Edison Pharmaceutical Company, Inc., Petitioner, v. Food and Drug Administration, Department of Health,education and Welfare, Andcaspar W. Weinberger,secretary, Department of Health,education and Welfare,respondents, 513 F.2d 1063 (D.C. Cir. 1975)

Before WILKEY, Circuit Judge, NICHOLS,^*  Judge, United States Court of Claims, and GASCH,^**  United States District Court Judge for the District of Columbia.

Opinion for the Court filed by District Judge GASCH.

GASCH, District Judge.

This is an appeal from a final order¹  of the Commissioner of Food and Drugs (Commissioner) that the new drug application (NDA) for Cothyrobal (NDA 16-865) submitted by petitioner Edison Pharmaceutical Company, Inc., is not approvable because of a lack of substantial evidence of its safety and effectiveness.

The Commissioner rejected petitioner's NDA without holding a hearing. He held that because petitioner had submitted no "adequate and well-controlled" studies in support of its NDA, the application was deficient on its face. Therefore there could be no genuine and substantial issue of fact outstanding which required a hearing.

Petitioner contends that it met the threshold evidentiary requirement to entitle it to a hearing on the approvability of its NDA before final action was taken by the Commissioner. It asks this Court to order the Commissioner to hold a hearing on its NDA for Cothyrobal.

The Court grants petitioner's request. We set aside the final order of the Commissioner and remand this case to the Food and Drug Administration (FDA) for a hearing on the approvability of petitioner's NDA 16-865.

The Federal Food, Drug, and Cosmetic Act of 1938,²  as amended³  (the Act), and the Regulations issued pursuant to the Act,⁴  establish, inter alia, a product pre-clearance system under which a new drug⁵  may not be marketed interstate until it has been approved by the Commissioner as "safe and effective" for its intended use. The burden is on the sponsor of the drug to demonstrate its safety and effectiveness.

Section 505 of the Act⁶  and pertinent Regulations⁷  set forth the procedure by which a drug manufacturer can have a new drug approved for marketing in interstate commerce.

Section 505(b)⁸  requires an NDA to be submitted to FDA. This application must include reports of investigations regarding safety and effectiveness, a statement of the composition and components of the drug, a description of the methods used in and the facilities and controls used for, the manufacture, processing, and packing of such drug, samples of the drug, and specimens of labeling.

Section 505(d)⁹  sets forth the grounds on which the Commissioner can refuse to approve an NDA. One of these grounds, on which Cothyrobal's NDA was determined to be deficient,¹⁰  is that the sponsor has failed to submit "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."

The section further defines "substantial evidence" asevidence 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.

A Regulation issued by FDA¹¹  defines the form such investigations should take. This form includes a "plan or protocol" setting forth the objective of the study, an adequate method of selecting appropriate subjects, an explanation of the methods of observation and steps taken to minimize bias, a provision for comparison by one of four "recognized" methods of the results of treatment or diagnosis with a control, and a summary of the methods of analysis, including any appropriate statistical method. Provision is made for the waiver of any or all of these requirements, if justified, upon petition for waiver.

Three of the four "recognized" methods of comparison for testing the effectiveness of a new drug consist of comparing the results of patients treated with the new drug with other patients (1) not receiving any treatment, or (2) receiving placebos, or (3) receiving non-drug therapy. The fourth method, used in testing the effectiveness of a new drug for diseases of high and predictable mortality, involves comparing the progress of the disease in patients treated with the new drug with "prior experience historically derived from the adequately documented natural history of the disease or condition . . .."

The Regulation also states that "uncontrolled studies or partially controlled studies are not acceptable as the sole basis for the approval of claims of effectiveness."¹² 

On May 19, 1969, Edison submitted to FDA an NDA for its drug, Cothyrobal,¹³  which is a combination of thyroid extract (levothyroxine) and vitamin B12 (cyanocobalamin) and recommended for the treatment of hypercholesterolemia, hypothyroidism, and for patients who became thyro-toxic with other types of thyroid medication. Petitioner claims that by combining thyroid extract with vitamin B12, the toxic effect of large doses of levothyroxine is overcome.

