BiDil, regulatory milestone but a tough sell for insurers and patients, charts uncertain course

December 13, 2006

Chicago IL - If there were ever a cardiovascular drug to inspire both dreamers and cynics, it might be BiDil (NitroMed, Lexington, MA), the isosorbide dinitrate-hydralazine (ISDN-H) combo medication approved by the FDA last year expressly for self-described African Americans with heart failure.

The BiDil bottle design is trademarked by NitroMed.

The landmark approval, the first with an indication specifying a single racial group, was acclaimed as a step toward a world in which drug treatments are selected for individual patients based on their genetic predispositions for a good response. But it also spotlighted a question that has weighed on BiDil since the early days of its development—whether the scientifically fuzzy construct of race is a viable basis for targeting drug therapies.

That very issue was debated publicly by two of the world's best-known HF specialists at last month's American Heart Association (AHA) 2006 Scientific Sessions. Their exchange on an idea that gives BiDil something of a mystique exposed the potential and the limitations of using an admittedly soft surrogate in place of focused pharmacogenetic responses.

One has to wonder how much of this is related to the inherent disquiet that we all realize when we're dealing with race-based issues, especially in healthcare.

The BiDil approval also challenged the drug's manufacturer to overcome an inherent Catch-22: its funding of the African-American Heart Failure Trial (A-HeFT) [1], which showed BiDil could significantly improve outcomes for an especially high-risk HF population, led directly to the FDA's approval of its proprietary fixed-dose formulation of ISDN-H. But now the company must market and profitably sell a product some clinicians and third-party payers see, accurately or not, as an expensive repackaging of two generic medications that can be used as effectively for a fraction of the cost. And those obstacles are in addition to another that would confront any new drug for HF—a crowded field serving patients with overfull medicine cabinets.

The evidence base and barriers to use in practice

As previously reported by heart wire , the A-HeFT trial randomized 1050 self-identified African Americans with documented LV systolic dysfunction and predominantly NYHA class 3 HF to receive either BiDil or placebo on top of state-of-the-art medical therapy. The active regimen initially consisted of one tablet three times daily, and the tid dosage was subsequently doubled as long as there were no side effects. Over a mean of 10 months, BiDil was associated with a 43% drop in relative mortality risk and a 39% decrease in the relative risk of HF hospitalization.

Dr Clyde Yancy

The A-HeFT findings were incorporated into formal guidelines, yet recent registry data suggest that no more than 20% of the target population is taking BiDil or its separate generic components, Dr Clyde Yancy (Baylor Heart and Vascular Institute, Dallas, TX), a member of the trial's steering committee, told heart wire . It appears in part a classic failure of evidence to penetrate clinical practice, but there are also challenges special to BiDil.

For starters, asked Yancy, "How do you add yet another compound, how do you justify yet another incremental cost, how do you manage an ever-increasingly complex medical regimen for patients who are chronically ill?"

Dr Mariell Jessup

According to Dr Mariell Jessup (University of Pennsylvania, Philadelphia) "Nobody has any question that BiDil was extraordinarily effective in the A-HeFT trial. Having said that, it's also true that very few African Americans are getting the drug." One reason, Jessup, not one of the trial's investigators, told heart wire , is that HF patients are likely to already be taking a burdensome number of pills and tend to resist taking more. Another is that many insurers and the Department of Veterans Affairs "won't pay for the drug." They will cover the generic components, she observed, but that means patients would have to add twice as many three-times-daily pills to their existing regimen.

Jessup said she often "tries to give BiDil" to her African American patients who are not doing well on standard medications. "Almost always their insurance company won't pay for it. And then I try to give them hydralazine and isosorbide dinitrate. In general, they don't usually take it," she said. "It's an unwieldy group of drugs to use if you don't have the combination drug of BiDil."

It does not mean every African American you see needs to be on BiDil.

Other options exist for blacks who are already maxed out on standard therapy, she said. They include aldosterone inhibitors and angiotensin-receptor blockers (ARBs). For whites, according to Jessup, there's no compelling reason to insist on direct vasodilators.

Dr Marvin A Konstam (Source: Tufts-New England Medical Center)

Even the evidence-based mandate for blacks could be less comprehensive than typically portrayed. Jessup observed that A-HeFT entered patients who were symptomatic despite best modern medical therapy and who had systolic heart failure. Yet frequently—more often than is true for whites—African Americans with HF have "normal" ejection fractions and are suboptimally on the other medications. "So it does not mean every African American you see needs to be on BiDil."

