Correspondence
FDA Drug Information That Never Reaches Clinicians
N Engl J Med 2010; 362:561-563February 11, 2010
- Article
To the Editor:
In their recent Perspective article (Oct. 29 issue),1 Schwartz and Woloshin state that critical safety data were omitted from the first Zometa (zoledronic acid, Novartis) label, and they make several misleading statements that require clarification. Zometa was first approved for hypercalcemia of malignancy at a “maximum recommended” dose of 4 mg.2 Although an 8-mg dose was tested in trials, clinical development of Zometa at this dose was discontinued when concerns arose about renal toxicity. The Food and Drug Administration (FDA) never approved the 8-mg dose. The authors incorrectly suggest that the first label's warnings about the 8-mg dose were inadequate; however, the label explicitly stated that “a single dose of Zometa should not exceed 4 mg” and that the risk of renal toxicity was significantly greater with 8 mg than with 4 mg.2 The authors further claim that the exclusion of mortality data from the label was a grave omission; however, the FDA advisory panel on Zometa concluded that “although the incidence of death was higher in the Zometa 8-mg group, there was no discernable pattern or cause of death that suggested drug-relatedness.”2 The FDA and Novartis included information in the label that they deemed critical in allowing physicians to make informed prescribing decisions.
John Hohneker, M.D.
Solveig Ericson, M.D.
Gabriela Gruia, M.D.
Novartis Pharmaceuticals
solveig.ericson@novartis. com Drs. Hohneker, Ericson, and Gruia report being employees of and owning stock in Novartis. No other potential conflict of interest relevant to this letter was reported.
2 References1
Schwartz LM ,Woloshin S . Lost in transmission -- FDA drug information that never reaches clinicians. N Engl J Med 2009;361:1717-1720
Full Text | Web of Science | Medline2
Drugs@FDA. Approval history of NDA 021223: Zometa. Silver Spring, MD: Food and Drug Administration. (Accessed January 21, 2010, at http://www.accessdata.fda.gov/Scripts/cder/DrugsatFDA/index.cfm?fuseaction=Search.Label_ApprovalHistory#apphist.)
To the Editor:
Sepracor supports the goal of providing consumers and health care providers with comprehensive information about the risks and benefits of prescription drugs. The findings of the Lunesta1 (eszopiclone, Sepracor) clinical program have been publicly disseminated through several venues, including peer-reviewed publications, presentations at medical and health care conferences, and promotional activities.
As Schwartz and Woloshin note, the FDA negotiates label content with drug companies and has the final authority to determine the level of detail included in each drug's label. More detailed information about prescription-drug products is available to health care providers who seek additional data. For example, Sepracor, like most pharmaceutical companies, has a scientific staff responsible for providing detailed safety and efficacy information on its products to health care providers in response to their unsolicited requests for such data.
Sepracor supports continued discussion and collaboration among the pharmaceutical industry, the FDA, and physicians to ensure that the most appropriate information is provided to both prescribers and patients in a way that ensures and encourages patient–physician communication.
Julio Casoy, M.D.
Sepracor, Marlborough, MA
julio.casoy@sepracor. com Dr. Casoy reports being an employee of Sepracor. No other potential conflict of interest relevant to this letter was reported.
1 ReferencesTo the Editor:
Schwartz and Woloshin highlight critical deficiencies in product labeling. However, issues in product labeling extend beyond phase 3 trial data. Another concern is labeling for drug–drug interactions, especially for drugs linked to time-dependent inhibition, which can be associated with quasi-irreversible or irreversible inactivation of the heme or protein of cytochrome P-450 enzymes. Administration of drugs with time-dependent inhibition can result in profound and prolonged impairment of drug-metabolizing enzyme activity. The market withdrawals of two drugs, cerivastatin and mibefradil, were associated with time-dependent inhibition.
The FDA's draft guidance on drug interactions recommends that clinical studies be performed when time-dependent inhibition is detected in vitro to determine its clinical relevance.1 During the development of two recently approved drugs, ixabepilone (in 2007) and iloperidone (in 2009), researchers found time-dependent inhibition of the important drug-metabolizing enzyme CYP3A4. No clinical trial of CYP3A4–drug interaction was performed for either drug. Yet, the labeling for ixabepilone suggests that it does not inhibit CYP3A4,2 and the labeling for iloperidone makes no reference to this enzyme at all.3 Neither label adequately informs clinicians regarding the uncertainty about or potential risk of CYP3A4 interactions.
Carol Collins, M.D.
