Correspondence
Clopidogrel and Thrombotic Thrombocytopenic Purpura
N Engl J Med 2000; 343:1191-1194October 19, 2000
- Article
To the Editor:
Clopidogrel is indicated for the reduction of atherosclerotic events (myocardial infarction, stroke, and death from vascular causes) in patients with atherosclerosis documented by recent stroke, recent myocardial infarction, or established peripheral arterial disease. This indication is based on the results of a clinical trial, involving 19,185 patients, in which clopidogrel was compared with aspirin in patients at risk for ischemic events. The study's steering committee estimated that clopidogrel would prevent about 24 major clinical events for every 1000 patients treated for one year.1
Clopidogrel was approved by the Food and Drug Administration (FDA) in November 1997 and has been used in more than 3 million patients worldwide. Bennett et al. (June 15 issue)2 describe 11 suspected cases of clopidogrel-associated thrombotic thrombocytopenic purpura. Information on these cases was obtained through solicited or spontaneous reporting or from published reports. Several of these suspected cases involve concomitant conditions, medications previously reported to be associated with thrombotic thrombocytopenic purpura, or both, and a clear causal relation with the use of clopidogrel could not be established in some of the cases. Nevertheless, Sanofi–Synthelabo and Bristol-Myers Squibb have included all suspected cases of thrombotic thrombocytopenic purpura in any estimates of incidence.
Sanofi–Synthelabo and Bristol-Myers Squibb had reported these cases to worldwide regulatory authorities before they were reported in the Journal, and on the basis of these cases, the FDA has approved new labeling for clopidogrel, which includes information on thrombotic thrombocytopenic purpura.
The report by Bennett et al. highlights the importance of post-marketing safety reporting in the detection of rare adverse events. Both Sanofi–Synthelabo and Bristol-Myers Squibb are committed to continued surveillance of these events and are evaluating the best methods for gaining a better understanding of thrombotic thrombocytopenic purpura and the incidence of this rare event.
2 ReferencesDominique Sallière, M.D.
Sanofi–Synthelabo, Paris 75031, FranceKenneth B. Kassler-Taub, M.D.
Bristol-Myers Squibb Pharmaceutical Research Institute, Princeton, NJ 08543-54001
CAPRIE Steering Committee
CrossRef | Web of Science | Medline2
Bennett CL ,Connors JM ,Carwile JM , et al. Thrombotic thrombocytopenic purpura associated with clopidogrel. N Engl J Med 2000;342:1773-1777
Full Text | Web of Science | Medline
To the Editor:
We write to offer minor corrections of the remarks by Bennett et al. and by Wood, in his accompanying editorial,1 about monitoring the safety of marketed drugs, as well as to describe the FDA's role in post-marketing surveillance of safety and to express enthusiasm for a comprehensive program of post-marketing safety surveillance. Within the FDA's Center for Drug Evaluation and Research, the Office of Postmarketing Drug Risk Assessment is responsible for monitoring and analysis of the approximately 300,000 reports of adverse events received annually. Approximately 6 percent of these reports come directly from consumers and health professionals through the MedWatch reporting program. Safety reports from all sources form the basis for the FDA's risk-management strategies, including product relabeling, notifications of health care professionals, drug withdrawals, and post-marketing drug studies and surveillance programs.
In the light of the post-marketing experience with ticlopidine, the Center for Drug Evaluation and Research was vigilant in its surveillance for thrombotic thrombocytopenic purpura and other adverse events associated with clopidogrel that may represent class effects of antiplatelet drugs. Ongoing surveillance of clopidogrel was uninformative until a cluster of reports from Bennett and others were received. When our review showed that the reported cases of thrombotic thrombocytopenic purpura were temporally associated with the use of clopidogrel and exceeded the expected numbers, the FDA promptly acted to warn practitioners by helping revise the product label so that it included a prominent warning.
We agree that strategies for active surveillance merit further development and funding. Adverse events that are frequently related to drugs, such as thrombotic thrombocytopenic purpura, may be particularly appropriate for active surveillance if there is a limited number of medical centers offering uniquely effective treatment, if patients generally survive long enough to receive treatment at such centers, and if the background rate of occurrence of the event is low. The FDA has not used independent reporting networks because of delays in reporting, difficulties in acquiring accurate drug histories, potential biases in the information obtained, and the large incremental expense associated with such strategies.
