Hss.medschool.ucsf.edu

Annals of Internal Medicine
BiDil for Heart Failure in Black Patients: The U.S. Food and Drug
Administration Perspective
Robert Temple, MD, and Norman L. Stockbridge, MD, PhD
Critics of the U.S. Food and Drug Administration (FDA) approval of trials in racially mixed patient populations that compared hydral- the fixed combination of hydralazine hydrochloride, 37.5 mg, and azine hydrochloride–isosorbide dinitrate with placebo or with ena- isosorbide dinitrate, 20 mg, for treating heart failure in black pa- lapril. Both trials showed little or no overall effect of hydralazine tients have suggested that data were insufficient to distinguish hydrochloride–isosorbide dinitrate in the mostly white patient pop- treatment effects in black and white people; that distinctions based ulation but hinted at a substantial effect in subsets of black patients.
on race, rather than pathophysiology, were scientifically unreason- Perhaps most critically, the criticisms do not appreciate the urgency able; and that a “race-based” approval could be a commercial ploy of strong scientific evidence of a substantial survival benefit in black to avoid a more expensive and prolonged full evaluation of a drug.
patients. A serious attempt to avoid race-based approval by man- The criticisms acknowledge that data supporting the approval came dating study of a mixed population to identify a possible white from a well-designed clinical trial in which self-identified black pa- patient–responder subset, particularly without a plausible hypothesis tients with heart failure who took hydralazine hydrochloride–isosor- as to what that subset might be, would have required years of bide dinitrate with standard therapy experienced a statistically sig- work, many thousands of patients, and wholly unreasonable delay nificant 43% (95% CI, 11% to 63%) reduction in mortality in approval of a treatment whose effectiveness had been well- compared with those who took only the standard therapy. The documented in the group for which it was intended.
criticisms do not always recognize that the decision to conduct the Ann Intern Med. 2007;146:57-62.
www.annals.org
trial in only black patients reflected careful analyses of 2 previous For author affiliations, see end of text.
Considerable discussion has followed the FDA approval patients and supported a differential effect in black and white
in June 2005 of BiDil, a fixed-dose combination of isosorbide dinitrate, 20 mg, and hydralazine hydrochloride, There was very strong evidence that hydralazine hy- 37.5 mg, for use in “the treatment of heart failure as an drochloride–isosorbide dinitrate was extremely effective in adjunct to standard therapy in self-identified black patients self-identified black patients and considerable evidence that to improve survival, to prolong time to hospitalization for the effects were far smaller, if present at all, in white pa- heart failure, and to improve patient-reported functional tients. This evidence has been substantially understated by status” (1). Approval followed the unanimous recommen- many commentators and critics (3– 6).
dation of the FDA’s Cardiovascular and Renal Drugs Ad- The principal trial of BiDil was the African American visory Committee, with 7 of 9 members supporting the Heart Failure Trial (A-HeFT) (7), a randomized compari- race-specific label and 2 of 9 members urging a broader son of hydralazine hydrochloride–isosorbide dinitrate and claim (2). While the importance of the effect shown in placebo in self-identified black patients with New York black patients has generally been recognized (3), critics Heart Association (NYHA) class III or IV (mostly III) have asked whether data were adequate to distinguish the heart failure. Treatment was added to standard therapy effects of hydralazine hydrochloride–isosorbide dinitrate in (94% of patients received diuretics, 87% received ␤-block- black and white patients; whether commercial rather than ers, 93% received angiotensin-converting enzyme [ACE] medical considerations led the drug manufacturer to re- inhibitors or angiotensin II blockers, 62% received digitalis strict the critical clinical trial of BiDil to an entirely black glycoside, and 39% received aldosterone antagonist).
patient population (3, 4); whether the FDA should have The A-HeFT investigators used a complex weighted allowed or encouraged such a trial; whether distinguishing combination of all-cause death, first hospitalization for drug responses in black and white patients by race rather heart failure, and change in quality of life at 6 months as than pathophysiology is scientifically reasonable (3–5); and the primary end point, with the individual components whether a “race-based” drug approval can be abused, lead- specified as secondary end points. The study was terminat- ing to suggestions of racial inferiority or stereotyping.
ed—after 1050 patients had been randomly assigned— on These questions and others are worthy of discussion, butthey did not cause the FDA to doubt that BiDil should beapproved.
