ASPIRIN ALCOHOL-SPECIFIC LABEL WARNING REJECTION URGED

ASPIRIN ALCOHOL-SPECIFIC LABEL WARNING REJECTION URGED by the Aspirin Foundation of America in Oct. 26 comments to FDA based on the absence of "reliable data showing that the combination of heavy drinking and aspirin use increases the risk of [gastrointestinal] bleeding." Responding to a recommendation by FDA's Nonprescription Drugs and Arthritis Advisory Committees to require an alcohol-specific label warning, AFA President Thomas Bryant, MD, maintained that "a review of the [meeting] record demonstrates that the committee recommendation is not supported by the underlying data." The two advisory committees voted 12-2 on Sept. 8 in favor of a warning for aspirin products regarding an increased risk of gastrointestinal bleeding for alcohol drinkers ("The Tan Sheet" Sept. 13, p. 11). In reaching its recommendation, the panel reviewed a package of data submitted by Tylenol (acetaminophen) manufacturer McNeil on gastrointestinal bleeding risks associated with aspirin, ibuprofen and naproxen. The Nonprescription Drugs Advisory Committee recommended in June that acetaminophen labeling contain a warning statement for liver toxicity directed at heavy drinkers ("The Tan Sheet" July 5, p. 1). The Aspirin Foundation asserted that "the committee's concerns and its recommendations [regarding aspirin] were produced not by actual evidence, but by inaccurate characterizations of available data by McNeil." According to AFA, McNeil "distorted" the data in its submissions to FDA and those "distortions had a significant impact on the committee's deliberations." The recommendations of the committee, AFA concluded, "were based on concerns over the large number of studies [in the McNeil submission] and a desire to err on the side of warning" even though "the actual evidence . . . failed to show any increased risk." AFA maintained that the data submissions by McNeil "selectively revealed only incomplete portions of study results and distorted the conclusions that can be drawn" from the studies conducted by the company. AFA also contended that McNeil's consultants did not provide the raw data for its submissions that would have allowed FDA to validate the multi-variate analyses used by McNeil to show increased risk with aspirin. In addition, AFA asserted that McNeil "frequently mischaracterized the conclusions" of published studies presented to the committee and "projected that thousands of additional UGI bleeds and hundreds of additional deaths would occur as a result of even a modest switch from acetaminophen to aspirin." To support its contention, AFA included two appendices to its letter to FDA. The first appendix summarizes comments made by the committee presenters of each study in the McNeil submission as well as comments made by an FDA epidemiologist, Raymond Alderfer, PhD. In a presentation to the panel, Alderfer reported that FDA does "not think that, the studies presented [by McNeil] were entirely conclusive" and said that the "magnitude" of bleeding from an aspirin/alcohol interaction is "uncertain" ("The Tan Sheet" Sept. 13, p. 14). The second appendix cites what AFA says are "serious errors and mischaracterizations contained in the McNeil submission." "By the sheer volume of its repeated submissions, McNeil appears to have convinced the committee that there must be a risk, despite the failure of well-controlled studies to demonstrate one," AFA maintained. According to AFA, the transcript of the meeting "reveals that the committee was influenced by such a concern and that it recommended a warning in spite of its own study-by-study rejection of the McNeil data." AFA also suggested that the advisory committee "was laboring under a misunderstanding of the current [aspirin] warning" for GI bleeding under the internal analgesic tentative final monograph. "Most of the data submitted by McNeil," AFA contended, "related to UGI bleeding by persons who had existing gastrointestinal disease for whom the TFM warning already warns against aspirin use." In urging FDA to reject the committee recommendation, AFA maintained that the data presented to the agency "contains no reliable evidence that would support the conclusion that an alcohol-specific warning is needed for aspirin." AFA added that "there is no evidence of an interaction between alcohol and aspirin"" and, "if anything," the data indicate "that alcohol may reduce any risk attributable to aspirin." The Aspirin Foundation also questioned whether FDA's acceptance of the committee's recommendation might set a precedent that could undermine future committee deliberations. "If FDA accepts the committee's recommendation," AFA asserted, "it may be inviting repeated abuse of the advisory committee process." AFA suggested that "interested parties will envision a process whereby companies can obtain label warning requirements for their competitors' products by submitting volumes of flawed or meaningless data to create the "impression that there is some danger there.'"