« Were The Democrats Weak Or Warlike? | Main | Four Congressmen »

January 18, 2007

Objectively Pro-Cancer

Digby lights on the sort of story that makes my blood boil:

It is expected there would be no problems securing funding to explore a drug that could shrink cancerous tumors and has no side-effects in humans, but University of Alberta researcher Evangelos Michelakis has hit a stalemate with the private sector who would normally fund such a venture.

Michelakis' drug is none other than dichloroacetate (DCA), a drug which cannot be patented and costs pennies to make.

It's no wonder he can't secure the $400-600 million needed to conduct human trials with the medicine - the drug doesn't have the potential to make enough money.

Michelakis told reporters they will be applying to public agencies for funding, as pharmaceuticals are reluctant to pick up the drug

As I've said before, using the government's monopsony and formulary powers to radically reduce the cost of pharmaceuticals probably won't hurt private R&D much (see here for another example of that). But even if it did, pumping those same savings into public sector research and development would more than balance the scales. So much of the genuinely original molecular research is being done on the NIH's dime in university labs -- but the NIH is underfunded, and so passing on promising and/or risky applications. Pumping cash into their coffers would be a wise, wise investment, and would open promising avenues for research that the drug companies won't currently pursue.

January 18, 2007 | Permalink


Just a note of caution, Ezra-- after going down the cancer path with a parent, I learned that there were a lot of dead ends in cancer research, and that some of them stalled for a variety of reasons (all gene therapy trials had been halted when Mom was sick, for example, due to that rash of deaths that got so much press). The observer of politics and literature in me likes to believe that a simple answer is right beneath our noses, but the battle-scarred relative is a lot more wary, IOW.

Posted by: latts | Jan 18, 2007 3:14:54 PM

I want to talk more on the effects of the proprietary nature of cancer clinical trials when the private sector is involved. You have university scientists and doctors at major research centers who should be sharing their research and steering patients who meet criteria for other protocals different from the one's of which they are part, not doing so. I have never done research to figure out whether this hampers the ability of the scientists to work toward better outcomes, but I am definitely curious after reading your post about how wide spread this issue is.

Posted by: akaison | Jan 18, 2007 3:20:07 PM


you make a really good point. most of these drugs prove to be only marginally effective, but I don't think your caution changes the nature of the deeper question- namely does proprietary interests hinder the maximazing of research outcomes? I also had a parent and had to deal with this issue. There were several things I learned about the medical system, and clinical trial protocals from this experience:

a) Doctors wanted to steer us away from protocals by others,a nd toward their own. They even provided older data to convince us.

b) Make sure the pay attention to health insurance. don't get sick before you paid up for a premium that month or else they will try to claim that you haven't made payment.

c) the medical administrative process is extremely disorganized. Sharing of recors, doctors talking to each other, etc are all an up hill battle.

d) race matters. Even if you have the coverage, this will be a factor in treatment because of perceptions of race. There is a Harvard Medical Journal article on this subject i think.

e) Gender matters. Same as d.

f) Have a clear plan BEFORE you get sick for powers of attorney, and other legal issues or you can spin weeks spinning in circles.

g) Just because a doctor is renowned, don't let that be a substitute for understanding what the doctors are doing or not doing. In our case, a failure to try chemotherapy because despite stronger lividity.

these are just a few lessons learned the hard way. but definitely one of the big one's is that the research side of the the profession is highly proprietary.

Posted by: akaison | Jan 18, 2007 3:29:02 PM

I think the $400-$600 million that is often estimated for clinical trials of things like this is overstated in this case, if we truly already know it has no side effects in humans (avoiding stage I clinical trials and making it much easier to recruit people for later trials) and is easy to make in bulk (no need for further investigation of the chemistry, aside from delivery vehicles into the body, some of which again is already known). Drug/chemical companies already make it, just not in a form manufactured under the safety standards needed to inject into humans.