We agree with Wood that funding of safety surveillance beyond spontaneous reports is necessary, and we welcome initiatives to increase the quality and quantity of data on drug safety. We believe the FDA must have a central role in the assessment of drug safety. The FDA alone has the statutory and regulatory authority to mandate drug-safety reporting and to control the labeling and marketing of prescription drugs.
1 ReferencesAnne E. Trontell, M.D., M.P.H.
Peter K. Honig, M.D., M.P.H.
Food and Drug Administration, Rockville, MD 208571
Wood AJJ . Thrombotic thrombocytopenic purpura and clopidogrel -- a need for new approaches to drug safety. N Engl J Med 2000;342:1824-1826
Full Text | Web of Science | Medline
To the Editor:
If the incidence of idiopathic thrombotic thrombocytopenic purpura is 3.7 cases per year per million persons, as Bennett et al. state, this amounts to 1 case per 270,000. Bennett et al. described 11 patients with an illness indistinguishable from thrombotic thrombocytopenic purpura, in a population of 3 million or more patients exposed to clopidogrel. Thus, the incidence of the condition among patients who took clopidogrel was virtually the same as 1 per 270,000. Although most of the clopidogrel-associated cases occurred within two weeks after the initiation of treatment, I wonder whether the authors have for the most part reported cases of naturally occurring thrombotic thrombocytopenic purpura in patients who happen also to have been given clopidogrel.
Jeff Z. Brooker, M.D.
Providence Hospital, Columbia, SC 29204To the Editor:
Bennett and colleagues have aroused unneeded concern among the general public and clinicians. A comprehensive assessment of the safety of all new drugs requires continual post-marketing surveillance, and clopidogrel is not an exception.
The association between the use of clopidogrel and the occurrence of thrombotic thrombocytopenic purpura appears to be weak and coincidental in many of the 11 reported cases. For example, Patient 10 had been taking clopidogrel for almost one year before thrombotic thrombocytopenic purpura developed, and he had three recurrences up to seven months after the discontinuation of clopidogrel. Thrombotic thrombocytopenic purpura developed within seven days of exposure to clopidogrel in 5 of the 11 patients, including 1 of 2 patients in whom von Willebrand factor–cleaving protease activity was undetectable and in whom IgG inhibitors of the protease were present in plasma samples. Thus, in these patients, thrombotic thrombocytopenic purpura was probably caused by an antibody-mediated mechanism, but the period of exposure to clopidogrel was too short for this mechanism.
The authors also point out that most of the patients were receiving other medications concomitantly and that the results of additional laboratory studies were not available to rule out other causes of thrombotic thrombocytopenic purpura, facts that further weaken the association between the use of clopidogrel and the occurrence of thrombotic thrombocytopenic purpura. Patients 1 and 4 are good examples. Patient 1 had a recurrence of thrombotic thrombocytopenic purpura after taking atorvastatin for 14 days. Patient 4 took clopidogrel for only 3 days but was taking other medications such as atorvastatin for another 21 days or more before thrombotic thrombocytopenic purpura developed. In addition, the 9 percent mortality rate in this group of patients was lower than that among patients with idiopathic or ticlopidine-associated thrombotic thrombocytopenic purpura.
The report by Bennett et al. should not affect the routine use of clopidogrel as a safer alternative to ticlopidine in suitable patients. Regular monitoring of complete blood counts may be an appropriate cautionary measure in patients who are starting to receive clopidogrel.
Raymond T.F. Cheung, M.B., B.S., Ph.D.
University of Hong Kong, Pokfulam, Hong Kong, ChinaTo the Editor:
I agree with Dr. Wood that we need better post-marketing surveillance of new drugs. However, another important issue with regard to drug safety should be addressed.
Direct-to-consumer drug advertising has become increasingly prevalent.1 Patients are bombarded by advertisements in various media and ask their physicians for specific drugs. Physicians feel pressured to prescribe these drugs.2 The result is the rapid and widespread use of new drugs and, therefore, the potential for the rapid appearance of adverse effects not detected in preclinical trials.
Perhaps direct-to-consumer advertising of new drugs should be prohibited until at least a year after they have been launched. This approach would allow a more gradual introduction of the drug and would allow post-marketing surveillance to identify dangerous adverse effects, if any, before large numbers of people were exposed to the drug.
