Why Race-Based Medicine is a Bad Idea

by Adriane Fugh-Berman, M.D.
The FDA recently approved BilDil for treatment of congestive heart failure in African-Americans, making it the first therapy approved for a specific racial group. There are a lot of things wrong with this decision, which has implications beyond racial profiling in medicine.
First, a bit of background. BilDil is a combination of two older, generic drugs: hydralazine, a high blood pressure medicine; and isosorbide nitrate, which relaxes arteries’ muscular walls, allowing more blood flow through arteries narrowed by heart disease. Until the 1990s, the combination was commonly used to treat congestive heart failure (heart muscle weakness that can follow a heart attack). A trial published in 1986, the Vasodilator Heart Failure Trial (VHeFT1), found hydral-azine/isosorbide nitrate (H/IN) therapy was barely superior to placebo; another drug called prazosin, however, was no better than placebo. In 1991, another trial (VHeFT2) compared H/IN to enalapril, a drug in the class called ACE inhibitors. Enalapril was superior to the isosorbide nitrate/hydralazine combination for prolonging survival, and ACE inhibitors subsequently became the primary treatment for heart failure. Some patients, however, cannot tolerate ACE inhibitors and are treated with H/IN.
One of the VHeFT investigators, Dr. Jay Cohn, patented a fixed combination of H/IN, now known as BilDil (the individual drugs that comprise BilDil were available generically, but it’s possible to patent a combination of older drugs in a new dosage form). A new drug application was filed with the FDA, but the FDA rejected BilDil as a treatment for congestive heart failure in 1997, because the application depended on the results from VHeFT 1, a decade-old trial with problematic outcomes. Drug patents are only good for 20 years after being filed, so rejection for drug approval, even if the drug is eventually approved, cuts years off its profit potential. The company that sponsored BilDil gave up on it. NitroMed, a biotechnology company, obtained intellectual property rights for BilDil from Cohn, and got a new patent specifically on the use of Bildil in African-Americans. Getting a new patent reset the patent clock.
Commentary in the New England Journal of Medicine notes that: ‘This patent, the first ever granted to a preexisting drug for a new, race-specific use, pushes back potential market entry by generic sellers of the fixed-dose combination from 2007 to 2020. Less than a month later, NitroMed went public, raising $66 million (even though isosorbide dinitrate and hydral-azine are available separately in generic formulations, making it possible to closely approximate NitroMed’s combination at a cost of about 44 cents per dose).’
BilDil was repackaged as an ethnic treatment based on a reanalysis of VHeFT2, in which 27% of the original group was African-American. Analyzing the subgroup separately indicated that some African-Americans did better on Bildil than enalapril. A new trial, the African-American Heart Failure Trial (A-HeFT), published in 2004, found that adding I/HN to standard therapy in African-American patients with heart failure substantially reduced deaths, reduced hospitalizations, and improved quality of life. It is not known whether adding I/HN to standard therapy would improve outcomes in other populations, as this hasn’t been tested. In 2005, BilDil was approved for the treatment of CHF in African-Americans. NitroMed plans to charge $1.80 a pill, about four times as much as taking the generic pills. Perhaps to deflect criticism, NitroMed will offer a discount program for those who can’t afford it.
This approval creates a bad precedent, and does not improve patient care. There is no biological basis for race. ‘Race’, whether self-identified or inferred by skin color, is only the roughest of surrogate measures for genetic differences. Some diseases and conditions are more common in one or another ethnic group, but there are no diseases or conditions that always include (or exclude) all members of any ethnic group. There is more variability in genetic differences within any population than there is among racial/ethnic populations.
Identification of a therapy as a African-American drug may cause doctors to automatically prescribe the drug to African-American patients — and withhold it from other patients — without taking into account the patient’s individual needs and responses. The only trial to date designed to look at outcomes in African-American patients found that BilDil was beneficial as an added treatment in those who were already on standard therapy for congestive heart failure.
Although it is often stated that ACE inhibitors don’t work as well in African Americans, a meta-analysis prepared for the Agency for Health Care Research and Quality examined the three placebo-controlled trials of ACE inhibitors stratified by race and concluded there was no mortality difference in the use of ACE inhibitors between African-Americans and Caucasians. There is no reliable evidence that BilDil as a single therapy is better than an ACE inhibitor in African-American patients, but promotion of BilDil may cause docs to prescribe BilDil alone.
BilDil’s approval is bad precedent. We can now expect drug companies scrambling to analyze both old and new clinical trials for subgroup benefit that can give new life to failed drugs. It has been said that if you torture data enough, they will tell you anything. Can it be long until a drug is approved for use in nonsmoking female poodle-owners between the ages of 27 and 33?
Adriane Fugh-Berman, M.D., is an associate professor in the department of physiology at Georgetown University School of Medicine, and a former chair of the NWHN.