Print
Related article. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52
THE FDA PERSPECTIVE
Web-Only
1. Data from 3 clinical trials showed dramatic effective- ness of hydralazine hydrochloride–isosorbide dinitrate in black www.annals.org
2 January 2007 Annals of Internal Medicine Volume 146 • Number 1 57
In the Balance BiDil: The U.S. Food and Drug Administration Perspective Table 1. African American Heart Failure Trial Results*
End Point
Hydralazine
Placebo Group
Hazard Ratio (95% CI)
Hydrochloride–Isosorbide
(n ؍ 532), %
Dinitrate Group (n ؍ 518), %
the basis of a survival advantage (a 43% [95% CI, 11% to trial had assay sensitivity: the ability to distinguish effective 63%] reduction in mortality) in the hydralazine hydro- treatment from ineffective treatment (10).
chloride–isosorbide dinitrate group. The primary compos- We (at the FDA) were asked to approve hydralazine ite end point showed a significant effect (P Ͻ 0.021), but hydrochloride–isosorbide dinitrate for black patients on the more critical results were mortality rate and time to the basis of the post hoc racial analyses of V-HeFT I and first heart failure hospitalization. Both outcomes statisti- II. Although we did not consider these analyses to be a cally significantly improved with treatment (Table 1).
sufficient basis for approval, the substantially different ef- The decision to conduct A-HeFT in black patients was fect of hydralazine hydrochloride–isosorbide dinitrate in supported by 2 earlier well-controlled studies that had white and black patients in 2 studies led us to believe that strongly suggested a differential mortality effect of hydral- there was a good case for conducting an additional study in azine hydrochloride–isosorbide dinitrate in black and black patients. We also believed that a study of reasonable white patients. Two Veterans Administration Cooperative size in a population of both black and white patients would Vasodilator Heart Failure Trials (V-HeFT) tested the have little chance of detecting a treatment effect in white drugs in patients with predominantly NYHA II and III patients, particularly if the treatment were added to current heart failure. The first trial (V-HeFT I [8]) compared hy- best therapy, which already reduces mortality substantially.
dralazine hydrochloride–isosorbide dinitrate with placebo A reasonably powered (␤ ϭ 80%) study in white patients, (and prazosin, not further discussed), while the second trial assuming an effect size of, say, 15%, would require about (V-HeFT II [9]) compared hydralazine hydrochloride–i- 16 000 patients if placebo mortality were similar to that sosorbide dinitrate with enalapril, which was by then an observed in A-HeFT. The FDA, therefore, advised Nitro- established heart failure treatment. Table 2 shows the re-
Med that the demonstration of a mortality benefit in self- sults of these studies. The first trial showed a nearly statis- identified black patients could be a basis for marketing tically significant effect of hydralazine hydrochloride– approval of hydralazine hydrochloride–isosorbide dinitrate.
isosorbide dinitrate on survival, and the second trial The mortality benefit of BiDil in black patients is, showed a nearly statistically significant advantage of enala- thus, supported by 3 well-controlled studies: most convinc- pril over hydralazine hydrochloride–isosorbide dinitrate.
