Congress adopted amendments to the FDA Act in 1962. See Drug Amendments of 1962, Pub. L. No. 87-781, 76 Stat. 780. Under the 1962 amendments manufacturers were required to show that a new drug product was effective as well as safe for its intended use. 21 U.S.C. § 355(a) and (b). A "new drug" was defined as "any drug . . . the composition of which is such that such drug is not generally recognized, among experts qualified by scientific training and experience to evaluate the safety and effectiveness of drugs, as safe and effective for use under the conditions prescribed, recommended, or suggested in the labeling thereof. . . ." 21 U.S.C. § 321(p).
Under the 1962 amendments a manufacturer seeking approval of an NDA was required to submit "substantial evidence" of the product's safety and effectiveness. The statute defined "substantial evidence" as "evidence consisting of adequate and well-controlled investigations, including clinical investigations . . . on the basis of which it could fairly and responsibly be concluded . . . that the drug will have the effect it purports . . . to have under the conditions of use prescribed . . . in [its] labeling. . . ." 21 U.S.C. § 355(d). Sandoz FWC products are exempt from the FDA's premarket clearance procedures only if they are "generally recognized" as safe and effective for their intended uses. 21 U.S.C. § 321(p).
The 1962 amendments required the FDA to evaluate the effectiveness of all drug products, not grandfathered, which had been placed on the market since 1938. To do this, the FDA enlisted the services of the National Academy of Sciences - National Research Council (NAS-NRC), which, after investigation, made recommendations to the FDA regarding the effectiveness of therapeutic classes of products. As the FDA reviewed the NAS-NRC recommendations, it published drug efficacy study implementation (DESI) notices in the Federal Register which rated products as to their effectiveness for specific labeled indications. Products that were found to be effective were still regarded as new drugs and manufacturers were required to supplement their NDAs to conform to conditions imposed by the DESI notices. In the period shortly after adoption of the 1962 amendments the FDA issued informal opinions with respect to some drug products that they found were not new drugs. However, in 1968 the FDA adopted a regulation revoking all such opinions. 21 C.F.R. § 310.100(d). Thereafter virtually all prescription drug products which had come on the market since 1938 would be regarded as a new drug requiring either a full or, in certain narrowly defined cases, an abbreviated new drug application (ANDA).
In 1976 the FDA published a Compliance Policy Guide (CPG) which established enforcement priorities for proceedings against DESI-related drugs that were being marketed without approved applications. The sixth category included unapproved combination drugs that are related to drugs which had been found effective in the DESI program.
As noted above before a product can be exempted from the statutory, new drug preclearance procedures, it must be "generally recognized" by qualified experts as safe and effective for its intended uses. 21 U.S.C. § 321(p). The "general recognition" requirement does not involve the actual safety or effectiveness of the product. Rather, it is the product's reputation in the scientific community that is relevant. Further, the experts' opinion as to general recognition must be based upon well-controlled, clinical studies which are published; the opinions of experts may not be based upon uncontrolled data or upon their own personal experiences. Weinberger v. Hynson, Westcott & Dunning, Inc., 412 U.S. 609, 619, 37 L. Ed. 2d 207, 93 S. Ct. 2469 (1973). The general recognition of a product's safety and effectiveness must be documented by at least the same quality and quantum of evidence that would suffice to obtain FDA's approval of the product in the first instance. Weinberger v. Bentex Pharmaceuticals, Inc., 412 U.S. 645, 652, 37 L. Ed. 2d 235, 93 S. Ct. 2488 (1973).
Certain FDA regulations are particularly pertinent to the present case. One is contained in 21 C.F.R. § 314.126. It sets forth in considerable detail the essential components of an adequate and well-controlled investigation. A study must meet these criteria in order to constitute substantial evidence of a product's effectiveness. Weinberger v. Hynson, Westcott & Dunning, Inc., supra, 412 U.S. at 617-20; Cooper Laboratories, Inc. v. FDA, 163 U.S. App. D.C. 212, 501 F.2d 772 (D.C. Cir. 1974).