The existence of alternative add-on drug choices may help explain low BiDil utilization, speculates Dr Marvin A Konstam (Tufts-New England Medical Center, Boston, MA), who didn't participate in A-HeFT. "When we just had ACE inhibitors, it was pretty simple. When we just had beta blockers and ACE inhibitors it wasn't too complicated. Once we started adding aldosterone blockers it became complicated," he told heart wire . "And now we've got this combination on top of it. So maybe people are getting a little fatigued in terms of their willingness to integrate more information into their practice."

Race's impact on utilization

Dr William Abraham

At the AHA sessions, Dr William Abraham (Ohio State University, Columbus) also observed that HF in blacks differs from HF in whites and noted that A-HeFT's population was unlike any other in clinical trials. He was publicly responding to comments made in a formal presentation by Dr Jay N Cohn (University of Minnesota, Minneapolis), who led the early research that led to A-HeFT and is more than anyone the inventor of BiDil. Abraham noted that the A-HeFT population, compared with other HF trials, had an increased prevalence of hypertension, obesity, and metabolic syndrome or diabetes and far less ischemic heart disease. "So is this a color-of-skin debate or is it really some other phenotypic marker . . . that predicts response?"

Dr Jay N Cohn

Cohn, agreeing that A-HeFT's population was unique in the world of clinical trials, said a multivariate analysis accounting for the relevant variables found none that influenced BiDil's efficacy. A-HeFT established blacks as BiDil's target population, but its scope could potentially be expanded. "A-HeFT was remarkably positive, but I suspect it would have required a certainly larger population to show a benefit in white people."

Cohn said he believes ISDN-H is "mandated" for blacks. "I would never have a patient with heart failure who says he or she is African American without using BiDil as part of their therapy. And I don't use that same approach in white people, but I certainly use the drug combination based on mechanistic insight."

Yancy said when interviewed that if physicians are not putting the A-HeFT evidence base into practice, "one has to wonder how much of this is related to the inherent disquiet that we all realize when we're dealing with race-based issues, especially in healthcare. It continues to be one of the more inflammatory topics in American culture, it continues to be something that we don't deal with very well. And it continues to be a very awkward friction point for practitioners and patients. . . . So, taking the safe way out, many practitioners may not like to bring it up at all."

Progress in genomics or the discovery of treatment-guiding biomarkers might free clinicians of the perception that they are being asked to treat patients according to skin color, Yancy observed. But for now, "We have a high-risk population and a new therapy to treat heart failure, period. Independent of race," he said. "Whether the practitioner goes with generics or goes with proprietary compounds, the benefit cannot be overlooked."

Is race a valid proxy for genetics in drug selection?

That question was the subject of a Controversy presentation at the recent AHA meeting, which assigned two renowned experts to represent and defend opposing answers [2]. Cohn's defense of race-based drug targeting was met by denouncements from Dr Milton Packer (University of Texas Southwestern Medical Center, Dallas).

Dr Milton Packer

As previously reported by heart wire , Packer—who is not on the A-HeFT paper's list of investigators—spoke in support of BiDil's clinical-trial track record at the June 2005 meeting of the FDA's Cardiovascular and Renal Drugs Advisory Committee, which unanimously recommended the drug for approval.

Cohn summarized the well-documented history of observations that culminated in the A-HeFT trial, including the finding of decreased mortality among patients who received ISDN-H in the first V-HeFT study [3]]; the signal of better survival with an ACE inhibitor as compared with ISDN-H in V-HeFT 2 [4]], followed by waning interest in ISDN-H; and subsequent indications of a possible survival advantage for blacks on the two vasodilators in both V-HeFT trials [5]. Together the findings suggested ISDN-H would be "exceedingly effective" in blacks with HF, Cohn said, and so A-HeFT was born.

You can say everyone should be treated with all drugs or devices that have improved outcomes in multicenter trials.   .   .   .   That means you give everyone an ACE inhibitor, a beta blocker, an ARB, spironolactone, ISDN, biventricular pacing, and maybe an ICD.