By letter dated December 1, 1969, the Commissioner denied the NDA on the ground that the information submitted in its support was inadequate under Section 505(b).¹⁴ 

On December 24, 1969, Dr. Murray Israel, the inventor of Cothyrobal, and petitioner Edison Pharmaceutical Company filed an antitrust suit in the United States District Court for the District of Columbia¹⁵  against Baxter Laboratories, Travenol Laboratories (a subsidiary of Baxter), Dr. David Kritchevsky, a consultant to FDA who was associated with Baxter, and Dr. Marion Finkel, an employee of FDA. The complaint alleged that defendants had conspired to keep Cothyrobal off the market and out of competition with Choloxin, a similar drug sold by Baxter and Travenol pursuant to an approved NDA, by influencing FDA to deny fair consideration to the NDA for Cothyrobal filed by plaintiffs. The District Court dismissed the complaint, holding that this case was exempt from the antitrust laws¹⁶  and that plaintiffs had not exhausted their administrative remedies.

On appeal¹⁷  this Court held in a prior opinion that the complaint, on its face, stated a cause of action,¹⁸  and we remanded all issues to the District Court. We directed that Court to remand the question of safety and efficacy of Cothyrobal to the FDA since it has primary jurisdiction over the approval of drugs for interstate sale. But we instructed the Court to retain jurisdiction over the plaintiffs' cause of action so that

if . . . plaintiffs do not then obtain full and fair consideration by the FDA as to the safety and efficacy of their drug Cothyrobal for interstate sale, they may obtain a full hearing in the District Court on all their allegations.¹⁹ 

This Court made it clear that "(t)he essential is that the plaintiffs have a full and fair presentation of all their contentions before an impartial tribunal."²⁰ 

Thereafter petitioner requested that its NDA for Cothyrobal be reactivated. It submitted no additional information in support of its application. After review by FDA personnel unconnected with the initial review of petitioner's NDA in 1969,²¹  the application was again found not approvable by the Commissioner.

Edison requested an opportunity to file its NDA over protest. The effect of this was to have the NDA re-examined by still another group of FDA personnel.²²  Again it was found not approvable.

On June 28, 1973, the Commissioner published in the Federal Register²³  a Notice of Opportunity for Hearing.²⁴  This Notice stated the Commissioner's intention to deny petitioner's NDA, the grounds for the decision, and petitioner's opportunity for a hearing "on the question of whether such application is approvable" before the final order would be issued. The Notice warned that "a request for a hearing may not rest upon mere allegations or denials, but must set forth specific facts showing that a genuine and substantial issue of fact requires a hearing."²⁵  Without such showing, the Commissioner may deny an NDA without a hearing.²⁶ 

In response to this Notice a written appearance and request for a hearing was submitted by Edward "Whitey" Ford, a member of the Board of Directors of the Vascular Research Foundation on behalf of himself and over two hundred other individuals, together with letters of a testimonial nature relating to Cothyrobal from each of these individuals.

A written appearance and request for a hearing was also submitted by Edison. The request was accompanied by no new data. It consisted of medical and legal arguments as to why data previously submitted met the requirements for approval of NDA 16-865.

Upon review the Commissioner decided and stated in his final order of October 26, 1973, that the requests contained no new data which would in any way correct any of the deficiencies which had been noted in the Federal Register.

Accordingly the Commissioner concluded that there was a lack of substantial evidence that Cothyrobal has the effect it purports or is represented to have for its intended use; that the legal arguments were insubstantial; and that petitioner had failed to set forth specific facts showing that there was a genuine and substantial issue of fact in dispute requiring a hearing. This appeal followed.

Petitioner is not challenging the validity of the procedure for approving NDA's. Nor is it challenging FDA's summary judgment procedure whereby an NDA may be rejected without a hearing if it does not contain the information required by Section 505(b). Instead petitioner contends that its submission more than meets what it characterizes as the "threshold evidentiary requirement" established by Hynson, Westcott and Dunning v. Richardson, 461 F.2d 215 (4th Cir. 1972), affirmed as modified on another issue²⁷  sub nom. Weinberger v. Hynson, Westcott & Dunning, 412 U.S. 609, 93 S. Ct. 2469, 37 L. Ed. 2d 207 (1973), thereby entitling it to a hearing on the approvability of its NDA.²⁸ 

Hynson involved the drug, Lutrexin,²⁹  which had been approved by FDA in 1952 when the Act required only a showing of safety. After the Act was amended in 1962 to require a showing of effectiveness as well as safety, Hynson's NDA was withdrawn³⁰  without a hearing because the clinical studies submitted to demonstrate Lutrexin's effectiveness were not "adequate and well-controlled."³¹  Thus the Commissioner held in Hynson, as he has held here, that since the clinical studies were not "adequate and well-controlled," substantial evidence of effectiveness was lacking, and the NDA was deficient on its face. Therefore there could be no genuine issue of fact in dispute requiring a hearing before denying the NDA.