ingly in A-HeFT and V-HeFT I and in V-HeFT II by an Thus, neither study showed an effect of the combination in informal noninferiority finding. Approval of BiDil was not based on a single trial where all data came from the black Post hoc subset analyses by race, however, indicated patient population, as has been suggested. The FDA’s en- that responses were not uniform in racial groups. In V- couragement of A-HeFT, a single-population trial, arose HeFT I, there was a large reduction in mortality in the from recognition that a larger study of black and white small subset of black patients (n ϭ 128) that just reached patients was not likely to yield any additional useful infor- nominal statistical significance (although it was not cor- rected for multiple end points). There was a much weaker 2. Not understanding the reasons for the difference in favorable trend in the larger subset of white patients. In treatment effect by race did not justify withholding the treat- V-HeFT II, hydralazine hydrochloride–isosorbide dinitrate ment from those who could benefit from it. was statistically significantly inferior to enalapril in white Race or ethnicity is clearly a highly imperfect descrip- patients (P ϭ 0.02). The magnitude of the difference was tion of the genomic and other physiologic characteristics not far from what one might expect in a comparison of that cause people to differ, but it can be a useful proxy for enalapril and placebo. In contrast, the mortality rates in the those characteristics until the pathophysiologic bases for enalapril and hydralazine hydrochloride–isosorbide dini- observed racial differences are better understood (11). The trate groups were almost identical in black patients. Al- history of treating heart failure should give pause to anyone though no formal noninferiority analysis was conducted, who imagines that developing effective therapies requires the effect size in black patients, in a setting in which ena- an understanding of pathophysiology or that pathophysio- lapril was almost superior to hydralazine hydrochloride– logic-based expectations reliably predict outcomes. Drugs isosorbide dinitrate in white patients, suggested that the that were once contraindicated in heart failure (␤-blockers) 58 2 January 2007 Annals of Internal Medicine Volume 146 • Number 1
www.annals.org
BiDil: The U.S. Food and Drug Administration Perspective In the Balance have become standard treatment and the most logical treat- might respond, the study would have had little chance of ments for heart failure, inotropic agents to strengthen the revealing predictors of racial differences. Even if anyone weakened heart muscle, have neutral effects on mortality at were willing to perform such a study, we believe that best (12) and have more often proved to be lethal (13–17).
spending years exploring the basis for the observed The effectiveness requirement of the Federal Food, black–white difference before approving hydralazine Drug, and Cosmetic Act asks only for evidence that a drug hydrochloride–isosorbide dinitrate for the population in will have its claimed effect, without reference to why it has which it dramatically reduced mortality would not have that effect. The idea that the FDA should have waited to approve hydralazine hydrochloride–isosorbide dinitrate 3. Race and other demographic characteristics have long until we understood why white and black patients differ in been important to consider in analysis of trials and as a matter response, despite the 3 well-controlled trials suggesting or showing a mortality benefit in black patients, has led to Since the early 1980s, there has been substantial con- cern about inadequate representation of women, elderly For example, Avorn (5), unimpressed by the 43% people, black people, and other groups in the drug devel- mortality risk reduction demonstrated in A-HeFT and ap- opment process. The interest was partly ethical, reflecting parently believing that the racial difference hypothesis was concerns about unequal access to potentially valuable treat- based on a post hoc analysis of a single trial, thought that ments, but it was far more a concern that lack of partici- “[t]his interesting observation could have been enormously pation of these subgroups would lead to incorrect conclu- important in helping us understand the pathophysiology of sions for those groups about benefits or adverse effects of [heart failure] in a particularly vulnerable population.” He treatments. Although, on examination, participation of suggested that instead of FDA approval of BiDil, a plausi- black people and both sexes approximately reflected the ble next step would have been to test the racial difference prevalence of the conditions being treated in the popula- hypothesis in a controlled trial that enrolled both black and tion (18), it was found that clinical databases in new drug white patients to look for differences in outcomes and pre- applications were not examined for potential differences in dictors of those differences, including “genetic markers, response among demographic subgroups, at least through self-identified race, diet, and other risk factors” (4). Under- standing pathophysiology is good, of course, but it ranks By 1993, the FDA had finalized specific guidelines on well behind a documented survival effect in importance, assessment of elderly persons (19) and of both sexes (18) in and the suggested study is more or less the same as V- drug development and more general guidelines on evalua- HeFT I. Without any plausible hypotheses (genetic, di- tion of demographic subgroups for safety, effectiveness, etary, and other risk factors) about which white patients and dose-response in new drug applications (20). Finally, Table 2. Vasodilator Heart Failure Trial I and II Results*
Variable
Patients,
Hazard Ratio
Mortality, %
Hydralazine hydrochloride–isosorbide dinitrate Hydralazine hydrochloride–isosorbide dinitrate Hydralazine hydrochloride–isosorbide dinitrate V-HeFT II
Hydralazine hydrochloride–isosorbide dinitrate Hydralazine hydrochloride–isosorbide dinitrate Hydralazine hydrochloride–isosorbide dinitrate * V-HeFT ϭ Vasodilator Heart Failure Trial.