Ventricular remodeling in HF, he said, appears to be mediated by beta receptors and receptors for angiotensin, aldosterone, and nitric oxide, but "to think that everyone has a similar degree of involvement of those receptor mechanisms is probably oversimplistic." Drug responses vary by individual, so should all HF patients receive drugs aimed at all of the receptors? It's common, he said, to treat selectively according to the continuous variables of age, LVEF, CAD severity, blood pressure, insulin resistance, cholesterol levels, and HF symptom severity. So, Cohn asked, why not race, which is also a continuous variable despite the superficial distinctions?

"You can say everyone should be treated with all drugs or devices that have improved outcomes in multicenter trials," Cohn said. "That means you give everyone an ACE inhibitor, a beta blocker, an ARB, spironolactone, isosorbide dinitrate-hydralazine, biventricular pacing, and . . . maybe an [implantable defibrillator]. And you have to accept the obvious variability in the benefit, cost, and adverse-event profile of these therapies in different individuals."

Based on the evidence, he said, all white HF patients should be on an ACE inhibitor and a beta blocker with the discretionary addition of an ARB, spironolactone, ISDN-H, or biventricular pacing. In contrast, "I believe it is inappropriate for any black patient not to be given this BiDil combination, probably with an ACE inhibitor and a beta blocker because that is traditional therapy and it does work." Other treatments in blacks would be discretionary, he said.

There is no consistent evidence that race meaningfully influences response to any drug treatment in patients with heart failure.   .   .   .   To suggest otherwise is to suggest that a specific racial group should be deprived of treatments for heart failure that have specifically shown to prolong life.

"There is no consistent evidence that race meaningfully influences response to any drug treatment in patients with heart failure," Packer countered. "My concern is that to suggest otherwise is to suggest that a specific racial group should be deprived of treatments for heart failure that have specifically been shown to prolong life, and I think we can collectively feel very uncomfortable with that."

Surrogate markers for racial characterizations, such as skin color, do not closely reflect genetic variations that may influence drug responses, Packer observed. He then poked holes in the published evidence cited as support for the race-specific efficacy of the three drug classes of proven survival benefit in patients with HF: ACE inhibitors, beta blockers, and direct-acting vasodilators.

A post hoc analysis that was famously interpreted as showing fewer HF hospitalizations among whites than blacks who took enalapril in the combined SOLVD treatment and prevention trials, he said, had statistical biases because it converted a randomized study into a case-control comparison [6]. But a later, more statistically appropriate SOLVD Prevention-trial subgroup analysis showed no variation in outcomes by race [7].

And a subgroup analysis of the BEST trial, which found no overall effect of the beta blocker bucindolol on all-cause mortality, was interpreted as pointing to a significant effect in whites but not in blacks [8]. Outcomes were also analyzed by sex, LVEF, NYHA class, and whether CAD was present.

The doctors have the awareness, they know about the drug, but they're not prescribing it.

The finding of a significant effect by race is suspect because the trial was negative for the prospectively defined primary end point, Packer said. Also, "I think there were so many subgroup analyses that some of these would have been significant by chance alone. . . . There are probably lots of reasons why this was a negative trial, but it hardly seems appropriate to think that race was one of them."

And, he said, there is "no persuasive evidence" that ISDN-H produces different HF outcomes in blacks and whites. The A-HeFT trial evolved from the V-HeFT-1 observation that survival curves for ISDN-H vs placebo separated to show a significant difference among blacks (p=0.04) but not among whites (p=0.47). But, he said, "The real question is not so much whether hydralazine-nitrates work in black patients. The real question is, is the effect in black patients really different from the effect in white patients?" Analysis of interaction by race produces a nonsignificant p value of 0.15, according to Packer.

Following his presentation, Packer disclosed for the audience that in practice, he would not use the combination of nitrates and hydralazine based on race. Rather, he said, he follows the same HF treatment strategy in blacks and whites.

A rough ride for NitroMed

To NitroMed and other BiDil advocates, the tablet containing 20-mg ISDN and 37.5 mg hydralazine, designed for three-times-daily dosing on top of contemporary standard HF drug therapy, is the only pairing of the two drugs with clinical trial evidence supporting efficacy in heart failure.

Jessup agrees. Third-party payers resist BiDil but almost universally cover the generics, she noted, but they have not been tested in HF patients already on beta blockers and ACE inhibitors. Still, BiDil's specific balance of the two drugs can be somewhat matched, at least on a milligram basis, using plain-vanilla pills, she observed. Both are available in at least four dose levels, from 5 mg to 30 mg for isosorbide dinitrate and from 10 mg to 100 mg for hydralazine.

