Join 3,363 readers in helping fund MetaFilter (Hide)


Race-based pharmaceuticals?
June 16, 2005 7:36 PM   Subscribe

Pharmaceutical company seeks approval for controversial heart drug BiDil for blacks only, even though the connection of a nitric oxide deficiency to the genetic makeup of the African-American population is an unsubstantiated hunch, in the words of the drug's own developer — and the drug had already once been denied approval. Is the FDA doing good science or ignoring science under the pressure of big business wanting to sell BiDil to a US$1Bln demographic?
posted by Rothko (27 comments total)

 
...not to mention us white guys who think it might make us better dancers.
posted by wendell at 8:38 PM on June 16, 2005


Rothko posted "Is the FDA doing good science or ignoring science"

The FDA's doing good medicine.

BiDil works in blacks, even though the mechanism is unknown. Plenty of drugs work without their mechanism being known. And patients don't care -- they just want something that works.

Doing medicine and doing science involves doing different things: good medicine involves taking an anecdotal, patient-centered approach that would be bad science.
posted by orthogonality at 8:57 PM on June 16, 2005


That is not good medicine, and I would be shocked if it were passed by the FDA. notwithstanding the approval of the panel. There's no way they should rely on any study from the 80s to support a new treatment. Cardiology has moved light years beyond what was happening then, not to mention the process and standards of cardiovascular medicines research.
posted by mikel at 9:11 PM on June 16, 2005


Doing medicine and doing science involves doing different things: good medicine involves taking an anecdotal, patient-centered approach that would be bad science.

Sorry to jump in early, but I can't parse this sentence to understand your distinction. Are you saying that anectodal evidence is a good basis both for rational drug treatment and for the functional mission of the FDA?
posted by Rothko at 9:26 PM on June 16, 2005


There's no way they should rely on any study from the 80s to support a new treatment.

The 1,050-patient study was initiated in June 2001 and halted July 19, 2004 (from last link)
posted by Edible Energy at 9:57 PM on June 16, 2005


I thought we were all equal?

Recently I was involved in a very interesting discussion with a very intelligent person on this science forum. He argued until he was blue in the face that there is absolutely no difference in genes between the races.

(It started off as a discussion on why people of West African decent dominate sprinting events – he maintains that it's not genetic)

Er, I duuno where I'm headed with this post. Here's some Richard Dawkins fo yo asses.

"It is an article of passionate faith among "politically correct" biologists and anthropologists that brain size has no connection with intelligence; that intelligence has nothing to do with genes; and that genes are probably nasty fascist things anyway."
posted by uncanny hengeman at 10:20 PM on June 16, 2005


There was considerable effort made by the government to get them hooked on crack.

Infecting them with laboratory AIDS was no small effort.

Do you think they'll just hand over cardiovascular health after all that work?




(I just hope all your sarcasm detectors are turned up extra sensitive...)
posted by sourwookie at 12:05 AM on June 17, 2005


mikel - That is not good medicine, and I would be shocked if it were passed by the FDA

The FDA has nothing to do with *good medicine" but in approving drugs that have shown equal or better efficacy than drugs already on market. Equal or lesser side effects are a secondary consideration. It is a *known* but lesser explored (for a variety if reason)s fact that there are ethnicity-based differences (epidemiologigically) in pharmaceutical responses.

orthagonalit - Doing medicine and doing science involves doing different things

Yeah, you have it right, but it doesn't necessarily support your position - doctors try whatever they thing might help. In experimental medicine, most docs have no idea why it works; they *think* a treatment might help, and sometimes it does, but fuck, a lot of time they're just talking out of their asses. When it works, great. When it doesn't, it's "The pathophysiology is complex and incompletely understood. We tried our best with current state of knowledge."

Fuck self-aggrandizing doctors (sorry to Mefi Docs; if it makes a diff, I haven't run into anyboddy here who fits the description) who think they know whats going on - it's fucking scientists who clean up after the docs - figuring out *why* something works or figuring out *how* to make something work to alleviate a disease.
posted by PurplePorpoise at 12:42 AM on June 17, 2005


The 1,050-patient study was initiated in June 2001 and halted July 19, 2004

They left out that the study was conducted on blacks only however. Quite easy to conclude that a drug works on a certain race if it's the only race tested. Quite the joke to then claim it works best for that race having not run similar tests on others.

Science is being ignored to a certain extent, but not by the FDA in this case.