University of Washington, Seattle, WA
carolc3@u.washington. edu No potential conflict of interest relevant to this letter was reported.
3 References1
Guidance for Industry. Drug interaction studies — study design, data analysis, and implications for dosing and labeling. Rockville, MD: Food and Drug Administration, September 2006. (Accessed January 21, 2010, at http://www.fda.gov/downloads/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/UCM072101.pdf.)
2
Drugs@FDA. Ixempra: prescribing information. (Accessed January 21, 2010, at http://www.accessdata.fda.gov/drugsatfda_docs/label/2009/022065s002lbl.pdf.)
3
Idem. Fanapt: prescribing information. (Accessed January 21, 2010, at http://www.accessdata.fda.gov/drugsatfda_docs/label/2009/022192lbl.pdf.)
To the Editor:
In addressing the issue of FDA information that never reaches clinicians, Schwartz and Woloshin put their finger on a sore spot. Since drug labels are written by drug companies (and only secondarily negotiated and approved by the FDA), it is in these companies' best interest to deemphasize data on harm. For instance, the package insert for amlodipine says that the incidence of pedal edema associated with the drug is 10.8%. In a large study in which pedal edema was actively assessed, however, the incidence was 36.8%.1 Similarly, dry cough, an annoying adverse effect of angiotensin-converting–enzyme inhibitors, has been consistently underreported in the Physicians' Desk Reference: the currently reported incidence associated with enalapril is 1.3% (or 0.4%, after subtracting incidents associated with placebo) — so low that it seems laughable to practicing physicians, even though in a study designed to show the superiority of losartan over enalapril, the incidence of cough was 15.1%.2 Thus, “lost in transmission” applies not only to FDA drug information that never reaches clinicians but also to available drug information that never reaches the FDA.
Franz H. Messerli, M.D.
St. Luke's Roosevelt Hospital, New York, NYSripal Bangalore, M.D., M.H.A.
Brigham and Women's Hospital, Boston, MA
sbangalore@partners.org No potential conflict of interest relevant to this letter was reported.
2 References1
Chrysant SG ,Melino M ,Karki S ,Lee J ,Heyrman R . The combination of olmesartan medoxomil and amlodipine besylate in controlling high blood pressure: COACH, a randomized, double-blind, placebo-controlled, 8-week factorial efficacy and safety study. Clin Ther 2008;30:587-604
CrossRef | Web of Science | Medline2
Tikkanen I ,Omvik P ,Jensen HA . Comparison of the angiotensin II antagonist losartan with the angiotensin converting enzyme inhibitor enalapril in patients with essential hypertension. J Hypertens 1995;13:1343-1351
CrossRef | Web of Science | Medline
To the Editor:
Additional sources of comparative-efficacy data that may be included in Schwartz and Woloshin's “Prescription Drug Facts Boxes” are the publicly available briefing documents from FDA advisory committees. A recent example is the Efficacy and Safety of Dronedarone versus Amiodarone for the Maintenance of Sinus Rhythm in Patients with Atrial Fibrillation (DIONYSOS) study (ClinicalTrials.gov number, NCT00489736), which compared the safety and efficacy of dronedarone (Multaq, Sanofi-Aventis) with the safety and efficacy of amiodarone in preventing the recurrence of arrhythmia in patients with atrial fibrillation. Briefing documents prepared by the study sponsor (Sanofi-Aventis) were presented to the Cardiovascular and Renal Drugs Advisory Committee on March 18, 2009. The primary efficacy end point of the DIONYSOS trial was the first recurrence of atrial fibrillation or premature drug discontinuation for lack of efficacy or intolerance at 12 months. The frequency of this end point was 75.1% in the dronedarone group and 58.8% in the amiodarone group (hazard ratio, 1.6; 95% confidence interval, 1.3 to 2.0; P<0.001 by the log-rank test).1 These results were not reported in the FDA review documents. Dronedarone was approved in July 2009 for selected patients with atrial tachyarrhythmias.2
G. Elliott Cook, Pharm.D., B.C.P.S.
Larry D. Sasich, Pharm.D., M.P.H.
Sana R. Sukkari, Pharm.D.
Lake Erie College of Osteopathic Medicine School of Pharmacy, Erie, PA
ecook@lecom.edu No potential conflict of interest relevant to this letter was reported.
