The heart drug BiDil exceeded all expectations when tested in African Americans.© Photodisc

A heart drug being tested in black patients is on course to become the first medicine approved for use in a specific ethnic group, challenging those scientists who believe that race is a bad basis for prescriptions.

The drug, made by Massachusetts-based pharmaceutical company NitroMed, was abandoned after a trial in the 1980s produced unimpressive results. But, because the data hinted at differences between white and black patients' responses, in 2001 NitroMed decided to carry out a further clinical trial using only African Americans.

This week NitroMed announced that the trial, in over 1,000 black heart-failure patients, has been stopped early because it appears so effective when used on top of normal therapy. "I'm so thrilled about it," says study leader Anne Taylor of the University of Minnesota, Minneapolis. If the drug, called BiDil, receives regulatory approval, the company says it will aim to launch it in early 2005.

But BiDil revives controversy about whether, and how, race should be used to prescribe medicines. In the clinic, for example, doctors will have to work out who is classed as African American in a racially mixed population. "It really becomes problematic," says Sandra Soo-Jin Lee, an anthropologist who studies race in science at Stanford University, California.

Skin deep

Doctors have long known that different ethnic populations can have different susceptibility to diseases or react differently to drugs. Drug labels for the common heart drugs called ACE inhibitors note that they may be less effective in black people.

BiDil contains two generic medicines that together boost production of nitric oxide, a molecule that relaxes blood vessels and eases strain on the pumping heart. Taylor believes that African Americans, who have a higher rate of heart disease, react better because they tend to have lower levels of nitric oxide than other ethnic groups.

Some scientists argue that race is a poor way of guessing a person's response to a drug. "I still think skin pigment is a lousy predictor of heart function," says Howard McLeod of Washington University in St Louis, Missouri.

McLeod argues that it is better to identify the one or more genetic variations that control whether the body reacts well to a drug and prescribe it to those people, regardless of race. In the case of BiDil, for example, the genetic difference responsible probably occurs more commonly in the African American population. But the same genetic difference could exist, at a lower frequency, in Caucasian, Asian or other ethnic groups.

Surrogate markers

The genetic argument was backed up by a study in 2001, in which British researchers divided a population into four different groups based on 40 genetic markers. They found that these groups were a better predictor of drug response than ethnic ones¹.

Taylor acknowledges these points and says that she plans to scan the genes of those patients who responded to BiDil to find those that will foretell a patient's response. Until these results are in, Taylor argues that race may serve as a reasonable surrogate for making prescriptions, when used alongside a patient's medical history. "We hold the trump card: it works," she says.

Future tests could show that other ethnic groups respond to the drug when it is given, as in the current trial, alongside conventional treatment. But Taylor is urging the traditionally white-dominated clinical trials to incorporate patients from all ethnicities now, so that differences in their responses can be picked up from the start. 

Fritz Haber Institute of the Max Planck Society, Berlin

University of Minnesota, Minneapolis

• References

  1. Wilson J.F. et al.Nature Genetics, 29. 265 - 269 doi:10.1038/ng761 (2001). | Article | PubMed | ISI | ChemPort |