Kenneth M Bate (Source: Nitromed)

What's to stop a manufacturer from selling a generic ISDN-H pill that matches BiDil's drug-to-drug milligram ratio? According to Kenneth M Bate, NitroMed's chief operating officer, "that would be a violation of the patent." The company owns long-term patents that cover the precise milligram ratio and the combination's use in African Americans with HF, he told heart wire .

Moreover, Cohn said at the AHA sessions, "The assumption that the generics will produce the same effect as the fixed combination in BiDil is without any support whatsoever." They may produce comparable effects, but past pharmacokinetic data on the separate drugs suggest they are "far less bioavailable" than what is in BiDil. "So given the vagaries of the pharmacokinetics, I think you cannot assume that the generic drugs, which are very difficult to get into the same combination that is in BiDil, would replicate the effect of A-HeFT."

Jessup said when interviewed, "I think that has really profound implications if it's true. Because then the insurance companies don't have a good excuse for saying they aren't going to pay for BiDil, in African Americans at least."

If it were just cost, the generic use should be off the charts and the proprietary use should be staying flat.

NitroMed has had nothing but an uphill battle selling its only product on the market. "We lost a lot of momentum after launch when the reimbursement coverage was so poor," according to Bate. Doctors didn't prescribe it because by and large, their patients couldn't or wouldn't cover the high copayments, he said. "That was problem number one and it's still an issue that we're dealing with."

Third-party coverage of BiDil has grown, according to Bate, who said his company's research suggests that about half of eligible patients have it now. But it's not reaching even the patients with coverage. "The doctors have the awareness, they know about the drug, but they're not prescribing it."

The company also has several "patient-assistance programs" that provide the drug free or at a greatly diminished cost for many who can't afford it. Bate wouldn't disclose how much of current BiDil use occurs under those programs, saying only that they have been "modestly utilized."

Yancy also observed that cost apparently isn't what's behind BiDil's low usage levels. "If it were just cost, the generic use should be off the charts and the proprietary use should be staying flat." But the numbers show otherwise; combined usage figures for the two separate generic drugs "aren't at the threshold to suggest they're being avidly prescribed," he said.

In fairness, Yancy said when interviewed, the experience with ACE inhibitors and beta blockers—"which, by the way, we continue to have arguments about as to why there's not 100% compliance"—suggests that acceptance of a new treatment strategy, "even for heart failure, a disease with compelling symptoms, is a slow, laborious process measured over not just one or two years but several years or longer."

The process remains slow today, about halfway into BiDil's second year on the market. "There is scant evidence that patients are receiving even the generics," Bate agreed, noting that a substantial increase in generic prescriptions might make a big difference to the company. "Frankly, that would be a victory," he said. "We may not like that the generics are being used, we do not believe they are comparable, but at least you could say doctors were responding to evidence." Then, he said, the company could focus on promoting what they see as BiDil's advantages.

Earlier this year, NitroMed stepped back to reassess its marketing approach. It replaced some top managers and got creative. In October, the company announced it would replace its entire sales force of about 140 people with an all-star team of 20 or so representatives with abundant CV-drug-sales experience, both to save money and refocus its message, according to Bate.

Despite clinical-trial evidence that should be compelling enough to convince clinicians to prescribe BiDil, Bate said, the company's research suggested they were taking their cues from regional opinion leaders. So the elite sales team will aim at leading HF specialists at major centers in the 20 states that encompass about 90% of the African American population. "We would like them to be prescribing more, and we're hopeful that their advice and their practice will trickle down into the community."

The A-HeFT paper states that "Cohn was the inventor on two patents for the use of combinations of hydralazine compounds and isosorbide dinitrate or isosorbide mononitrate in heart failure. In return for equity and potential future royalties, NitroMed licensed from him the patent for mortality reduction of the drug combination in heart failure." Yancy said he has consulting relationships with NitroMed, GlaxoSmithKline (GSK), and AstraZeneca. Jessup said she consults for GSK, Medtronic, Acorn, and Ventracor. Konstam has reported significant grant support and research relationships with GSK, Pfizer, and Merck and is employed as medical director by Orquis Medical, a device company in which he also has ownership interest. Abraham has reported receiving research support and honoraria from Amgen and Scios. Packer reported that he has served as a consultant to NitroMed, AstraZeneca, and GSK.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: