Turning the tables on drug companies; exposing deficiencies in FDA regulation.

Pharmaceutical manufacturers decry the purported burdens imposed by federal regulation when it suits their interests. Yet they sing a different tune when defending products liability suits. As defendants, they seek the protective womb of Food and Drug Administration (FDA) regulation. They argue that agency approval of their drug and its labeling precludes their product from being regarded as defective or their conduct as tortious.

Most courts have rejected the claim that compliance with FDA regulations preempts a finding of liability for negligence, defective design, and inadequate warnings under state tort law. FDA regulation of prescription drugs is "generally viewed as setting minimum standards, both as to design and warning."(1)

Even where courts have rejected federal preemption as a defense to a pharmaceutical products liability suit, a drug company will not hesitate to urge deference to the purportedly preeminent knowledge of FDA, the agency that previously determined that the drug's benefits outweighed its risks; justified its marketing; and approved the warnings, precautions, contraindications, and other safety-related information contained in the drug's labeling. Defendants commonly maintain that they have met or exceeded all FDA requirements and have not deviated from any regulation, statute, or guidelines to which they were subject.

Courts that do not honor the preemption defense nonetheless often permit the trier of fact to take defendants' alleged compliance with FDA standards into account.(2) Agency approval may be accorded whatever weight the trier of fact determines it deserves,(3) and may be disregarded where the jury sees fit.(4)

Implicit in the "FDA approval defense" is the assumption that agency regulation protects the public from the dangers of marketed drugs. Congressional oversight of FDA's performance, however, challenges the validity of this assumption. Since enactment of the Drug Amendments of 1962 - which overhauled drug regulatory laws - shortcomings in FDA regulation of new drugs have been repeatedly exposed. So where a drug company has argued that it has complied with government standards, it behooves the plaintiff's attorney to know about instances of serious deficiencies in agency regulation.

Before approving new drugs, the agency has not always known of significant adverse information regarding their use, even when this information was in its own files. Witness the case of the antiarthritis drug Oraflex (benoxaprofen), withdrawn by manufacturer Eli Lilly & Co. a few months after it was approved in 1982 because of a large number of serious and sometimes fatal liver and kidney reactions associated with its use. Because of a filing backlog in the documents room of the FDA division responsible for reviewing the safety of Oraflex, the agency was unaware of four clinical trial reports of serious reactions to the drug - including hepatitis and jaundice - that Lilly had submitted to the agency's investigational new drug (IND) file.(5)

After reviewing the matter, a House committee concluded that "had such reports been noted, it is unlikely that FDA would have originally approved labeling for the drug which made no mention of liver disease and which confined Oraflex-associated liver reactions to |liver function test abnormalities' which were |usually transient.'"(6)

The committee also found that Lilly had prominently reported to the FDA six cases of drug-associated kidney disease before the drug's approval.(7) Yet Lilly proposed, and FDA approved, labeling that denied any "evidence ... of renal toxicity in the clinical studies."(8) As of a congressional hearing on August 3, 1982 - three and a half months after Oraflex was approved - FDA still could not confirm Lilly's assertion that it had reported serious drug-associated adverse effects before the drug's approval.(9)

FDA also did not know more than five months after approving Oraflex and seven weeks after its withdrawal from the market whether Lilly had reported any Oraflex-associated deaths before approval. The House Committee on Government Operations concluded that this signaled that "the agency had not thoroughly examined the sponsor's IND and [new drug application] submissions before approving the drug for marketing."(10)

Glaring omissions in the agency's discharge of its statutory responsibilities have a long history. On January 13, 1956, FDA approved McNeil Laboratories' new drug Flexin (zoxazolamine) for the relief of muscle spasms in conditions like musculoskeletal disorders and neurological diseases.(11) On October 13, 1961, the agency suspended the Flexin new drug application after receiving a large number of reports of severe and sometimes fatal hepatitis associated with the drug's use.(12) As with Oraflex more than two decades later, FDA officials were unaware of the drug's association with hepatitis and jaundice.

Lilly's failure to report serious and sometimes fatal liver reactions outside the United States contributed to FDA's assumption that Oraflex had not been implicated in the onset of hepatitis and jaundice.(13) While FDA's ignorance of Flexin-associated hepatitis and jaundice was similarly attributable, in part, to McNeil's failure to report adverse liver reactions,(14) the agency conceded - as it was forced to do years later with Oraflex - that it had overlooked several reports of subreactions in its own files that had been submitted by McNeil.

For example, the agency neglected to note the manufacturer's acknowledgment in June 1958 of "occasional reports of patients who have developed jaundice."(15) FDA also overlooked a statement included in a supplemental new drug application submitted for a Flexin-containing drug on September 29, 1958, that "during the past 3 1/2 years there had been a total of 32 reports to McNeil suggesting that the administration of the drug [Flexin] may have been associated with development of hepatitis with jaundice."(16)

Then-FDA Commissioner George Larrick admitted in congressional testimony that this information strongly suggesting the relationship between drug-induced hepatitis and the drug had been received by the new drug branch but had not been reviewed.(17) FDA's failure to consider reports of drug-associated liver injury was ascribed to a failure to examine documents comprising a backlog of submissions received from regulated manufacturers.(18)

That the Oraflex experience so strongly resonates with echoes from the agency's regulation of Flexin is particularly ironic in that these drugs had a similar molecular structure and were two of only three such drugs ever approved by FDA.(19) In fact, the House Committee on Government Operations concluded that "Oraflex's chemical similarity to Flexin ... should have alerted FDA to its potential liver toxicity."(20) The Flexin and Oraflex sagas underscore the enduring truth of George Santayana's admonition that those who cannot remember

In the wake of revelations regarding deficiencies in the pre-approval review of Oraflex, Congress chronicled oversights in the agency's pre-market assessment of the nonsteroidal anti-inflammatory drug Zomax (zomepirac sodium). The originally approved Zomax labeling did not mention the drug's reported association with allergic/anaphylactoid reactions. Later, a congressional committee found that before approval FDA had received but failed to notice clinical trial reports of such reactions.(21)

The agency was also unfamilliar with important data it had received from the manufacturer of Versed (midazolam hydrochloride), a drug approved to induce conscious sedation. The information suggested the originally approved doses were excessive.(22) Until FDA required the manufacturer to reduce the dosage levels many months after the drug's approval, a number of Versed patients suffered serious and sometimes fatal respiratory depression and cardio-respiratory arrest. The agency eventually attributed these episodes to an originally approved and recommended dose that was "too high."(23)

Medical Illiteracy

In clinical pharmacology, the published medical literature is recognized as the most basic and readily accessible information about the safety of new drugs. Unfortunately, FDA has repeatedly been shown to be unaware of important aspects of the toxicity of new pharmacological agents discussed in medical publications. For example, congressional hearings conducted as early as 1964 exposed FDA's lack of knowledge of important information linking the use of Orabilex (bunamiodyl sodium), a drug enabling X-ray determination of gallbladder disease, to renal failure.(24) Similarly, before approving the antidepressant Merital (nomifensine maleate), FDA was unfamiliar with important published papers suggesting that drug's toxicity to the human immune system.(25)

The Versed case is also exemplary. FDA approved conscious sedation doses of Versed of 0.1 to 0.15 mg/kg and, if necessary, up to 0.2 mg/kg.(26) The House Government Operations Committee found that, before approving Versed at these doses, the agency was unaware of studies published in prominent medical journals demonstrating that these does were excessive.(27) The agency was equally unfamiliar with published studies indicating the efficacy of Versed at conscious sedation doses substantially below those that had been approved.(28)