Posted by: Cryptic Ned | Jan 18, 2007 3:36:59 PM

Part of the high cost ($400-600M?) is because MD's are paid for participation - how much I don't know. Yes, they have greater record keeping duties with in-trial drugs, but how does that cost compare to what they are paid?

If ways could be found through legislation, and using government insured patients (medicare, medicaid, VA, etc.) to lower this cost, then the drugs could be issued initially as generics, making the cost to the patient very reasonable.

I don't know what this would take, but having a government run clinical trial setup alongside the private sector seems like a good way to keep the drug companies more responsible and less greedy.

Right now, the high cost of this trials is dragged out each time the drug industry is subject to investigation by Congress, so clipping their wings in this area seems fruitful.

Posted by: JimPortlandOR | Jan 18, 2007 4:21:34 PM

Well, so far there's been no triumphalism about how R&D goes down the tubes without massive profit. Good. As a way to inoculate against that, I'd like to point out that CTV and the Express News of the University of Alberta itself are not nearly as pessimistic about the drug's prospects as is AHN.

I don't really know how things work in Canada, but in the USA a drug that is shown to have a certain effect in one study is not automatically ready for human testing. latts is right, there's a lot of promising paths in cancer research that have led to dead ends. I would imagine that some pharmaceutical company would be more willing to bite after another study or two.

Also, since it is already used as a drug, I find it hard to believe that drug companies don't find it profitable enough to make. Seems to me that Michelakis is trying to play politics a little bit, pressure some company to invest in further study. He's just not very good at it.

Posted by: Stephen | Jan 18, 2007 4:39:26 PM

This story is almost identical to the story of GenoMed, a St. Louis startup that is trying to tell the world about its success in using genomic therapies to avoid or delay dialysis for end-stage renal disease, West Nile virus encephalitis, coronary obstructive pulmonary disease, and type 2 diabetes. The problem, according to founder and nephrologist David Moskowitz, M.D., is that ESRD generates about $25 million a year in dialysis treatments, about 90 percent of which would go away if GenoMed's therapies were applied.

The guy has actually sold his therapy in India, but can't get beyond cult status here, apparently because too much profit would disappear if he's successful.

Posted by: Rick | Jan 18, 2007 4:46:26 PM

Reminds me of a certain researcher who discovered that most ulcers were caused by a bacteria that could be easily cured with an old antibiotic. It took him ten years to get the information out as he was constantly blocked by pharmaceutical companies reliant on the billion-dollar antacid industry.
Sometimes you need the government funded studies to prove one way or the other who is in the right.

Posted by: Hawise | Jan 18, 2007 4:48:02 PM

see here for another example of that

Is "here" supposed to be a link?

Posted by: Sanpete | Jan 18, 2007 4:50:30 PM

The problem, according to founder and nephrologist David Moskowitz, M.D., is that ESRD generates about $25 million a year in dialysis treatments, about 90 percent of which would go away if GenoMed's therapies were applied.

I'm very suspicious of claims like this as explanations why a treatment isn't catching on.

It took him ten years to get the information out as he was constantly blocked by pharmaceutical companies reliant on the billion-dollar antacid industry.

How did they block him?

Posted by: Sanpete | Jan 18, 2007 4:54:43 PM

this is another aspect of this. that companies will try to block other cheaper products from entering the market.

Posted by: akaison | Jan 18, 2007 4:57:13 PM

Orac had a good response to Digby's post. Most of these kinds of claims amount to utter nonsense (see Clark, Hulda). A proper trial of this medication would cost far less than $400M, could be financed through the NIH (or various cancer foundations), and numerous examples of similar trials can be easily located (e.g. aspirin, vitamin E). There is absolutely no reason why Glaxo or whomever should finance trials for every crackpot idea that some quack comes up with. We even have an office of quackery, um, I mean alternative medicine, under the NIH that funds trials for various non-scientific medical theories.

