a manufacturer of an "old" antibiotic, who had no reason to suppose the efficacy of his product was in question, with the need of producing an adequate and well-controlled investigation within 30 days. This is partially answered by the statement of FDA policy made in promulgating the May 8 Regulations which we quote below.17 In any event the point is inapplicable to this case. Pfizer knew since the spring of 1969, if not before, that the efficacy of Signemycin was being seriously questioned for lack of evidence that it lived up to its labeling as more effective than either of its components. On April 30 of that year Pfizer proposed a program of what would appear to be "adequate and well-controlled" in vitro, animal and clinical studies that would have fully met FDA requirements. Its present problem arises from its apparently unexplained failure to do anything to pursue them. If it had initiated such studies and needed a bit more time than general agency policy, see fn. 17, allowed and the FDA had then cracked down, we would have a different case.

Little need be said with respect to Pfizer's argument that even if adequate and well-controlled investigations could lawfully be required as a condition to an evidentiary hearing, it met the test. While such a determination is peculiarly within the FDA's expertise and we would be reluctant to intrude into medical matters we do not truly understand, it is apparent even to us laymen that the FDA had a reasonable basis for considering that Pfizer's submissions did not comply with the requirement of the statute and the Regulation. We quote in the margin the passages of its final order relevant to the Italian, Thomas-Burchell and Cullen-Isenberg studies.18,19,20 The Ente study was sufficiently disposed of by the obvious point that all it could possibly demonstrate was that Signemycin was as effective as Ampicillin; this had nothing to do with the issue whether Signemycin had effects not attainable as well or better by either of its components alone. The contrast between what Pfizer produced and what it had undertaken to produce in the way of adequate and well-controlled investigations was also significant in this context.21

Pfizer's final contention is that since § 507(a) speaks only of "efficacy of use," the FDA is not authorized to withdraw approval of an "old" antibiotic drug, not shown to be without efficacy simply because it lacks efficacy "as a fixed combination." Here again such problem as there is arises from the fact that the 1962 Congress did not choose to incorporate in the portions of § 507 dealing with "old" antibiotics the more elaborate language used in the amendment of the "new drug" section, § 505 (e), and the provision, § 507(h), dealing with antibiotics originally approved as "new drugs." Section 505(e) permits withdrawal of approval of a "new drug" if on the basis of new information with respect to the drug,22 there is lack of substantial evidence "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," and § 507(h) is to the same effect with respect to an antibiotic originally certified as a new drug. A good case could certainly be made that, quite apart from this, the "efficacy" of a drug is necessarily related to the use recommended. Be that as it may, it was within the FDA's rule-making power to flesh out the word "efficacy" in § 507(a) so as to apply these more precise requirements, which Congress imposed on new drugs and on antibiotics originally certified as new drugs, to the old antibiotics as well. While the FDA was thus justified in requiring proof "that the response to the combination is greater than to either component alone," 35 F.R. 11625, it found also, that "[t]he unnecessary use of an antibiotic, in a fixed combination dosage form, when one or the other component would suffice, favors the emergence in the environment of resistant microorganisms," 35 F.R. 11624, and, "that both components of the fixed combination drug have adverse effects which must be taken into account," 35 F.R. 11625. Such findings support a conclusion of lack of "safety and efficacy of use" even without the gloss placed on that phrase of § 507(a) by the May 8 Regulations.

We therefore deny Pfizer's petition to review. Since the company may desire to pursue the matter further and there is no evidence that Signemycin constitutes a serious safety threat, we will continue the stay of the FDA's order until 30 days from the filing of this opinion and, if within that period an application for a stay of longer duration is made to the Supreme Court, until determination thereof.