† There were 7 other minority group patients.
‡ Effect favored enalapril.
§ There were 15 other minority group patients.
www.annals.org
2 January 2007 Annals of Internal Medicine Volume 146 • Number 1 59
In the Balance BiDil: The U.S. Food and Drug Administration Perspective the regulations describing new drug application submis- tion. These limitations would represent a loss if, in fact, the sions (21) were changed in 1998 to require analyses of drug were actually more broadly effective. The FDA has, safety and effectiveness in demographic subsets of the pa- therefore, encouraged broad inclusion of patients in trials.
tient population. Recent changes in labeling (package in- In that case, if treatment is ultimately directed toward a sert) regulations (22) also describe the need to include per- group that benefits most, this is done with the knowledge tinent demographic subset information in many sections of that the treatment performs less well in other groups. We believe that this describes the situation with hydralazine The FDA was hardly alone in this interest in possible hydrochloride–isosorbide dinitrate. But there could be differences among demographic subsets of the population.
cases in which little information was available for the In 1993, Congress passed the National Institutes of Health “other” group, a growing possibility in an era of “targeted” (NIH) Revitalization Act (23), asking the NIH to ensure therapy. When a therapy is shown to be effective for a that inclusion of women and minorities in phase 3 trials responsive subgroup, critical questions include how much was sufficient to allow for valid analyses of differences in data should be expected on the drug’s effects in other intervention effect. A conspicuous example of the effect of groups; how small an effect needs to be detected or ex- the law is the Antihypertensive and Lipid-Lowering Treat- cluded in those groups; when should the data on those ment to Prevent Heart Attack Trial (ALLHAT) (24), a other groups be expected (before or after approval and how 40 000-patient trial of blood pressure–lowering and lipid- long after approval); and to what extent the FDA can insist lowering interventions in which the goal patient popula- on the conduct of further studies. These questions are un- answered because we are still in the early days of individ- Finally, despite legitimate concerns about overinter- ualization, but they are being actively considered.
preting subset analyses (25), published reports of outcome We believe that, in most cases, interest in individual- studies in recent years almost always show forest plots of ized therapy need not conflict with the desire for broad effects in a wide variety of subsets, invariably including experience and information about the effects of drugs in a demographic subsets (such as sex, age, and race) and other range of patients. Even if therapy is directed at a responsive subsets (for example, disease severity and concomitant ill- subgroup, providing adequate directions for clinical use in ness). These presentations illustrate the growing interest in the drug label—a requirement of law—will generally call possible subset differences in response, even where these for a reasonable amount of data on excluded patient pop- cannot be pathophysiologically explained. Also, epidemio- ulations or clear evidence that the treatment cannot work logic studies regularly examine racial aspects of disease in those people. In the case where a drug has shown an prevalence and outcome. Bloche (3) notes, for example, important effect in a particular group, however, it seems that “[t]here is wide agreement that blacks die from heart hard to argue that the group can ethically be denied the failure at rates disproportionate to those among whites.” therapy while investigators seek to determine whether the One might ask why, if etiology and prevalence of diseases effect applies more broadly. In that situation, data on the can be racially linked, a difference in treatment response effect of the drug in other populations would often come would be surprising, even if we did not understand the from phase 4 (postmarketing) studies. What one hopes, of reason for it. Given the long history of urgent interest in course, is that the pathophysiologic basis for a differential searching for racial and other demographic differences, response can be defined in future studies, allowing rational which surely accepted the possibility that such differences might be discovered, it seems surprising that there would Fortunately, in the case of hydralazine hydrochloride– be so much discomfort when one was found.