2 References1
Sanofi-Aventis. Multaq (dronedarone): briefing document — Advisory Committee Meeting of the Cardiovascular and Renal Drugs Division of the US Food and Drug Administration, March 18, 2009. (Accessed January 21, 2010, at http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/CardiovascularandRenalDrugsAdvisoryCommittee/UCM134981.pdf.)
2
FDA approves Multaq to treat heart rhythm disorder. News release of the Food and Drug Administration. (Accessed January 21, 2010, at http://www.fda.gov/NewsEvents/Newsroom/PressAnnouncements/ucm170276.htm.)
Author/Editor Response
We chose the Zometa example to highlight information about drug harms missing in the label. As Hohneker et al. point out, in 2001 the FDA approved only the 4-mg dose of Zometa to treat hypercalcemia of malignancy — not the 8-mg dose — because of concerns about harm. As we highlighted, all-cause mortality was 1.7 times higher with the 8-mg dose than with the 4-mg dose (33% vs. 19%, P=0.03). Yet the approved label never mentions death — it only warns that “the risk of renal toxicity was significantly greater with 8 mg than with 4 mg.”
The omission of mortality data from the label matters, since physicians sometimes use drugs “off-label” at higher-than-approved doses. The article reporting on the preapproval trial1 — which was funded and coauthored by Novartis — actually recommends 8 mg for “refractory cases” and omits the mortality data. Clinicians reading the article might reasonably be led to use the 8-mg dose. That Novartis changed the label in 2007, specifically stating “Do not use doses greater than 4 mg,” suggests that there was concern about such use. Nonetheless, a higher rate of death was not mentioned. We strongly believe that Novartis should correct these errors.
Contrary to what Hohneker et al. say, the only FDA panel meeting for Zometa was held in 2002 and was for a different indication: preventing skeletal fractures in people with bony metastases.2 In one of three trials reviewed, the higher dose (one group was randomly assigned to receive 8 mg but was switched to 4 mg in the middle of the study because of problems observed in the hypercalcemia trial) was again associated with a higher rate of death (22% vs. 14.5%, P=0.04).3 Even if there was uncertainty about the “drug-relatedness” of these deaths, it seems irresponsible to ignore the increased rate of death observed in two independent trials.
We appreciate Casoy's comments. Though it is true that physicians can request safety and efficacy information from the manufacturer, this should not be necessary; the information should already be on the label.
We thank Messerli and Bangalore for pointing out that some important drug information may never reach the FDA. Ross et al.4 have also highlighted the usefulness of cumulative meta-analysis in identifying drug harms. Whether the FDA or independent investigators should take on this role is an open question.
Finally, Cook et al. highlight that another good source of information — FDA advisory committee documents — should be made more accessible. The FDA needs a mechanism for efficiently summarizing what they know about drugs — from preapproval trials, from advisory committee meetings, and from postapproval studies.
Steven Woloshin, M.D.
Dartmouth Institute for Health Policy and Clinical Practice, Hanover, NHLisa M. Schwartz, M.D.
Veterans Affairs Medical Center, White River Junction, VTSince publication of their article, the authors report no further potential conflict of interest.
4 References1
Major P ,Lortholary A ,Hon J , et al. Zoledronic acid is superior to pamidronate in the treatment of hypercalcemia of malignancy: a pooled analysis of two randomized, controlled clinical trials. J Clin Oncol 2001;19:558-567
Web of Science | Medline2
sNDA21-386, Zometa (zoledronic acid for injection). Presented by Novartis to the Food and Drug Administration Oncologic Drugs Advisory Committee, January 31, 2002. (Accessed January 21, 2010, at http://www.fda.gov/ohrms/dockets/ac/02/briefing/3827b1.htm.)
3
sNDA21-386, Zometa: appendix 1 — review of zoledronate safety in prostate cancer. Study 039. Presented by Novartis to the Food and Drug Administration Oncologic Drugs Advisory Committee, January 31, 2002. (Accessed January 21, 2010, at http://www.fda.gov/ohrms/dockets/ac/02/briefing/3827b1_04_FDA-Part2%20.htm.)
4
Ross JS ,Madigan D ,Hill KP ,Egilman DS ,Wang Y ,Krumholz HM . Pooled analysis of rofecoxib placebo-controlled clinical trial data: lessons for postmarket pharmaceutical safety surveillance. Arch Intern Med 2009;169:1976-1985
CrossRef | Web of Science | Medline
- Citing Articles (1)
Citing Articles
1
(2010) Current awareness: Pharmacoepidemiology and drug safety. Pharmacoepidemiology and Drug Safety 19:8, i-xiii
CrossRef