From the CBC:

Others consider Cohn's explanation to be overly simplistic, noting since BiDil was tested only on African-Americans, the company can't say whether it works in other populations.

and:

Although there are diseases that occur more often among some races, such as sickle cell anemia in blacks or Tay-Sachs in Ashkenazi Jews, these are genetic, not racial disorders. The diseases aren't exclusive to one race.
posted by juiceCake at 4:52 AM on June 17, 2005


The genetic differences between "races" is marginal, or perhaps better described, there is so much overlap in the mix of genes among races in America, the distinction is muddied so that making pharmaceutical decisions based on these distinctions will lead to bad medicine.

There are legitimate studies in pharmacogenetics and people who carry specific genes can respond differently to some drugs (although relatively few). But who carries those genes is not well defined by race. If one was to say, I have this genetic test and a person is carrying this gene, then this is the drug for that person, that would make sense.

The FDA regularly approves marginally beneficial drugs (or non-beneficial drugs). Depending on the disease, they don't have to prove they are as good or better than what's out there, just that they are better than placebo.

There is a huge economic incentive to get a failed drug reconsidered for some (any) subgroup for which it showed efficacy. Recently several drugs are approved for women-only or even studied in women only -- even though the diseases are common to men. The reason for this is trying to salvage a failed drug by looking for a pocket of people that show statistical differences. Often these differences appear just because the analysis is shaved in so many ways until it does appear.

Then you play the race card, say the FDA is trying to prevent my (wonderful) drug from helping a group.

Not surprisingly, it's about the money.
posted by dances_with_sneetches at 6:46 AM on June 17, 2005


The future in pharmaceutical science lies in drugs adapted to differing genetic make-ups. A doctor, or perhaps the pharmacist, will examine your DNA profile to select an appropriate drug or variant of a main drug which suits your genetic profile. It will lessen side effects and perhaps enhance efficacy. This current controversy is but the tip of the iceberg.
posted by caddis at 6:58 AM on June 17, 2005


Huh, strange.

It seems strange to me that anyone would claim that "good science" and "good medicine" are diffrent things. I certanly would not want to be treated by someone who didn't belive in science or the scientific method.

That said, not knowing how something works is not the same thing, from a scientific standpoint, as knowing that it does work. I mean, Scientists don't even know how gravity works.

For a long time Scientists didn't know how Marijuana affected the brain. Does that mean that at the time smoking weed in order to get high was a bad idea? Does that mean that once endocanabanoids were discovered marijuana suddenly started getting people high? Not hardly.

---

Look, you do a study and if there's a corrilation between African decent and being positivly affected by a drug, then that is science. It means that there is some genetic marker, some unknown combination of genes that makes this drug more effective.

That unknown combination is belived to be more common in Blacks, and the makers have shown that Given a black person, this drug is likely to help them.

I don't see what, exactly, makes that 'unscientific'. You don't know why something happens, only that it does happen.

---

There are probably a lot of black people who would not be helped by this drug, and a lot of white people who would be. Hopefully some day they can devise a real genetic test that would tell you which drugs will work and which won't.
posted by delmoi at 9:26 AM on June 17, 2005


They left out that the study was conducted on blacks only however.

The large-scale, Phase 3 clinical trial was only on blacks. Earlier, smaller clinical trials (Phase 1 and 2) had shown benefits in black patients, but no statistically significant improvement in other races. And if you can't show a benefit after Phase 2, you can't go forward into Phase 3.

The FDA regularly approves marginally beneficial drugs (or non-beneficial drugs). Depending on the disease, they don't have to prove they are as good or better than what's out there, just that they are better than placebo.

This is a little true, but phrased in a totally false way. You only have to prove "better than placebo" if NO alternate treatment exists for the condition. Because in that case, placebo is the only alternate treatment. In EVERY OTHER CASE, you have to compare against "what's out there," and your drug will not be approved if it is not better in some way than existing treatments (although better can mean a lot of things).

For example, you might get a drug approved which is statistically equally or less effective than existing treatments, if it works by a different mechanism, to provide an alternative treatment for patients who are non-responsive to current treatments. You'd be hard-pressed to paint that as unreasonable.

Also, what PurplPorpoise and Caddis said.
posted by iron chef morimoto at 9:30 AM on June 17, 2005


Also, if you think BiDil is controversial, you're an tard. Sorry, but that's just the way it is.
posted by iron chef morimoto at 9:50 AM on June 17, 2005


An "an tard," of course, being that very special sort of tard who fails to proof-read before posting.
posted by iron chef morimoto at 9:51 AM on June 17, 2005


Genetics based treatment of individuals is the future of medicine. Ask any serious medical researcher what a SNP is. Standing in the way of this is the politically correct thinking of half educated blowhards.