The stories about Barry Marshall and his rejection by conventional medicine are urban myth. Once he proved his theory, conventional medicine signed on cheerfully. Proton pump inhibitors are used even more widely since the H. pylori/ulcer connection was identified. Alternative medicine advocates like to use Barry Marshall as an example of a rejected prophet. The real story is that he was a conventionally trained scientist working with convetional scientific methods whose work was accepted by other scientists without any problem.

I am an M.D., my husband is a Ph.D. He works in industry in research. He makes a whole lot more money than me. 'Splain that JimPortlandOR. M.D.s are not particularly highly paid if you compare their incomes to the incomes of other highly educated American professionals. And if you look at incomes normalized for hours worked, medicine is an even worse deal. Hubby does not take call nights, weekends, and holidays.

Posted by: J Bean | Jan 18, 2007 5:59:30 PM

J. Bean: I wasn't suggesting that MD's are overpaid, only that some MD's have practices are essentially dedicated to testing new drugs; that they are paid normal fees by patients or insurers, but they also get paid by the companies conducting the research (and they don't disclose the drug company payments). I worked briefly in the office of one of these MD's almost 15 years ago, and I know from personal knowledge that the MD's spent no more time with patients in drug trials than with other patients. So the payments from the drug trials were gravy on the practice - at what amount I don't know.

I have great respect for the medical profession and believe MDs should be compensated fairly for the amount of education and training they have undertaken.

I just don't think that drug companies should be paying MDs to test their medicines - it is a compromise of their ethics and casts doubt on the results. Since clinical testing is supposed to cost the drug companies a lot of money, I'm wondering if the costs are for the payments to MDs, or what? Why are these costs so high?

Posted by: JimPortlandOR | Jan 18, 2007 9:43:55 PM

Ezra seems to think that the government can easily pick up the drug industry's R&D. 'Fraid not. What the drug industry does is very different than the NIH.

The NIH is like a university. It employs ace senior types and a large staff of juniors. It retains the aces on academic terms: low pay, but prestigious self-directed work and consulting. It runs the juniors hard for a few years, and lets most of them go. Some end up in academia, some in pharma, and some in law school.

This model works okay, but it is not infinitely scale-able. You can't have everybody do self-directed work. You can't give every senior researcher a large staff of hungry ambitious juniors. The old fallacy of composition.

That's #1. Now for #2.

A lot of primary drug development is really boring stuff: whip up an enormous number of molecules, feed them to rats (or lawyers), and see what works. (Rational drug design may be the Holy Grail, but I haven't yet met Professor Parsifal.) Nobody skilled would do this grinding large-team work if they weren't paid good money. And if something works on rats, then you have to see if it works on people. Very expensive, and not much opportunity to park a prestigious article in Nature magazine. Again, you don't do this kind of work for the glory, unless perhaps you are the head of a large medical lab. The staff is doing it because they are paid.

This is not to say that we are stuck with the status quo in drug development. However, you can't just scale up the NIH. The government has had some experience with scaled-up lab work, such as the weapons labs. It hasn't been all that great.

Some economists suggest awarding multibillion prizes instead of patents. Some folk suggest direct government contracting with private labs. Beats me. But we can't scale up the NIH.

Posted by: Joe S. | Jan 18, 2007 11:26:02 PM

Some folk suggest direct government contracting with private labs.

And the NIH would oversee this, right? That seems sensible.

Posted by: Sanpete | Jan 19, 2007 12:08:49 AM

The 400-600 million figure that Ezra cites, if true, speaks towards something that should be acknowledged: the FDA as it currently stands is very harmful. For every unsafe drug that it helps the market avoid, it reduces the speed and ability of life saving drugs to come to market. It also reduces the potential profitability of drugs so that the sort of thing Ezra is talking about happens. And of course there is the Plan B saga.

