FDA approves first 'ethnic medicine'

The FDA, as expected, has followed the recommendation of an advisory panel last week and approved the heart drug BiDil, which clinical data have shown works well in African-Americans but poorly in other populations.

The drug, produced by Massachusetts-based NitroMed, leapt to prominence last year when a trial of 1050 African-American patients was halted prematurely, with results showing a dramatic 43 per cent reduction in mortality. An earlier trial in a mixed population, primarily of European descent, had shown little positive impact.

BiDil's approval is controversial on several accounts. The drug is targeted strongly at a single ethnic group, marking an important step towards an era of more personalised medicine. Medical data suggest that African-Americans generally have lower levels of nitric oxide, increasing the drug's efficacy. Critics, including The New York Times' editorial page, point out that NitroMed was able to extend the drug's patent some 13 years, to 2020, based on its FDA-sanctioned decision to target a specific population group. Some analysts predict that BiDil could become a blockbuster (US$1 billion in sales) by 2010, although the price of the drug -- a combination of two generically available drugs -- has not been announced.

The drug has also come in for criticism for giving credence to biological notions of race. Some prominent geneticists argued that the FDA should postpone approval until more sophisticated pharmacogenetic tests could be devised to identify which patients would best respond to the therapy.

The drug label will note that BiDil is intended for patients who self-identify themselves as black. However, members of the FDA regulatory panel argued that there is reason to think that it should also work for some patients of other ethnic backgrounds. There is a good chance it will be prescribed for non African-Americans. The FDA's associate director, Robert Temple, said, "In the future, we hope to discover characteristics that identify people of any race who might be helped by BiDil."

NitroMed says it has a sales force of nearly 200 representatives in place, and is ready to market the drug almost immediately.

Interview with NitroMed chief medical officer Manuel Worcel

Late last year, Bio-IT World editor-in-chief Kevin Davies interviewed NitroMed's chief medical officer (and former president and CEO) Manuel Worcel about the prospects for, and controversy surrounding, the potential approval of BiDil. Here are some extracts of that conversation.

Bio-IT World: How did NitroMed acquire this drug? Worcel: NitroMed has the objective to develop drugs that either improve safety or efficacy by adding nitric oxide to the molecule. In 1998, when Dr [Jay] Cohn was looking for a follow-up developing BiDil... he came naturally to us. We discussed the lessons to learn from existing data on the efficacy and safety of BiDil in the patient population. Very early in the game, we understood the effects observed in the overall population were determined in many ways by a specific beneficial response in patients self-designated as African-American.

What is your view of personalised medicine? It has been obvious for many years to cardiologists that drugs that are currently used for treating cardiovascular disease, hypertension, and heart failure, do not appear to have the same effect in all patients across all populations. There's a wide consensus -- and we have data showing -- that African-American patients respond in a different way to drugs like renin-angiotensin antagonists and beta-blockers, which are widely used to treat hypertension and heart failure. That said, when you have to deal with the treatment of heart failure, you cannot determine what is the response of any individual patient to the treatment of heart failure... In hypertension, I can determine if my patient responds to my prescription by measuring blood pressure. In heart failure, I'm not going to know individually if the drug works, because I'm looking for clinical endpoints like survival, days in the hospital, and so on.

What happened in last year's AHeFT African-American trial? The data safety monitoring board recommended very strongly that we stop the trial because there was a clear-cut benefit in patients treated with BiDil, a very clear reduction in mortality. After discussions with the FDA, we announced we were stopping the trial.

Why were such benefits not seen in earlier trials? Did the lower efficacy of the drug in people of European descent mask the positive benefits in African-Americans? This is the right way to put it. There was a benefit in the overall population that did not attain statistical significance... The reanalysis of mortality data shows that there's an improvement in mortality of 47 per cent, which is very close to what we saw in the AHeFT trial.

How would you characterise the support of the Congressional Black Caucus? They supported us in many press releases, saying that they welcome the fact that for the very first time, we were obtaining important information that would guide the future treatment of African-American patients. The Association of Black Cardiologists co-sponsored the trial with NitroMed, they've been extremely helpful in helping us identify principal investigators and also in refining the message that we send to the clinical centres to explain the motivation for the trial.

Is there a sharp distinction between the response to BiDil in African-Americans and in European-Americans? I believe that not all African-American patients respond to the drug, and I believe that a subset of non-African-Americans, whites, Asians, whomever, do respond to BiDil. In the very near future, we're going to have data that may allow us to see more closely the genotype and the phenotype of the responders to BiDil. Another clinical trial we've been running is called GRAHF (genetic risk assessment in heart failure), which is going to analyse the phenotype of different markers for cardiovascular disease. We have close to 400 patients, and the DNA is being analysed. We're going to link these results with the data we obtained in AHeFT to give a very powerful database.

Beyond that, we need to explore possibilities of finding other BiDil markers for response. There are very strong suggestions that a proportion of African-American patients have nitric oxide deficiency. We are working to identify other physiological biomarkers beyond the genetics -- blood pressure, heart size, response to vasodilation stimuli. We are building a Phase IV plan to identify what would be the genotype characteristics to go beyond skin colour that we all admit is a very imperfect surrogate of NO deficiency, and allow us to test in the future BiDil in other populations and narrow down the specificity of response in African-Americans.

Do you feel any concerns about BiDil being marketed based on something as superficial as skin colour? No, not at all. I understand the concerns -- I would be absolutely against any use of valuable data for African-American patients to conclude anything regarding historic values of African-Americans. Being African-American is about as diverse as being white. If you are a Caucasian from Northern Europe, compared to the Mediterranean, there are differences in occurrence of many diseases.

What are the reasons why a particular disease occurs more frequently in a particular ethnic group? Genetics may play a role, but not all the roles. It is clear that African-Americans present cardiovascular disease and heart failure in a different way. There may be a genetic predisposition, but other environmental factors impact on the phenotype. If my patient smokes, or is overweight, or has a metabolic disease, or bad dietary habits, or is underserved by the healthcare system because the patient is poor, and arrives to medical centres when disease has evolved too far to be prevented -- all these contribute to when the disease presents.

Should you conduct further tests to find pharmacogenetic markers of response before marketing BiDil? Doing [that] would be to postpone the use of BiDil that improves survival by 43 per cent in African-American patients, and hospitalisation by 40 per cent! And wait for how long? I hope that pharmacogenetic testing will happen soon, but it would be a disservice to my patients to wait for theoretical reasons... We all agree that pharmacogenomics could improve clinical testing of patients, but there is a lot of work to do -- we are not yet there.

Some African-Americans reportedly associate a black drug with the notorious Tuskegee experiment. How are you going to educate and reassure the African-American population that you're doing this for the right reasons?We've dealt with this Tuskegee question for the last four years. We approach African-American patients with the help of black doctors and we're very direct explaining to our patients the context of our trial, the risks and benefits. I have to say, we enrolled 1656 patients in this trial and more than 10,000 patients wanted to participate in the trial... I cannot quote a better approval of acceptance than that number.

One reason why the AHeFT trial is exceptional is because we enrolled 1050 patients. You know how many African-American patients have been enrolled in other heart failure trials? No more than 2800, most are in AHeFT. Most heart drugs have been approved from international trials that do not include African-American patients.