In originally approving conscious sedation doses for Versed comparable to those it had approved for intravenous Valium (diazepam), FDA was unaware of a Swedish study published before the approval of Versed strongly suggesting that Versed was three times as potent as Valium when used intravenously as an anesthetic. Therefore, Versed should have been approved at substantially lower conscious sedation doses than those recommended for Valium.(29) Incredibly, as late as a House hearing on May 10, 1988, more than two years after FDA had approved Versed for marketing, agency regulators were still not aware of this study.(30)

Legally Questionable Approval

A products liability suit is not a forum for adjudicating the legality of FDA decision making. However, if faced with claims that the agency only approves drugs where clinical evidence complies with all legal requirements, counsel may want to point out that FDA in the past has endangered patients' health and safety by approving drugs which, by its own account, had not been shown to be safe within the meaning of pre-market approval requirements.

The history of Triazure (azarabine) is illustrative. FDA approved it for treating is on February 28, 1975. In August 1976, the agency issued a press release warning patients taking Triazure to discontinue its use immediately because of its capacity to produce fatal blood clots.(31)

Before approval, the agency had determined that the law required it to disapprove the Triazure new drug application unless additional testing illuminating the apparent risk of drug-induced blood clotting was completed. The agency advised the manufacturer, Calibiochem, that the application was inadequate because the drug had not been shown to be safe within the meaning of the law, and that further testing was necessary.(32)

Even though the manufacturer refused to conduct additional testing, FDA approved Triazure. The results were tragic. When Triazure was removed from the market, it had been linked to serious injury and even death resulting from blood clotting.(33)

There are faint echoes of this experience in FDA's more recent approval of Tambocor (flecainide acetate), a drug intended to treat cardiac arrhythmias or irregular heart rhythms. By letter dated August 29, 1985, FDA advised the manufacturer, Riker Laboratories, Inc., that Tambocor should be approved only for life-threatening arrhythmias because of its propensity to provoke new or more serious arrhythmias. FDA further advised the company of the need to undertake additional study of Tambocor in patients with less severe arrhythmias before it could be recommended for use by such patients.(34)

Nonetheless, on October 31, 1985 - only three months later - FDA approved Tambocor for patients with less severe as well as life-threatening arrhythmias.(35) After approval, clinical trials sponsored by the National Heart, Lung, and Blood Institute revealed a two-and-a-half-fold increase in mortality among patients treated with Tambocor or Enkaid (encainide HCl) - another antiarrthythmic agent - compared to patients receiving a placebo. As a result, FDA requested labeling changes in April 1989 recommending against use of Tambocor and Enkaid in less severe arrhythmias.(36)

FDA's lack of awareness of important drug-safety data is not confined to pre-market reports of adverse effects. For example, important information revealing Versed to be far more likely to oversedate patients - and thus expose them to a serious risk of respiratory depression - than injectable Valium was contained in a clinical trial report submitted to the agency on September 26, 1986.(37) That FDA is not always aware of potentially important safety data may sometimes simply reflect limitations on its already overworked and understaffed personnel. Thus, "buried as it was in a voluminous" submission to the Versed IND file made after FDA approval, a congressional committee observed of the September 1986 report that it "not surprisingly, apparently went unreviewed by FDA."(38)

Although the agency requires that certain post-market reports of drug-associated adverse reactions be submitted to it in timely fashion,(39) it has been publicly exposed as not knowing about large numbers of these reports that have been forwarded to it. For example, an FDA employee testified at a 1983 congressional hearing that he was surprised when McNeil Pharmaceutical told FDA at a February 11, 1983, meeting that it had submitted to FDA 908 reports of allergic/anaphylactoid reactions associated with the use of Zomax since its approval. An FDA memorandum of a February 28, 1983, meeting with McNeil indicated that shortly before the drug was withdrawn from the market, the agency's computerized tracking system had 270 reports of these reactions.(40)

These revelations confirmed what had earlier been found by the General Accounting Office (GAO). In a 1982 report, GAO found that FDA's Division of Drug Experience - the agency unit then responsible for tracking post-market reports of drug-associated adverse effects - took an average of 3.3 months and sometimes more than a year to enter such reports into its computerized monitoring system.(41) Before Zomax was taken FDA also failed to analyze data it had that strongly intimated that Zomax was associated with a higher incidence of anaphylactoid reactions than other drugs in its class.(42)

Following what it thought was a temporary marketing suspension of Zomax, McNeil proposed remarketing the drug with labeling emphasizing that a "majority" of life-threatening and fatal anaphylactic reactions occurred in "individuals without a prior allergic history."(43) That such reactions could unforeseeably strike users without an allergic drug history clearly heightened the drug's dangers. Congressional investigators found that while Zomax was still being marketed, FDA neglected to analyze data it had collected suggesting that these patients were at highest risk of experiencing such reactions.(44)

As a result, FDA permitted McNeil, before the drug's removal from the market, to drop from the package insert a proposed warning about the risk of anaphylactoid reaction to patients who had previously manifested no allergic reactions to Zomax or other drugs in its class.(45) This is noteworthy, since the Government Accounting Office concluded that the agency's review procedures may be inadequate for identifying drug-induced risks before approval.

In an April 1990 report,(46) GAO found that of the 198 drugs approved between 1976 and 1985 for which data were available, 51.5 percent had serious post-approval risks. GAO implied in its recommendations that some serious risks "discovered" following approval could possibly have been identified and addressed during pre-market review.(47)

Disarming the Defense

Large pharmaceutical concerns that must regularly deal with the FDA acquire expertise in agency regulation. Defendant's counsel is therefore well-positioned to argue that the agency's allegedly painstaking regulatory process forecloses a finding that the company marketed a defective product or engaged in tortious conduct.

In one case, a defendant argued that FDA certification of its vaccine should be conclusive evidence of "non-negligence per se." The court rejected this:

FDA certification represents only the FDA's opinion, albeit an informed one, of the safety and efficacy of the drug. Regrettably, drugs occasionally prove not so safe as the FDA first believed.(48)

Accordingly, the court held that "FDA certification of a drug is evidence but not conclusive evidence of the drug manufacturer's reasonableness; the trier of fact may assign FDA approval the weight it deserves."(49)

That a drug eventually shows itself as dearly more hazardous than FDA knew or assumed may sometimes be explained by the agency's inattention or lack of knowledge due to limitations or deficiencies in its scientific-review procedures.

If the defendant introduces evidence that it complied with government standards, plaintiff's counsel can cite authority that FDA approval only constitutes compliance with minimum standards, which do not preempt liability otherwise accruing under state tort law. Counsel should also try to introduce or rely on case histories of shortcomings in FDA's review process that have placed the health, safety, and lives of unwitting patients at risk.

(Notes)