Note that you could also reduce the life of a patent quite easily if you shrunk the patent time, reducing not only the cost of testing, but increasing the speed at which new life saving drugs become generics.

I say up the funding for the NIH, shrink the FDA, negotiate for drugs for Medicaid Part D, and we're going a long way towards fixing the problems.

Posted by: Joel W | Jan 19, 2007 3:43:03 AM

Joe S.: I'd probably believe you if you could find, say, a director of the NIH saying "Actually we don't need any more money. We couldn't use it. No way we could expand - it just wouldn't work. No, we're fine as we are."
Or, indeed, anyone with any knowledge of the situation. At the moment it's just reasoning de recto.

Posted by: ajay | Jan 19, 2007 6:30:30 AM

Listen, when Reagan cut the CDC budget, nothing bad happened, except of course the massive spread of HIV and AIDS without the necessary research to inform people how to stop that spread. But if we learned anything for that experience, it was that private industry would step up, except they didn't. But it's not like the "socialized" nations made any contributions on that initial research, except France, Canada, etc. . . .

Posted by: MDtoMN | Jan 19, 2007 8:32:18 AM

I wasn't arguing that basic research was overfunded; I was arguing that the basic research model of the NIH does not transpose well to the kind of applied work that the drug companies do. But even if I was arging that basic research is in fact overfunded, your rebuttal is weak.

No bureaucrat--public or private sector--will ever say that their fiefdom is getting too much money. So a lack of bureaucratic suicide is not evidence of underfunded labs; it is evidence of human nature.

But by the way, there is some history indicating that basic research is sometimes overfunded. Let me pick an example: the NIH. Nixon pumped a huge amount of cancer money into the NIH in his first term: his "War on Cancer." The result? 1) Cancer wasn't cured. 2) Nixon was reelected--only a weak causal relationship, I'm sure. 3) The National Cancer Institute (the cancer part of the NIH) became notorious for the relatively low quality of its tenured researchers. They just hired too fast, because they had too much money and had to fill the shiny new labs with warm bodies. I think that they've fixed this by now.

I don't think that, in general, basic research is overfunded. In general.

Posted by: Joe S. | Jan 19, 2007 4:04:38 PM

We can cut the FDA costs quite simply. Manufacturers (Big Pharma, genercis, charities, NIH, whatever) only have to prove that the drugs are safe. Remove the "effective" part and Stage III clinical trials (the big scale stuff that really costs the money) become much cheaper.

The ulcer guy: didn't he get The Nobel for that work? There are other incentives for drug research than corporate profits, obviously.

Posted by: Tim Worstall | Jan 20, 2007 2:12:14 PM

Cynicism about the gaming of the system must be setting in. Who will give odds that government review of drugs doesn't mainly act as a brake on liability torts ? ( And keep out new/smaller players ? )
This isn't from study or anything but perversity.

Posted by: opit | Jan 20, 2007 11:22:16 PM

I have not found the research to back up the claims of dichloroacetate sucess. All I have found was that it is used to induce liver cancer in rats. The articles of its success seem to be in newspapers by journalists. I do not believe that funding for a cancer cure is not there for some diabolical reason, because the prestige of coming out with the cure would be tremendous. I suspect this whole thing will turn out to be bunkum.

Posted by: AG | Jan 24, 2007 4:53:00 AM

The january issue of Cancer Cell has the article by Bonnet et el 2007. Professor Michelakis is a legitimate professor in cardiac research at the University of Alberta, at Edmonton. He is the last author on the article. Try Google to locate information.

Posted by: Mark | Jan 26, 2007 2:43:54 PM

hey Ezra,

You are being called out for being an idiot when challenging the patent based medical industry over on


via reddit

You might want to challenge this. And inform Digby as well. Apparently you can't know anything, about this subject.


Posted by: patience | Jan 30, 2007 12:13:38 PM


Posted by: jamo4000 | Mar 16, 2007 9:45:33 PM

The comments to this entry are closed.