2 ReferencesMark R. Goldstein, M.D.
466 Crescent Dr., West Chester, PA 193821
Hollon MF . Direct-to-consumer marketing of prescription drugs: creating consumer demand. JAMA 1999;281:382-384
CrossRef | Web of Science | Medline2
Greenhalgh T ,Gill P . Pressure to prescribe. BMJ 1997;315:1482-1483
CrossRef | Web of Science | Medline
Author/Editor Response
The authors reply:
To the Editor: In response to Sallière and Kassler-Taub, it should be noted that after the publication of the reports on the clinical trials of ticlopidine, 4 million persons received the drug, before ticlopidine-associated thrombotic thrombocytopenic purpura had become widely recognized.1 An immune-mediated deficiency of von Willebrand factor–cleaving protease was found in seven consecutive patients with ticlopidine-associated thrombotic thrombocytopenic purpura and in two with clopidogrel-associated thrombotic thrombocytopenic purpura.2,3 Prior exposure to ticlopidine, which differs from clopidogrel in one carboxymethyl side chain, or a metabolite, may have caused sensitization, the formation of antibodies, and the rapid onset of thrombotic thrombocytopenic purpura in one patient.2
As Trontell and Honig note, the FDA coordinates drug-safety assessments, including 1192 MedWatch reports concerning clopidogrel and 4892 concerning ticlopidine. One fifth of these reports are from health professionals (as compared with 6 percent of reports on other medications). The thienopyridine derivatives ticlopidine and clopidogrel are the drugs that are first and third in terms of the frequency with which they are reported to be associated with thrombotic thrombocytopenic purpura. We have estimated the frequency of thrombotic thrombocytopenic purpura on the basis of the FDA formula for ticlopidine-associated thrombotic thrombocytopenic purpura: the number of cases ÷ (10 percent estimated underreporting × the number of users).4 We estimate that there is 1 case of thrombotic thrombocytopenic purpura per 15,000 clopidogrel-treated patients: 20 cases ÷ (10 percent underreporting × 3 million users).
MedWatch data indicate that the odds ratio for reports of thrombotic thrombocytopenic purpura associated with ticlopidine, as compared with clopidogrel, is 5.29 (Table 1Table 1
Odds Ratios for Selected MedWatch-Reported Adverse Events Associated with Ticlopidine as Compared with Clopidogrel for the Period from January 1998 to May 2000.). The incidence of ticlopidine-associated thrombotic thrombocytopenic purpura is about 1 case per 1600 to 5000 ticlopidine users.4 Thus, the MedWatch data suggest that the incidence of clopidogrel-associated thrombotic thrombocytopenic purpura is 1 case per 8500 to 26,000 clopidogrel-treated patients. This incidence is higher than the estimate of 11 cases per 1 million patient-years included in the revised product label.5 Since our report was published, active surveillance has identified 9 additional clopidogrel-associated cases of thrombotic thrombocytopenic purpura, for a total of 20 cases (75 percent of the cases occurred within two weeks after the initiation of treatment with the drug; 2 cases were fatal, 2 required more than 20 total plasma exchanges, and 3 involved relapses). After four years of ticlopidine use, 25 cases of ticlopidine-associated thrombotic thrombocytopenic purpura were reported.4 We agree with the FDA and the manufacturers of clopidogrel that physicians should be aware of the possible development of thrombotic thrombocytopenic purpura in patients taking ticlopidine or clopidogrel. We think that a weekly platelet count during initial therapy would be prudent.
5 ReferencesCharles L. Bennett, M.D., Ph.D.
Veterans Affairs Chicago Healthcare System, Lakeside Division, Chicago, IL 60614Jean M. Connors, M.D.
Harvard Medical School, Boston, MA 02115Joel L. Moake, M.D.
Baylor College of Medicine, Houston, TX 770301
Bennett CL ,Weinberg PD ,Rozenberg-Ben-Dror K ,Yarnold PR ,Kwaan HC ,Green D . Thrombotic thrombocytopenic purpura associat-ed with ticlopidine: a review of 60 cases. Ann Intern Med 1998;128:541-544
Web of Science | Medline2
Bennett CL ,Connors JM ,Carwile JM , et al. Thrombotic thrombocytopenic purpura associated with clopidogrel. N Engl J Med 2000;342:1773-1777
Full Text | Web of Science | Medline3
Tsai H-M ,Rice L ,Sarode R ,Chow TW ,Moake JL . Antibody inhibitors to von Willebrand factor metalloproteinase and increased binding of von Willebrand factor to platelets in ticlopidine-associated thrombotic thrombocytopenic purpura. Ann Intern Med 2000;132:794-799
Web of Science | Medline4
Ticlid (ticlopidine hydrochloride). Nutley, N.J.: Roche Laboratories, 1998 (package insert).