Another particularly pertinent regulation is set forth in 21 C.F.R. § 300.50 and provides in part:
(a) Two or more drugs may be combined in a single dosage form when each component makes a contribution to the claimed effects and the dosage of each component (amount, frequency, duration) is such that the combination is safe and effective for a significant patient population requiring such concurrent therapy as defined in the labeling for the drug.
These, then, are the statutory and regulatory provisions which must be applied in determining the critical factual issue in the case, namely, whether FWC No. 1 and FWC No. 2 are generally recognized on the basis of well-controlled, published clinical studies to be safe and effective for their intended use.
III. The FWC Products
Fiorinal is an analgesic marketed by Sandoz. Originally it contained the peripheral analgesic aspirin (200 mg), phenacetin (130 mg), the analgesic adjuvant caffeine (40 mg) and the barbiturate butalbital (50 mg). In 1983 pursuant to an FDA directive pertaining to products containing phenacetin, the phenacetin was removed and the quantity of aspirin was increased to 325 mg.
The FWC product consists of the Fiorinal ingredients plus varying amounts of codeine. FWC No. 1 contains 7.5 mg codeine; FWC No. 2 contains 15 mg codeine; FWC No. 3 (which is not a subject of seizure in the present case) contains 30 mg codeine. When phenacetin was removed from Fiorinal it was also removed from the FWC products. FWC is a widely prescribed combination product useful for severe tension headaches, post-surgical pain, and other forms of pain.
In 1962 Sandoz received from the FDA an informal written opinion that the FWC products were not new drugs. In July 1963 Sandoz began marketing FWC. In 1968 the FDA revoked all its informal opinions. 21 C.F.R. § 310.100(d).
In response to a 1973 DESI notice Sandoz submitted data to support its NDA for Fiorinal, but did not submit any data for its FWC products. On November 15, 1977, in response to this submission, the FDA published a DESI follow-up notice which stated that Fiorinal "has been evaluated as effective for . . . tension. . . headache" and that "such drugs are regarded as new drugs (21 U.S.C. § 321(p)."
In 1977 and 1978 Sandoz filed ANDAs with the FDA for its FWC products, the difference between an abbreviated and full NDA being that the abbreviated application need not contain full reports of clinical studies to prove the product's safety and effectiveness. 21 C.F.R. § 314.55(a). On both occasions the FDA rejected the applications because it had not made a specific finding that such short-form applications were appropriate for the FWC products and because the approved products, plain Fiorinal and Trigesic with Codeine, upon which Sandoz sought to "piggy back" its ANDA applications for the FWC products, contained different ingredients from the FWC products. FWC contains codeine; Fiorinal does not. FWC contains butalbital; Trigesic with Codeine does not.
Sandoz did not appeal these decisions but in 1978 filed a Notice of Claimed Investigational Exemption for a New Drug (IND). The IND has been maintained since that date.
In March 1978 FDA officials informed Sandoz that the FWC products were new drugs, requiring a full NDA. Sandoz was also informed, among other things, of the studies that would be required for approval of an NDA and that FWC No. 1 containing 7.5 mg codeine would probably not show effectiveness.
A June 24, 1980 FDA DESI order stated that no evidence had been presented to show that two products containing butalbital, acetaminophen and other analgesics were effective. On November 15, 1984 FDA sent Sandoz a Regulatory Letter reminding it that the FWC products were regarded as new drugs, that the FWC products were identical, similar or related to the barbiturate-analgesic combination drug product found to lack substantial evidence of effectiveness in the June 24, 1980 DESI order, and that continued marketing of the FWC products without an approved NDA would violate 21 U.S.A. § 355.
