Cautionary tales: An occasional look at events with potential long-term impact for biotechnology
Personalized medicine takes a hit — Scientists have spent more than a decade scouring the human genome to identify genetic alterations that might predict your risk of developing, say, heart trouble or cancer. Now, however, a new study (subscription required) Wednesday in the Journal of the American Medical Association suggests that many of these disease “biomarkers” identified so far may be little more than junk. (For a lay review of the report, click here.)
The main problem here is that “personalized medicine” — a future in which doctors anticipate medical problems and prescribe tailored treatments based on your individual genetic profile — was supposed to be one of the revolutionary consequences (not to mention big businesses) of decoding of the human genome project. If biomarkers, which are effectively the building blocks of personalized medicine, don’t actually tell us what we thought they did, that’s not particularly encouraging.
In the JAMA study, a research team led by Washington University’s Thomas Morgan re-tested 85 biomarkers that previous studies had identified as risk factors for heart attacks and other problems related to atherosclerosis and other forms of coronary artery disease. To do so, the team scanned the genes of 811 heart patients and 650 healthy people in order to confirm the purported links between disease and particular genetic variations. Of the 85 variations retested, however, the researchers found that only one — a mutation in the gene for a protein called beta-fibrinogen — appeared to qualify as a true risk factor.
The problem seems to be that many biomarkers are “discovered” in small studies, where it’s difficult to tell the difference between a real gene-to-disease association and an imposter that appears in the data by random chance. Heart-disease specialists now face the disheartening task of starting all over with larger, more rigorous and more expensive trials. Since scientists claim to have identified dozens of biomarkers for other conditions — cancer in particular — in more or less the same way, it’s entirely possible that the current edifice for personalized medicine may be much shakier than just about anyone had imagined.
The little drug that could, and did, too much — Hands down, the most successful biotechnology drug in the world is erythropoietin, or EPO, a genetically engineered version of a natural human protein that stimulates the production of red blood cells. EPO and its next-generation cousin racked up $6.6 billion in 2006 sales for Amgen, helping make it the world’s biggest biotechnology company. EPO is primarily used to treat anemia in patients with kidney failure or those undergoing chemotherapy, and has long been a major cost item for Medicare, which in the U.S. pays for most of its use in kidney disease.
All, however, is not well in EPOland. Since late last year, the EPO-related drugs have started to pile up safety concerns, with studies showing that they may increase the risk of heart attacks and strokes in kidney patients at high doses and cut short the lives of cancer patients (some of whom were receiving radiation therapy and others who weren’t). Amgen’s travails with the drug are ably recounted by Marilyn Chase in this recent WSJ article (subscription required).
Although the WSJ notes concerns related to Amgen’s support for higher EPO doses and its advertising strategy, it stops short of accusing the company of engaging in Big Pharma-style overpromotion of its drugs. For that, you have to turn to a recent editorial in Nature Biotechnology (hat tip: Derek Lowe), which, among other things, blasts the company for a “massive marketing blitz” aimed at getting oncologists to use more of its drugs despite increasing fears that EPO-related drugs might actually promote tumor growth. As the editorial notes:
Amgen does not come out of this well. Although seeking new indications for existing medicines is clearly a valid strategy, the company appears to have miscalculated the balance between expansion and the risks to its existing business—and potentially opened itself to charges that it has recklessly endangered patients’ lives.
This is not territory any biotech — nor any drug company, for that matter — wants to find itself in.
By way of contrast, commercial hopes were high last year when Pfizer launched Exubera, the first inhalable form of insulin for diabetics. Co-developed by biotech Nektar Therapeutics, Exubera was supposed to relieve diabetics of the inconvenience associated with injecting insulin multiple times a day. Pfizer predicted the drug would eventually pull in blockbuster annual sales of $2 billion.
That, of course, hasn’t happened. The Exubera inhaler, it turns out, is cumbersome to use, the drug costs more than regular insulin, and patients appear spooked by lingering concerns about its impact on lung function. Not to mention the fact that new insulin-injection systems now use smaller needles and are far less painful than they used to be. Pfizer last week re-launched Exubera with a new marketing campaign. The moral here: Sometimes fear itself isn’t the only thing to fear in drug development.
Non-profit drugs? — Makers of drugs for HIV/AIDS have reluctantly started to embrace the notion that they can’t sell drugs intended for use in the developing world at First World prices, sometimes with the assist of an activist government like that of Thailand. Now comes news that GlaxoSmithKline plans to launch a combination vaccine called Globorix for use only in Africa from which it “never expects to make money”. A detailed but somewhat wishy-washy analysis (from a Fleishman-Hillard “consultant,” to boot) follows at Ethical Corporation, a self-described independent publisher and conference organizer in the U.K. that focuses on “how companies relate to the world around them” (hat tip: Pharmalot).
This is a tough one to puzzle out at this point. GSK may just be doing the right thing, as it appears to insist, or it may be trying to buy goodwill with drug activists and international NGOs to gain leverage in some other controversial area, which seems more likely. Given increasing drug-pricing pressures on pharmas and biotechs alike, though, it’s certainly intriguing and a trend worth watching.
VB's research team is studying web-personalization... Chime in here, and we’ll share the results.