(1) Graham v. Wyeth Lab., 666 F. Supp. 1483, 1991 (D. Kan. 1987) (emphasis omitted). (2) For example, one court has held that "compliance with FDA requirements, though admissible to demonstrate lack of negligence, is not conclusive on this issue, just as violation of FDA requirements is evidence, but not conclusive evidence, of negligence." MacDonald v. Ortho Pharmaceutical Corp., 475 N.E.2d 65, 70-71 (Mass.), cert. denied, 114 S. Ct. 304 (1993). (3) Graham, 666 F. Supp. 1483, 1491. (4) Tobin v. Astra Pharmaceutical Prods., Inc., 993 F.2d 528, 538 (6th Cir. 1993). (5) All of the Oraflex clinical trial patients who contracted liver disease had also developed kidney disease. Fourteenth Report, HOUSE COMM. ON GOVERNMENT OPERATIONS, DEFICIENCIES ON FDA'S REGULATION OF THE NEW DRUG ORAFLEX, H.R. REP. NO. 511, 98th Cong., 1st Sess. 9-12 (1983) [hereafter Oraflex Rep.]. (6) Id. at 9. (7) In addition to four reports of combined kidney and liver disease, two other clinical trial reports of Oraflex-associated kidney disorders were submitted to Oraflex IND. Id. at 10. (8) Id. (9) Id. at 12. (10) Id. at 11-12. (11) See Drug Safety: Hearings Before a Subcomm. of the Comm. on Government Operation, 88th Cong., 2d Sess., pt. 2, at 625 (1964). (12) Id. at 582. By July 13, 1961, the manufacturer reported to FDA that it had received a total of 54 reports of hepatitis in patients, many of which involved death. Id. at 580. (13) Daniel W. Sigelman, Drug Safety Information from Foreign Countries, TRIAL, Mar. 1992, at 20. (14) For example, then-FDA Commissioner George Larrick testified that the agency, later learned that McNeil had received "a report of fatal hepatitis in a patient on this drug even before we allowed the [new drug] application to become effective, but that report was not relayed to us until 1961." Drug Safety, supra note 11. (15) On June 11, 1958, a supplemental NDA for Flexin was submitted for the drug's use as a uricosuric agent to treat gout. The quoted statement was made in an accompanying brochure. Id. at 572-73. (16) Id. at 574. (17) Id. (18) Id. (19) Specifically, the Oraflex and Flexin molecules contained a "benzoxaxole" ring system as their nucleus. Oraflex Rep., supra note 5, at 9. (20) Id. (21) Thirty-first Report, HOUSE COMM. ON GOVERNMENT OPERATIONS, FDA'S REGULATION OF ZOMAX, H.R. REP. NO. 584, 98th Cong., 1st Sess. 15 (1983) [hereafter Zomax Rep.l. (22) Seventy-first Report, HOUSE COMM. ON GOVERNMENT OPERATIONS, FDA'S DEFICIENT REGULATION OF THE NEW DRUG VERSED, H.R. REP. NO. 1086, 100th Cong., 2d Sess. 21-22 (1988) [hereafter Versed Rep.]. (23) Id. at 8. (24) See Drug Safety, supra note 11, at 685, 785-86, 795. (25) Fifteenth Report, HOUSE COMM. ON GOVERNMENT OPERATIONS, FDA'S REGULATION OF THE NEW DRUG MERITAL, H.R. REP. NO. 206, 100th Cong., 1st Sess. 37 (1987). (26) Versed Rep., supra note 22, at 10-11. (27) Id. at 10-11, 20. (28) Id. at 19-20. (29) Id. at 20-21. (30) Id. at 20. (31) FDA's Regulation of the New Drug "Triazure": Hearings Before a Subcomm. of the Comm. on Government Operations, 94th Cong., 2d Sess. 1 (1976). (32) Id. at 4. (33) Id. at 1. (34) Problems with FDA's Regulation of the Antiarrhythmic Drugs Tambocor and Enkaid: Hearings Before the Human Resources Intergovernmental Relations, Subcomm. of the Comm. on Government Operations, 102d Cong., 1st Sess. 144-45 (1991). (35) Id. at 125, 130. (36) Id. at 189. (37) Versed Rep., supra note 22, at 33. (38) Id. (39) See 21 C.F.R. [sections] 314.80 (1993). (40) Zomax Rep., supra note 21, at 11-12. (41) Id. at 12. (42) Id. at 13. (43) Id. at 16. (44) Id. at 16-17. (45) Id. at 19. (46) U.S. GEN. ACCOUNTING OFFICE, FDA DRUG REVIEW: POSTAPPROVAL RISKS, 1976-1985, at 57 (1990). (47) Id. at 57-58. (48) Toner v. Lederle Lab., 732 P.2d 297,311 n.12 Idaho 1987), cert. denied, 485 U.S. 942 (1988). (49) Id.

Daniel W. Sigelman practices with the Atlanta law firm of Pope, McGlamry, Kilpatrick & Morrison. From 1981 to 1988, he conducted oversight of FDA regulation of new drugs as counsel to the House Subcommittee on Human Resources and Intergovernmental Relations.