The Fourth Circuit reversed the Commissioner, however, holding that the standard of proof required to secure the right to a hearing is less than that required to prove effectiveness.

Neither due process nor the Administrative Practice (sic) Act permits an arbitrary denial in any case where it can be fairly said there are "genuine and substantial issues of fact" in dispute.

Only if it can be fairly said that the clinical tests and medical studies and investigations submitted by the applicant, if credited and accepted (emphasis added), will not support a finding that they provide "substantial evidence" of effectiveness was it proper for the Commissioner to deny the appellant a hearing before entering a final order of withdrawal.³² 

The Court then noted that "no qualified expert has given an opinion that Lutrexin is ineffective for the uses intended," and that the Commissioner, in pointing out the deficiencies in the investigations submitted by Hynson, "did not impugn the competency or qualifications of the scientists and medical experts whose investigations were cited by the appellant in support of its claim."³³  Therefore

(a)ssuming that all the objections by the Commissioner to these clinical studies, conducted as they were by competent medical authorities, may have some validity (emphasis added), they do not justify a final conclusion, made ex parte, without a hearing, that it "clearly appears" that there is no genuine issue of fact on the effectiveness of Lutrexin, which is the test under the Commissioner's own regulation for denial of a hearing; at most they merely create a genuine question of fact to be resolved at a hearing upon proper evidence.³⁴ 

On certiorari, the Supreme Court upheld FDA's summary judgment procedure for NDA's.³⁵  It more precisely defined those circumstances in which it might be determined that it "clearly appears" that there is no genuine issue of fact in dispute as being occasions

where it is apparent at the threshold that the applicant has not tendered any evidence which on its face meets the statutory standards as particularized by the regulations. . . . We cannot impute to Congress the design of requiring, nor does due process demand, a hearing when it appears conclusively from the applicant's "pleadings" that the application cannot succeed.³⁶ 

The Court then set forth the scope of review of an order denying a hearing:

In reviewing an order of the Commissioner denying a hearing, a court of appeals must determine whether the Commissioner's findings accurately reflect the study in question and if they do, whether the deficiencies he finds conclusively render the study inadequate or uncontrolled in light of the pertinent regulations.³⁷ 

But then the Court, without any discussion of the studies submitted by Hynson or the question of how accurately the Commissioner had discussed the studies in his findings, affirmed the Fourth Circuit's decision that Hynson's submission was sufficient to warrant a hearing.³⁸  Thus the Supreme Court, like the Fourth Circuit, said that even granting the validity of the Commissioner's objections to Hynson's investigations, Hynson's NDA would still survive "summary judgment" and be entitled to a hearing.

In the case at bar, petitioner submitted five clinical studies in support of its NDA for Cothyrobal. Like the submission in Hynson, these studies were deemed deficient by the Commissioner because none was "adequate and well-controlled" within the meaning of the Act.

Three of these studies the Commissioner characterized as anecdotal, and therefore insufficient by themselves to demonstrate effectiveness. At best they might provide corroboration for well-controlled studies.³⁹  The two remaining studies were found to contain several major deficiencies. Neither study verified any of its subjects as hypothyroid, even though one of the problems for which Cothyrobal was recommended was hypothyroidism; subjects received oral thyroid as well as injections of Cothyrobal, thereby making the evaluation of the effects of Cothyrobal difficult; and there was no control group employed. In the Commissioner's words,

There are no well-controlled studies using blind and double-blind cross-over and randomnization techniques or any other kind of control specified in 21 CFR 130.12(a) (5) (ii), so that neither the clinical nor statistical significance of the reported results can be evaluated.⁴⁰ 