That said, targeting Africans as a "race" is amusing. As a group Africans are anthropologically older and consequently have greater single nucleotide polymorphisms (differences) within their group than other "races" that emerged as narrow gene pools from Africa more recently. How hilarious is it that a "race" targeted drug is aimed at the least homogenous "race"?
posted by paleocon at 12:17 PM on June 17, 2005


Standing in the way of this is the politically correct thinking of half educated blowhards.

Or perhaps the question is how drugs are being approved, given the science for the drug in question is incomplete and possibly misleading.

Nowhere in this post did I suggest that genetics-based treatments are not useful (see the Tay-Sachs quote, for example, or any other gene therapy research), but it's unclear that blacks are genetically homogenous enough for this kind of blanket classification by the FDA, and it raises the question of why this controversial classification should be allowed, and why a particular drug should be approved, when the science didn't support it the first time around, and there is significant profit to be made from its approval.

Also, if you think BiDil is controversial, you're an tard.

If mainstream medical journals believe it to be controversial to assign drug approval based on hazy racial classification, does that make mainstream medical journal editors "tards" as well?
posted by Rothko at 12:44 PM on June 17, 2005


To respond to iron chief

The large-scale, Phase 3 clinical trial was only on blacks. Earlier, smaller clinical trials (Phase 1 and 2) had shown benefits in black patients, but no statistically significant improvement in other races. And if you can't show a benefit after Phase 2, you can't go forward into Phase 3.

First of all, Phase 1 trials can not show benefit by definition. Phase 1 trials use non-diseased subjects. Secondly, phase 2 trials do not require statistical proof of benefit in order to go to phase 3. They can be used to determine the necessary sample size of phase 3 trials in order to determine statistical significance at that time.

The phase 3 trial(s) is the place where you get the statistical significance that it has efficacy. And no you don't have to prove that it is better than what's on the market. There is something nice about the notion that each new drug is onwards and upwards better, but that standard would not be enforceable.

When I said the race card is played, I was suggesting that the political correctness card is used in favor of race-based drug approvals.
posted by dances_with_sneetches at 1:31 PM on June 17, 2005


Rothko:

1. Black people are at very high risk for heart failure, more so than any other race in the US. Statistically provable

2. One of the major causes of this is low nitric oxide levels, which is common in black patients but very rare in the general population. Many drugs on the market are ineffective in treating black heart failure patients. Statistically provable

3. BiDil is an effective treatment for low nitric oxide levels. It is therefore effective in most black patients, but not in most of the general public. Statistically provable

Conclusion: As a hard and fast rule,* BiDil should be prescribed to most black heart-failure patients, who will respond to it, and not to most other patients, who almost certainly will not respond to it.

*Obviously, doctors don't have to give the drug to black patients who don't have low nitric oxide levels, and they can off-label prescribe it to non-black patients who do have low nitric oxide levels. Hopefully you're rich and have a good doctor, but that's a separate issue.

What, exactly, is the problem here? That black people tend to respond differently to different heart failure drugs? That the drug isn't 100% effective in black people and 100% ineffective in everyone else? That there's disagreement over what "black" means? Where, exactly, is our disagreement?
posted by iron chef morimoto at 1:37 PM on June 17, 2005


dances:

Phase 1 trials use non-diseased subjects

This is how the clinical trial were designed originally, but this not always true any more. You're probably right with regards to BiDil, however. I withdraw that statement.

...phase 2 trials do not require statistical proof of benefit in order to go to phase 3.

You'd really have to warp that statement around to make it true. Phase 2 studies set the effective dose, which is calculated by comparing side effects vs. effectiveness. Obviously, you need to see strong evidence of effectiveness. Statistical proof? Maybe, maybe not.

The phase 3 trial(s) is the place where you get the statistical significance that it has efficacy. And no you don't have to prove that it is better than what's on the market.

Big, bold, type: Yes, you do. Unless you're talking about a baldness cure. If you are studying a drug to be given to patients with a life-threatening disease, the FIRST QUESTION the FDA will ask you, before you even enter Phase 1, is "Why is this drug better than what's already on the market?" And you need an answer. "It's more effective. It has less side effects. It looks like it will work in black patients who aren't responding to conventional treatment." And you have to prove that argument in your clinical trials.

Whatever. "Better" is a malleable term, but you need to make that exact argument. That it's better. There are lots of drugs that have never, and will never, enter into clinical trials because the current treatments are extremely effective.