isosorbide dinitrate, we did not need to face the most dif- 4. Regulatory and other concerns associated with drug ficult aspects of this question because data indicating a approval for narrow patient populations did not justify with- much smaller effect in white patients were available at the holding BiDil from those who could benefit from it. time of marketing. Whether there is any effect in white The growing interest in targeting therapy to specific patients, and exactly who the responsive white population subgroups (“individualization”) raises legitimate questions might be, remains to be determined. Given the lack of any about implications for the broader population, drug man- plausible predictor of such a responsive subset and the mas- ufacturer incentives, and possible stigmatization of a target sive study needed to explore effectiveness in an unselected group. Not all the questions are yet answerable, but none white population, the FDA did not seek specific commit- has justified denying the benefits of hydralazine hydrochlo- ments from NitroMed to conduct studies in white pa- ride–isosorbide dinitrate to the subpopulation in which it tients, although the company has expressed interest in find- ing a pharmacologic response predictor.
Approval of a drug for a specific subgroup means that Some critics have suggested that the FDA approval of its use in other groups would be considered “off-label.” drugs in narrow subgroups will allow drug manufacturers Although off-label use is not barred by law, third-party to “get off cheap,” encouraging them to seek out narrow payers may not pay for such use and, of course, manufac- niche populations that are easy to study and suggest nov- turer promotion must be directed at the approved popula- elty, and that other populations will be deprived of the 60 2 January 2007 Annals of Internal Medicine Volume 146 • Number 1
www.annals.org
BiDil: The U.S. Food and Drug Administration Perspective In the Balance opportunity for benefit. The FDA does not regulate the come of disease whose biological or other basis we do not economics of drug development and generally does not understand are countless. Adverse consequences of these evaluate drug manufacturers’ motives, but the issues are observations have not been identified. Indeed, some of complex. There is great interest in efficient drug develop- these observations have led to appropriate efforts to im- ment, and one step toward efficiency is studying drugs in patient populations with high event rates or greater respon-siveness (26). This thought is not new. The ability to en-rich populations to be studied has, for example, facilitated CONCLUSIONS
development of such critical treatments as ACE inhibitors The FDA approval of a fixed combination of hydral- for heart failure. The first mortality study of enalapril (Co- azine hydrochloride–isosorbide dinitrate to treat heart fail- operative North Scandinavian Enalapril Survival Study ure in self-identified black patients was a scientifically rea- [CONSENSUS]) enrolled an extremely sick population sonable, data-based decision, one that provided a major with high mortality, allowing a successful study in just 253 benefit in a group that is particularly burdened by conges- people (27). When these results were added to the labeling tive heart failure. The evidence of benefit in black patients for enalapril, we did not know the effect of the drug in less is very strong, and the evidence that white patients have ill patients. Much larger studies in such patients were per- less, if any, benefit, is also strong. We hope that further formed later, but the importance of the CONSENSUS research elucidates the genetic or other factors that predict finding can hardly be overstated. In some cases, there may the usefulness of hydralazine hydrochloride–isosorbide di- be no reason to expect anyone to study the broader patient nitrate. Until then, we are pleased that one defined group population until an effect can be shown in some patient has access to a dramatically life-prolonging therapy.
population, providing “proof of concept.” While we sharethe community’s interest in broadly developing important From the U.S. Food and Drug Administration, Silver Spring, Maryland.
new therapies, we also would not want to stifle innovationand efficient studies and deprive the community of valu- Potential Financial Conflicts of Interest: None disclosed.
able treatments. Again, this issue did not arise for hydral- Requests for Single Reprints: Robert Temple, MD, U.S. Food and
azine hydrochloride–isosorbide dinitrate because the Drug Administration, 10903 New Hampshire Avenue, Silver Spring, broader patient population had been studied and the white MD 20993-0002; e-mail, [email protected].
patients clearly had a small response at best. There waslittle likelihood that BiDil’s developer would have been Current author addresses are available at www.annals.org.
willing to conduct the 16 000-patient study that wasneeded to show a small effect in white patients, and therewas no evidence of interest in such a study by any other References
private or government sponsor before A-HeFT.