5
Plavix. New York: Bristol-Myers Squibb/Sanofi Pharmaceuticals Partnership, 2000 (package insert).
Author/Editor Response
The editorialist replies:
To the Editor: I agree with Dr. Goldstein that direct-to-consumer advertising should be restricted until sufficient data on safety have been accumulated after marketing to provide confidence in a drug's safety. I recently argued strongly for this approach.1
Although I am encouraged that Drs. Trontell and Honig, on behalf of the FDA, agree with my suggestions on intensive post-marketing surveillance, I am disturbed by their failure to acknowledge that the “cluster of reports from Bennett and others” that were received by MedWatch were in fact the very reports first highlighted in the article by Bennett and his colleagues.2 These reports were received by the FDA as a result of a specific search by Bennett and his colleagues for such adverse events, and they speak to the need, highlighted in my editorial, for innovative approaches to the detection of adverse drug effects in the future. The scientific community and the FDA need to develop initiatives to work together in order to develop and fund such novel approaches.
2 ReferencesAlastair J.J. Wood, M.D.
Vanderbilt University, Nashville, TN 37232-66021
Wood AJ . The safety of new medicines: the importance of asking the right questions. JAMA 1999;281:1753-1754
CrossRef | Web of Science | Medline2
Bennett CL ,Connors JM ,Carwile JM , et al. Thrombotic thrombocytopenic purpura associated with clopidogrel. N Engl J Med 2000;342:1773-1777
Full Text | Web of Science | Medline
- Citing Articles (7)
Citing Articles
1
DAVID WEEDON, GEOFFREY STRUTTON, ADAM I RUBIN. 2010. The vasculopathic reaction pattern. , 195-244.
CrossRef2
2010. 8 The vasculopathic reaction pattern. , 167-216.
CrossRef3
Anaadriana Zakarija, Hau C Kwaan, Joel L Moake, Nicholas Bandarenko, Dilip K Pandey, June M McKoy, Paul R Yarnold, Dennis W Raisch, Jeffrey L Winters, Thomas J Raife, John F Cursio, Thanh Ha Luu, Elizabeth A Richey, Matthew J Fisher, Thomas L Ortel, Martin S Tallman, X Long Zheng, Masanori Matsumoto, Yoshihiro Fujimura, Charles L Bennett. (2009) Ticlopidine- and clopidogrel-associated thrombotic thrombocytopenic purpura (TTP): review of clinical, laboratory, epidemiological, and pharmacovigilance findings (1989–2008). Kidney International 75, S20-S24
CrossRef4
Charles L. Bennett, Benjamin Kim, Anaadriana Zakarija, Nicholas Bandarenko, Dilip K. Pandey, Charlie G. Buffie, June M. McKoy, Amul D. Tevar, John F. Cursio, Paul R. Yarnold, Hau C. Kwaan, Davide De Masi, Ravindra Sarode, Thomas J. Raife, Joseph E. Kiss, Dennis W. Raisch, Charles Davidson, J. Evan Sadler, Thomas L. Ortel, X. Long Zheng, Seiji Kato, Masanori Matsumoto, Masahito Uemura, Yoshihiro Fujimura. (2007) Two Mechanistic Pathways for Thienopyridine-Associated Thrombotic Thrombocytopenic Purpura. Journal of the American College of Cardiology 50:12, 1138-1143
CrossRef5
Ronald S. MacWalter, Colin P. Shirley. (2002) A Benefit-Risk Assessment of Agents Used in the Secondary Prevention of Stroke. Drug Safety 25:13, 943-963
CrossRef6
Scott A. Harding, Darren L. Walters, Igor F. Palacios, Stephen N. Oesterle. (2001) Adjunctive pharmacotherapy for coronary stenting. Current Opinion in Cardiology 16:5, 293-299
CrossRef7
(2001) Current Awareness. Pharmacoepidemiology and Drug Safety 10:2, 173-188
CrossRef
