ON JULY 19 a clinical trial involving more than 1,000 patients with heart disease was halted prematurely. A phase III trial, as this was, is usually the final, expensive step before a therapy is licensed or canned, and an early end can spell very bad or very good news.

In this case, the news was spectacularly good. The data on BiDil, a heart drug developed by a Massachusetts company called NitroMed, had become “so compelling”, according to those steering the trials, that it had become unethical to withhold the medication from the patients receiving a placebo, as opposed to BiDil, on top of their existing medication.

But there was another factor that marked this particular investigation as extraordinary. Every single patient in the BiDil experiment, held at 170 medical centres throughout America, was black. A Food and Drugs Administration licence will mean that BiDil can be prescribed only to black patients. BiDil is thus likely to become the first “ethnic medicine”, a therapy designed to be prescribed to one race, in this case African-Americans, with moderate to severe heart failure.

The drug is now a leading protagonist in the debate over what role, if any, race should play in medicine, and whether frontline health workers really should judge patients by the colour of their skin. The topic is a hot one: only last week, Nature Genetics revealed research from University College London showing that 29 medicines have safety or efficacy profiles that vary between ethnic or racial groups. The list is peppered with familiar names: the anticoagulant warfarin, for which ideal dosage differs across ethnicity; interferon, which shows a lesser response in African Americans; insulin, to which Hispanics and African Americans show lower sensitivity than Europeans.

BiDil has been welcomed by some as a dramatic way of bridging the health crevasse that separates whites from non-whites — African-Americans are twice as likely to suffer heart failure than white Americans, and more likely to die from it. The Association of Black Cardiologists in America, for example, supported Nitro-Med’s study, and has embraced BiDil as a potential corrective to the disturbingly high death rates among African-Americans with certain types of heart disease.

The trial results are due to be disclosed at the meeting of the American Heart Association in New Orleans in early November.

Dr Michael D. Loberg, president of NitroMed, is optimistic that an FDA licence will be granted next year. “It is in the FDA’s hands now but we are very excited. If you could see me now, you would see that I am smiling.” Others, however, view the BiDil study as misguided, cynical or downright dangerous. Their objections are many. The palette of skin colours that falls under the umbrella of “black” is expansive, and so colour may not be a reliable indicator of underlying health. So while ancestry appears to be important genetically — meaning that certain ethnic groups are particularly afflicted by certain diseases — skin colour can be a misleading way of measuring ancestry.

And while chemically plugging the health gap that undeniably exists between races seems like a good idea, it risks diverting attention away from the complex reasons — for example, dietary, environmental and socio-economic — why the gap has opened up in the first place.

A more sinister outcome is that, potentially, it endorses the view that races are biologically different, a view which is, genetically speaking, fallacious, and, politically speaking, explosive. Dr Jonathan Kahn, assistant professor of law at Hamline University, Minnesota, and a bioethicist, recently published a highly critical analysis of BiDil’s history and its implications in the Yale Journal of Health Policy, Law, and Ethics.

Kahn believes that the repackaging of BiDil as an ethnic drug was driven by legal and commercial, rather than biomedical, concerns (the formulation has been around for 20 years and was previously rejected by the FDA as ineffective for patients in general). Kahn says that NitroMed’s failure to test it in whites, and establish whether it does indeed act differently, constitutes bad science.

“It’s great that the drug looks so promising but this trial was carried out only with African-Americans and so tells you nothing about how BiDil works in anyone else,” Kahn says. “Even though NitroMed is seeking race approval for BiDil, there’s nothing in this trial that says it doesn’t work in other groups.

“It lends credence to the idea that blacks and whites are genetically different. We know that with regard to genetics, race is a sloppy construct.”

Kahn says he is not making a stance for colour-blind medicine — after all, certain conditions do appear to cluster along racial lines, such as hypertension and heart disease — but that greater care is needed to establish that assumed biological difference do genuinely exist. The danger, Kahn says, is that “it’s easier to fix molecules than social inequality”.