By letters dated December 5, 1984 and January 11, 1985, Sandoz responded to the November 15, 1984 regulatory letter and asked that the agency not take immediate regulatory action against FWC and further requested that the FDA allow FWC products to remain on the market pending review and approval of a soon-to-be submitted NDA. On February 1, 1985 Sandoz submitted an NDA for the FWC products.
On February 20, 1985 Sandoz submitted a third request that the FDA accept an ANDA for the FWC products, stating among other things that "Fiorinal with Codeine is identical to . . . Fiorinal with the exception of a single active ingredient, codeine phosphate, . . . the ingredients of Fiorinal with Codeine are in the same therapeutic class as the ingredients in Fiorinal, [and] there are ample data to establish that the addition of codeine to . . . Fiorinal can be expected to have the same effect as Fiorinal."
On April 16, 1985 the FDA advised Sandoz that a preliminary review of its NDA for the FWC products indicated that the application was not approvable.
On April 26, 1985 Sandoz wrote the FDA and advised that it was taking the position that the FWC products were not new drugs. It cited a long history of safe and effective use and referred to "an extensive series of clinical investigations which have given rise to a general recognition of the safety and effectiveness of the product and the contribution thereto of each of its active components."
The April 26 letter was accompanied by the declarations of Dr. Donald Dalessio and Dr. Neil Raskin. Dr. Dalessio based an opinion about the safety and effectiveness of FWC products upon his experience in prescribing the drug for a 15 year period and upon published and unpublished material which Sandoz provided him. He referred specifically to the published studies of MacDonald and Ehrenreich and of Madore and Chiricosta and to a Sandoz multicenter study which had been submitted to the FDA in support of Sandoz' NDA.
On May 16, 1985 Sandoz submitted to the FDA the declaration of Dr. Arnold P. Friedman attesting to the general recognition of the safety and effectiveness of FWC products. None of the three declarations which Sandoz submitted differentiated between the three FWC formulations.
On October 3, 1986 the complaint in this action was filed and on January 12, 1987 the United States Marshal seized the FWC products.
IV. The Published Clinical Studies
The critical issue in this case is whether Sandoz' FWC products are generally recognized among qualified experts as safe and effective. Safety does not appear to be a contested issue, and thus the present inquiry is confined to the FWC products' effectiveness and to the effectiveness of each of their constituent drugs. The opinions of the experts as to general recognition must be based upon well-controlled clinical studies which are published. The parties in this proceeding refer to six such studies in support of and in opposition to the motion for summary judgment.
A. Madore, P. and Chiricosta, A., "Analgesic Effect of Fiorinal with Codeine," Anesthesia and Analgesia, Vol. 46, No. 4 July-Aug. 1967 : This study used FWC No. 3 which contained 30 mg of codeine. The FWC No. 3 formulation differed from that which is presently marketed in that it contained 130 mg phenacetin and 200 mg aspirin. Today's FWC No. 3 (like today's FWC No. 1 and No. 2) contains no phenacetin and 325 mg aspirin.
In the Madore and Chiricosta study 41 patients who had undergone various types of operations and complained of acute pain episodes were given, in double-blind, single dose, random-sequence fashion, one of three treatments: (a) placebo, (b) one capsule FWC No. 3, or (c) two capsules FWC No. 3. Thereafter, at hourly intervals and for 6 hours each patient's pain relief was evaluated. The authors concluded that both quantities of active drugs were superior to placebo at hours 2 through 4, and that two capsules of FWC No. 3 were superior to one capsule of the same product only at the first hour.
B. MacDonald, C. and Ehrenreich, K.A., "A Double-Blind Comparison of Two Analgesics with Placebo Control," Canadian Med. Assoc. J. 95:1072-75, Nov. 19, 1966 : This study compared two codeine combination analgesics prescribed in the 1960s. One was FWC No. 2 as then formulated which contained butalbital (50 mg), phenacetin (130 mg), caffeine (40 mg), aspirin (200 mg) and codeine (16 mg). The other product, APC with Codeine, contained all of those ingredients except butalbital. One purpose of the study was to assess the contribution of butalbital.
In this study pain relief was measured in 83 post-surgical patients who reported a total of 206 episodes of severe or moderate pain. Patients estimated their level of post-operative pain just prior to treatment and their pain relief approximately one hour after treatment. Three treatments were assigned randomly to 206 severe or moderate pain episodes.
FWC No. 2 (having butalbital) provided greater pain relief than APC with Codeine (having no butalbital) in moderate and severe pain episodes and both were more effective than placebo.
C. Desjardins, P.J., Cooper, S.A., and Finizio, T., "Efficacy of Low Dose Combination Analgesics: Acetaminophen/Codeine, Aspirin/Butalbital/Caffeine/Codeine, and Placebo in Oral Surgery Pain," Anesth. Prog. 1986, May-June, 33(3):143-6 : This was a double-blind, randomized, single-dose study performed to compare the efficacy and safety of two combination analgesic products to placebo. The products were Tylenol with Codeine No. 3 containing acetaminophen (300 mg) and codeine (30 mg) and FWC No. 3 containing aspirin (325 mg), butalbital (50 mg), caffeine (40 mg) and codeine (30 mg). The study showed that FWC No. 3 was significantly more effective than placebo for total pain relief, peak relief and global evaluation. FWC No. 3 was numerically superior to Tylenol with Codeine No. 3 for every measure of analgesic efficacy but the difference did not achieve statistical significance. The report stated, "whether the slight improvement in analgesia is therapeutically exploitable or whether it can be attributable to a caffeine or butalbital effect cannot be answered in the present study design."
D. Forbes, J.A., Jones, K.F., Smith, W.K., and Gongloff, C.M., "Analgesic Effect of an Aspirin-codeine-butalbital-caffeine Combination and an Acetaminophen-codeine Combination in Postoperative Oral Surgery Pain," Pharmacotherapy 1986, Sept.-Oct. 6(5):240-47 : This study was a randomized, double-blind, placebo-controlled comparison between current formulas FWC No. 2 and Tylenol with Codeine No. 3. The efficacy of FWC No. 2 was compared to Tylenol with Codeine No. 3 and placebo in outpatients who had moderate or severe pain after the surgical removal of impacted third molars.
The report stated, among other things:
. . . Although the rationale for many combinations, for example, aspirin and codeine, is readily apparent, other combinations have less surface validity. One such combination is that of aspirin 325 mg, caffeine 40 mg, codeine phosphate 15 mg with butalbital 50 mg, a barbituate. . . . The butalbital constituent of this combination is presumed to decrease anxiety and act as a muscle relaxant.
MacDonald et al demonstrated the statistical superiority of an aspirin-phenacetin-caffeine-codeine-butalbital combination to the combination without butalbital in the treatment of postoperative pain. The study, however, did not attempt to measure the effect, if any, of the medications on anxiety or relaxation. Although the study was conducted double-blind and employed a concomitant placebo control group, it was unusual in that (1) patients could participate more than once, (2) only one measure of effect, pain relief, was used, and (3) patients were evaluated only once, that is, at one time point, after the administration of the study medication.
The Forbes study leads to the conclusions that each medication was significantly superior to placebo for measures of analgesia and relaxation and that although the butalbital containing combination provided consistently greater analgesia, the differences between the active medications were not statistically significant. The report further stated that "one can only speculate as to the relative contribution of the constituents of these combinations since a factorial study was not conducted."
E. Friedman, A.P., et al., "Assessment of Fiorinal with Codeine in the Treatment of Tension Headache," Clinical Therapeutics, Vol. 8, No. 6, 703-21, 1986 : This was a double-blind, randomized multicenter study in which 39 patients with specific symptoms of tension headache (pain, psychic tension, and muscle-stiffness) and 15 patients with any two of those symptoms were given one of four treatments: the early version of FWC No. 3 (containing phenacetin and 200 mg aspirin); the early version of plain Fiorinal (also containing phenacetin and 200 mg aspirin); codeine phosphate (30 mg); or placebo. Before taking one of these treatments and at.5, 1, 2, 3 and 4 hours following treatment patients rated six items on a four point scale. The physicians also made global evaluations of the effectiveness of treatment.