1. BiDil [package insert]. Lexington, MA: NitroMed Inc.; 2005.
While the FDA cannot free a manufacturer from the 2. Proceedings of the U.S. Food and Drug Administration Center for Drug obligation to provide legally sufficient data to support drug Evaluation and Research Cardiovascular and Renal Drugs Advisory Committee[transcript]. 16 June 2005. Rockville, MD: U.S. Food and Drug Administration; approval, judgments are involved in determining what con- 2005. Accessed at www.fda.gov/ohrms/dockets/ac/05/transcripts/2005-4183T1 stitutes sufficient data. The striking effects in black patients in A-HeFT and V-HeFT I, the need for heart failure treat- 3. Bloche MG. Race-based therapeutics. N Engl J Med. 2004;351:2035-7.
ments for black patients, and the substantial delay involved [PMID: 15533852]
4. Kahn J. How a drug becomes “ethnic”: law, commerce, and the production of
in conducting an all-race study—if it could have been done racial categories in medicine. Yale J Health Policy Law Ethics. 2004;4:1-46.
at all and the near certainty that it would not be done—all argued strongly for the path that the FDA took.
5. Avorn J. FDA standards— good enough for government work? N Engl J Med.
Finally, the possibility that drug approval for a partic- 2005;353:969-72. [PMID: 16148281]
6. Haga SB, Ginsburg GS. Prescribing BiDil: is it black and white? J Am Coll
ular subgroup could lead to stigmatization of that sub- group has led Kahn (4) to suggest that evidence of a ge- 7. Taylor AL, Ziesche S, Yancy C, Carson P, D’Agostino R Jr, Ferdinand K, et
netic basis for a racial distinction must clearly be shown al. Combination of isosorbide dinitrate and hydralazine in blacks with heart
failure. N Engl J Med. 2004;351:2049-57. [PMID: 15533851]
before a race-specific approval, even in the face of compel- 8. Cohn JN, Archibald DG, Ziesche S, Franciosa JA, Harston WE, Tristani FE,
ling benefit in one race. Aside from the dubious ethics of et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure.
that proposal, the concern itself seems far-fetched. Differ- Results of a Veterans Administration Cooperative Study. N Engl J Med. 1986; ences in drug response by sex (for example, women’s 314:1547-52. [PMID: 3520315]
9. Cohn JN, Johnson G, Ziesche S, Cobb F, Francis G, Tristani F, et al. A
greater susceptibility to torsade de pointes), race (for exam- comparison of enalapril with hydralazine-isosorbide dinitrate in the treatment of ple, reduced blood pressure response to ACE inhibitors chronic congestive heart failure. N Engl J Med. 1991;325:303-10. [PMID: and angiotensin II antagonists and greater susceptibility to ACE inhibitor–induced angioedema in black people [28]), 10. Temple R, Ellenberg SS. Placebo-controlled trials and active-control trials in
the evaluation of new treatments. Part 1: ethical and scientific issues. Ann Intern
and age (too many to count) are reasonably well-estab- lished, and demographic differences in prevalence and out- 11. Harty L, Johnson K, Power A. Race and ethnicity in the era of emerging
www.annals.org
2 January 2007 Annals of Internal Medicine Volume 146 • Number 1 61
In the Balance BiDil: The U.S. Food and Drug Administration Perspective pharmacogenomics. J Clin Pharmacol. 2006;46:405-7. [PMID: 16554447] an application. Rockville, MD: U.S. Food and Drug Administration; 1988. Ac- 12. The effect of digoxin on mortality and morbidity in patients with heart cessed at www.fda.gov/cder/guidance/statnda.pdf on 25 October 2006.
failure. The Digitalis Investigation Group. N Engl J Med. 1997;336:525-33.