Petitioner contended in his request for a hearing as it has contended before us that such studies comparing the use of Cothyrobal, levothyroxine, and cyanocobalamin in separate groups of subjects would involve a "risk of death to patients" already taking the thyroid extract, levothyroxine. Thus even though its studies may have been technically deficient under Section 505(d), they were "as scientifically sound and objective as it was humanly possible to make."⁴¹ 

In his final order the Commissioner dismissed petitioner's claim that double-blind studies might be dangerous to subjects by saying that "textbook medicine" demonstrates "that the L-thyroxine dosage contained in Co-Thyro-Bal is not ordinarily toxic."⁴² 

Petitioner does not deny that the amount of levothyroxine in Cothyrobal could be safely administered to a control group. But it points out that most patients taking injections of Cothyrobal are also taking oral thyroid. Therefore the amount of levothyroxine which a control group would take would equal the sum of levothyroxine in Cothyrobal and the oral thyroid already being taken.⁴³  It is this amount of thyroid extract which petitioner claims would be toxic.

This illustrates the inability of the parties to come to grips with the essential issue in this case. If the petitioner had been given a hearing, it is at least more likely that it could have made its point more clearly. At least it could have been given the attention it deserves.

The Commissioner did not consider the toxicity of this combined amount of levothyroxine in his final order. Instead he attempted to circumvent the issue by saying that

it should be stressed . . . that whether such toxicity exists or not and whether a study of the L-thyroxine alone would be dangerous or not is in part irrelevant, since the Commissioner finds that not even an adequate and well-controlled study comparing Co-Thyro-Bal itself with a placebo has been performed.⁴⁴ 

From this statement, repeated four times in FDA's brief⁴⁵  and at oral argument, we inferred that if Edison had conducted a test comparing Cothyrobal with a placebo, and the test had demonstrated effectiveness, it would have been sufficient to at least entitle petitioner to a hearing. For then it could not be said, under Hynson, that Edison had not "tendered any evidence which on its face (met) the statutory standards."⁴⁶  Upon questioning, however, government counsel informed us that such a test would not have been enough.

Thus it appears to us that the factual question of whether the double-blind tests comparing levothyroxine and Cothyrobal are too dangerous to perform is still "relevant." Indeed, we deem this a sufficiently material fact in dispute to require an evidentiary hearing on Edison's NDA before the Commissioner may issue a final order. The Commissioner cannot have it both ways. Either the double-blind tests comparing Cothyrobal with a placebo would be sufficient so that the question of whether levothyroxine⁴⁷  alone is toxic is irrelevant, or such tests would not be sufficient, in which case whether tests comparing Cothyrobal and levothyroxine are "humanly possible" becomes not only relevant, but also an issue of material fact in dispute.

Therefore, for the aforementioned reasons, we hold that the threshold question whether it is "humanly possible" to acquire evidence which meets the statutory standards (as interpreted by FDA regulations) requires a full evidentiary hearing.

We also find it appropriate to require the Commissioner to hold a full evidentiary hearing on all relevant issues relating to the approvability of petitioner's application, whichever way he decides the threshold issue. In so doing we are not establishing a new requirement for a full hearing whenever only a threshold issue is in dispute. We are mindful of the fact, noticed by the Supreme Court in Hynson,⁴⁸  that if FDA were required automatically to hold a hearing on every NDA it would not have time to carry out its statutory mandate.

We make an exception here only because of the facts of this case. Petitioner first filed NDA 16-865 over six years ago. In the interim its application has been denied on three separate occasions without an opportunity for hearing despite direction to the contrary from this Court.⁴⁹  By requiring the Commissioner to hold a full evidentiary hearing on all relevant issues relating to petitioner's NDA, even if he decides against petitioner on the threshold issue, we enable this Court on review finally to bring this proceeding to a close.

We are re-enforced in our decision by the fact that between May 8, 1970, when FDA's summary judgment regulations went into effect, and April 4, 1973, when the government submitted its reply brief in Hynson, the Commissioner had not granted a single request for a hearing on an NDA.⁵⁰  Even after the Supreme Court's decision in Hynson, which was cited by petitioner in its request for a hearing,⁵¹  the Commissioner still denied Edison's request for a hearing. In our eyes this failure to grant a hearing to any applicant⁵²  casts doubt upon the good faith we would ordinarily impute without question to a decision of the Commissioner.