FWC No. 3 was found to be generally significantly more effective than placebo or Fiorinal alone in improving all patient-evaluated items between the second and fourth hours after administration. FWC No. 3 was also found to be generally significantly better than codeine alone with respect to pain severity, pain relief, the ability to perform daily activities and average pathology. The three variables rated by the physicians also were found to be generally reduced significantly more with FWC No. 3 than with placebo or Fiorinal alone.
F. Dar-shong Hwang, Ph.D., et al., "Fiorinal with Codeine in the Management of Tension Headache: Impact of Placebo Response," Clinical Therapeutics, Vol. 9, No. 2, 1987 : This study analyzed a subset of 67 patients admitted under a protocol identical to the Friedman, et al., study. It compared the old Fiorinal (containing phenacetin and 200 mg aspirin), the old FWC No. 3 (containing phenacetin and 200 mg aspirin), codeine and placebo. The report noted that "in the original analysis of the study data, no distinction was apparent between patient response to Fiorinal with Codeine and the response to the individual components, a finding that appeared to conflict with the results of a similar earlier study." The authors of the report attributed the discrepancy to a high placebo response in the later study. To adjust for this factor they identified a subset of less anxious patients with mild to moderate pain severity who, according to the authors, were least likely to respond to placebo.
Based on the data from this less anxious subset of patients the authors found that FWC No. 3 was significantly better than placebo in improving patients' self-ratings of various symptoms of tension headache at 0.5, 1, 2, 3, and 4 hours after ingestion of the study medication and that FWC No. 3 was also consistently superior to Fiorinal alone and codeine alone in improving patients' self-evaluation items, and differences between FWC No. 3 and its components were generally of statistical or borderline significance during the last half of the study.
The investigators' assessments of the effect of treatment on the three principal variables in tension headache (headache pain, psychic tension, and muscle contraction of the head, neck and shoulders) at the final patient visit also showed FWC No. 3 to be significantly superior to placebo and consistently superior to Fiorinal and codeine. The superiority of FWC No. 3 over Fiorinal alone achieved borderline significance for headache pain and psychic tension.
V. Summary Judgment Standards
In order to prevail on a motion for summary judgment, a moving party must establish that "there is no genuine issue as to any material fact and that [it is] entitled to a judgment as a matter of law." Fed.R.Civ.P. 56(c). The moving party has the initial burden of informing the court of the basis of its motion and identifying those portions of the record which it believes demonstrate the absence of a genuine issue of material fact. Celotex Corp. v. Catrett, 477 U.S. 317, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986).
Once the moving party has carried its burden under Rule 56, "its opponent must do more than simply show that there is some metaphysical doubt as to the material facts." Matsushita Electric Industrial Co., Ltd. v. Zenith Radio Corp., 475 U.S. 574, 106 S. Ct. 1348, 1356, 89 L. Ed. 2d 538 (1986). A plaintiff opposing a summary judgment motion must set forth specific facts showing a genuine issue of material fact. "The inferences to be drawn from the underlying facts . . . must be viewed in the light most favorable to the party opposing the motion." Id. 106 S. Ct. at 1357. However, the Matsushita Court explained that in evaluating the evidence presented, "where the record taken as a whole could not lead a rational trier of fact to find for the nonmoving party, there is no 'genuine issue for trial.'" 106 S. Ct. at 1356.
In addition, should it appear from the affidavits of a party opposing the motion that the party cannot for reasons stated present by affidavit facts essential to justify the party's opposition, the court may refuse the application for summary judgment or may order a continuance to permit affidavits to be obtained or discovery taken. Fed.R.Civ.P. 56(f).