22. Requirements on content and format of labeling for human prescription drug 13. Packer M, Carver JR, Rodeheffer RJ, Ivanhoe RJ, DiBianco R, Zeldis SM,
and biological products. Final rule. Fed Regist. 2006;71:3921-97. [PMID: et al. Effect of oral milrinone on mortality in severe chronic heart failure. The
PROMISE Study Research Group. N Engl J Med. 1991;325:1468-75. [PMID: 23. NIH Revitalization Act, Pub. L. No. 103-43 (1993).
24. Davis BR, Cutler JA, Gordon DJ, Furberg CD, Wright JT Jr, Cushman
14. Packer M, Rouleau J, Swedberg K, Pitt B, Fisher L, Klepper M. Effect of
WC, et al. Rationale and design for the Antihypertensive and Lipid Lowering
flosequinon on survival in chronic heart failure: preliminary results of the PRO- Treatment to Prevent Heart Attack Trial (ALLHAT). ALLHAT Research FILE study [Abstract]. Circulation. 1993;88(suppl I):I-301.
Group. Am J Hypertens. 1996;9:342-60. [PMID: 8722437] 15. Cohn JN, Goldstein SO, Greenberg BH, Lorell BH, Bourge RC, Jaski BE,
25. Yusuf S, Wittes J, Probstfield J, Tyroler HA. Analysis and interpretation of
et al. A dose-dependent increase in mortality with vesnarinone among patients
treatment effects in subgroups of patients in randomized clinical trials. JAMA.
with severe heart failure. Vesnarinone Trial Investigators. N Engl J Med. 1998; 339:1810-6. [PMID: 9854116]16. Xamoterol in severe heart failure. The Xamoterol in Severe Heart Failure 26. Temple R. Special study designs: early escape, enrichment, studies in non-
Study Group. Lancet. 1990;336:1-6. [PMID: 1694945] responders. Communications in Statistics: Theory and Methods. 1994;23:499- 17. Just H, Hjalmarsson A, Remme WJ, et al. Pimobendan in congestive heart
failure: results of the PICO trial [Abstract]. Circulation. 1995;92(suppl I):I-722.
27. Effects of enalapril on mortality in severe congestive heart failure. Results of 18. Guideline for the study and evaluation of gender differences in the clinical the Cooperative North Scandinavian Enalapril Survival Study (CONSENSUS).
evaluation of drugs; notice. Fed Regist. 1993;58:39406-16. [PMID: 11645233] The CONSENSUS Trial Study Group. N Engl J Med. 1987;316:1429-35.
19. Guideline for the study of drugs likely to be used in the elderly. Rockville, MD: U.S. Food and Drug Administration; 1989. Accessed at www.fda.gov/cder 28. Brown NJ, Ray WA, Snowden M, Griffin MR. Black Americans have an
/guidance/old040fn.pdf on 25 October 2006.
increased rate of angiotensin converting enzyme inhibitor-associated angioedema.
20. Guideline for the format and content of the clinical and statistical sections of Clin Pharmacol Ther. 1996;60:8-13. [PMID: 8689816] 62 2 January 2007 Annals of Internal Medicine Volume 146 • Number 1
www.annals.org
Annals of Internal Medicine
Current Author Addresses: Drs. Temple and Stockbridge: U.S. Food
and Drug Administration, 10903 New Hampshire Avenue, Silver
Spring, MD 20993-0002.
www.annals.org
2 January 2007 Annals of Internal Medicine Volume 146 • Number 1 W-9

Source: http://www.hss.medschool.ucsf.edu/PDF/Fernandez_Bidil%20FDA%20response.pdf