And my point (edited a little) is still the same: In the Phase 2 trials, BiDil showed effectiveness in black patients, but not in other races. And considering the mechanism it works by, that makes perfect sense.
posted by iron chef morimoto at 2:03 PM on June 17, 2005


I'm studying for CRA certification right now, so we can argue this until you guys get bored. It's like a nice little review for me.
posted by iron chef morimoto at 2:26 PM on June 17, 2005


...so we can argue this until you guys get bored.

No, but thank you for offering. Points 1 and 2 are irrelevant and the third point does not follow from the data (correlation is not causation). In any case, the 'tard' comment was sufficient for me.
posted by Rothko at 2:39 PM on June 17, 2005


Third point does not follow the data? Correlation is not causation? Fine. Whatever. Rephrase it as such:

BiDil is an effective treatment for low nitric oxide levels. Statistically proven It is therefore effective in most black patients, but not in most of the general public. Statistically proven

Seriously, the clinical trials results should be posted online somewhere (they're supposed to be "made available to the public," and that's the most common method used to disseminate them these days). I'll try to post a link some time tonight.

I can't believe that with all the perfectly legitimate complaints to be made against pharmaceutical companies, people are trying to make a controversy out of BiDil. You have me agreeing with someone posting under the name "paleocon," for chrissakes.
posted by iron chef morimoto at 2:56 PM on June 17, 2005


3. BiDil is an effective treatment for low nitric oxide levels. It is therefore effective in most black patients, but not in most of the general public. Statistically provable

No statistically significant proof exists for that statement. The earlier test which found some difference had too small of a sample size to be statistically relevant. The later, larger study had only blacks, at least as those patients identified themselves. It showed a statistical difference. Might it work just as well outside of the black population? It probably would but since the study was restricted to blacks who knows. Because the study was so limited, and for no other reason, there is a push to label the drug as effective for blacks. Of course, once it is approved an on the market doctors can prescribe for whomever they want and for whatever they want. What labelling it will receive is still in question. It may very well get a labelling that approves it for hear failure with a note to the study being conducted on self identifying blacks. No one knows for sure at this point if the drug works better in one racial population over another. Further, as has been pointed out above, the "black" population is hardly homogeneous.
posted by caddis at 5:43 PM on June 17, 2005


I can't find the results of the first clinical trials at any of the major clinical resources, but there was a casual mention that, for the patients in the Phase 2 study, BiDil showed "only 7% improvement over placebo." I don't know if that refers to patient survival or what, it assume it refers to a general population sample. But it was considered too low, by the FDA, to enter a Phase 3 trial.

For the most recent trial, targeting African-Americans:
Trial results demonstrate that the drug met its primary efficacy endpoints, with a 43% improvement in patient survival (p=0.01), a lower sudden death rate (6.2% vs. 10.2%; p=0.02), a 33% reduction in first hospitalization for heart failure (p=0.001), and a significant improvement in quality of life (p=0.02), vs. placebo.

When I studied BiDil a couple of months ago, the description of the first clinical trials was that they had shown significant improvement in a few patients, but no change or a decline was seen in most of the patients tested. And since there was reason to believe that BiDil shouldn't be used in conjunction with other heart failure drugs, which were more effective, the trial was terminated before Phase 3.

Logically, the problem here was predicting which patients would benefit from BiDil, and which patients would not. From animal studies, it has been shown that BiDil increased nitric oxide levels, and dilated blood vessels. People of African descent are uniquely susceptible to low levels of nitric oxide, which is believed to be the principle reason that black patients with heart failure die about 50% faster on average than white patients in the US, even after controlling for factors such as diet, income, etc.

If NitroMed thought their drug would work well enough on the general population to pass another trial, they wouldn't have targeted only African Americans. As it is, they can only market it to 12% of the US population (and yes, the FDA is cracking down on marketing for off-label uses). And most physicians won't try it on non-black patients, except maybe as a last-resort.

If NitroMed is some sort of evil corporation, they're the most incompetent one I've ever seen.
posted by iron chef morimoto at 7:46 PM on June 17, 2005


And most physicians won't try it on non-black patients, except maybe as a last-resort.

Right. We'll see how that works out.

This drug may very well work better in the black population, and even for the reasons you state. Statistically, we have no proof of that yet. If I were a betting man, I would say you may be right, but then again, the proof . . . .
posted by caddis at 10:17 PM on June 17, 2005


studying for CRA? Clinical Research Associate? You're not getting a job in my lab. Beyond the mind-boggling lack of understanding of the drug approval process, you have a really snotty arrogance about your misinformation.
posted by dances_with_sneetches at 1:42 PM on June 18, 2005


« Older InstantSOUP is good for the android's soul...  |  Outcasts in Their Own Villages... Newer »


This thread has been archived and is closed to new comments