VI. The FDA's Demonstration
The FDC Act provides that no new drug may be introduced into interstate commerce unless and until the FDA has approved an application for that drug. 21 U.S.C. §§ 355(a), 331(a). Before products such as FWC No. 1 and FWC No. 2 can be exempted from the statutory new drug preclearance procedures they must be "generally recognized" by qualified experts as safe and effective for their intended uses. 21 U.S.C. § 321(p). The "general recognition" requirement does not involve the actual safety or effectiveness of the product. Rather, it is the product's reputation in the scientific community that is relevant.
An expert's opinion as to general recognition must be based upon published well-controlled clinical studies meeting the requirements of 21 C.F.R. § 314.126 and may not be based on uncontrolled data and personal experience. Further, when a product contains more than one active ingredient, the contribution of each ingredient in the product must be scientifically demonstrated.
It is FDA's contention that there are no well-controlled clinical studies on the basis of which it can be concluded that FWC No. 1 and FWC No. 2 are effective for their labeled indications and there are none which show that each ingredient of the FWC products contributes to the effectiveness of these two products. This being so, FDA urges, it is entitled to summary judgment.
To establish the absence of a factual issue, the FDA has submitted the declaration of six highly qualified experts, William Forrest, M.D., Thomas Kantor, M.D., Michael Weintraub, M.D Michael Lockshin, M.D., Richard Black, M.D. and Raymond Dionne, D.D.S., Ph.D. The FDA asked each of them to give his opinion regarding the general recognition of the safety and effectiveness of FWC No. 1 and FWC No. 2 and whether each component of the two drugs contributes to their effectiveness. Their factual statements and opinions were generally to the same effect. They can be summarized as follows:
The doctors described the contents of the two FWC products and referred to their labels which set forth the conditions for which the products may be used. They noted their familiarity with the requirements of 21 C.F.R. § 314.126 and 21 C.F.R. § 300.50.
As the manufacturer of a combination drug product Sandoz must show that each ingredient in FWC No. 1 and FWC No. 2 contributes to the claimed effects. Typical of the opinion of the six doctors is that of Dr. Forrest:
8.a. . . . To establish that FWC No. 1 complies with [§ 300.50 governing combination drug products], Sandoz must show that the addition of codeine (7.5 mg.) will significantly increase the effectiveness of the combination of aspirin (325 mg.), caffeine (40 mg.), and butalbital (50 mg.) to treat mild to moderate pain as indicated in this product's label. This must be done by conducting adequate and well-controlled studies in which the effect of the combination of aspirin, caffeine, and butalbital is compared with the effect of the combination of aspirin, caffeine, codeine, and butalbital (i.e., FWC No. 1) in patients who have mild to moderate pain. The addition of 7.5 mg. codeine is justified only if there is significantly greater relief of pain in the latter group.
b. Because FWC No. 1 also contains butalbital and caffeine, Sandoz must further show that the addition of each of these ingredients, as formulated in FWC No. 1, significantly increases the effectiveness of the FWC No. 1 combination. These showings must be made by comparing the pain relief achieved with FWC No. 1 against the relief achieved with additional control groups which eliminate each of the ingredients whose effect is to be measured (e.g., to measure the contribution of butalbital, a group of patients must receive aspirin, caffeine and codeine; to measure the contribution of caffeine a group of patients must receive aspirin, codeine, and butalbital). (In my view, it is necessary to separately establish the contribution of caffeine in this product, even though there is reasonable evidence which shows that caffeine in larger doses (65 mg.) enhances the effectiveness of clinical doses of aspirin. I am not aware of any adequate and well-controlled studies which show that caffeine increases the effectiveness of aspirin when codeine and butalbital are present in the product.) The studies outlined in paragraphs 8a and 8b could be combined.
c. Because the foregoing reasoning applies with equal force to FWC No. 2, each of the showings described in paragraphs 8a and 8b above would have to be made with the FWC product which contains 15 mg. codeine phosphate.