HEALTH - FEBRUARY 11, 2004

The downside of genetic testing

In my last column I discussed some of the recently developed genetic tests and talked about when and for whom they might be most useful. Here I want to discuss the unfortunate features of many of these new genetic tests and sound a warning note. The downside has less to do with genetic testing per se, and more to do with the commercialization of the whole process of genetic research and development that has led to these tests now being available.

I’ll take as text for today’s sermon an article by T.A. Badger of The Associated Press, “Test May Spare Women Chemo,” in the Dec. 6 issue of the Detroit Free Press. The article follows an announcement by Genomic Health Inc. (described as a Silicon Valley biotech company) that it had identified a set of about two dozen genes. Taken together, this set of genes is a very good predictor of which women whose early breast cancer has just been diagnosed, will in later years have a recurrence of breast cancer.

The practical importance of this finding seems to be immense. Today, any woman in this situation is faced with a tough choice – whether to follow up the initial surgery with a potentially agonizing course of chemotherapy. Supposedly, once the new genetic test panel is available in 2004, a woman can elect to have the genetic test done. If the results show a high risk of recurrence of cancer, she can proceed with chemo knowing that she’s among the group really likely to need it. If she tests out in the group with a low risk of recurrence, she can avoid the chemo with confidence that she’ll probably do well without it.

The first people who might want to rain on this parade are the advocates of evidence-based medicine. They’d point out that there are several untested assumptions in this comforting scenario. To take the worst case, suppose that the genetic test predicts not only those women with a high likelihood of recurrence, but also those cancers that are most resistant to chemo? Instead of selecting for the women most likely to benefit, the testing would actually pick out those least likely. The only way to be sure about the actual health outcomes is to follow women for some years after they have had the test and the chemo, compared to the strategy in use today.

The problem with this evidence-based approach is that women need to make decisions now and the research results would not be available for maybe five or 10 years. No one could blame women if they refused to be randomly assigned to testing vs. no testing, which would be the ideal way to do this research.

But there’s another downside to this whole system. How sure can we be that the gene test actually performs as well as it’s supposed to? And, if new information becomes available in the next few years that shows that the test is not performing up to par, how confident can we be that this information will become known?

The answer is that in today’s setting, all information of this type is basically proprietary. As the Mafiosi said in “The Godfather,”it is not about our health or our lives, “it’s just business.” The bottom line for Genomic Health Inc. is that it probably paid for the research done so far on its genetic test, and it has legal ownership rights to data it collects in the future. If word ever gets out that the test is not as good as it was supposed to be, one thing can be predicted with 100-percent certainty – Genomic Health Inc. will see its stock take a beating. What company is willing to take that risk?

We have already seen instances in the biotech industry of a test that was initially marketed as being perhaps 80 percent accurate in predicting a risk for cancer, and information later came out to suggest that the accuracy was actually perhaps as low as 40 percent. We’ve seen numerous instances in the pharmaceutical industry of companies suppressing negative data about risks and side effects of drugs – and the pharmaceutical industry is much more tightly regulated compared to the biotech firms that market genetic testing.

Our nation has decided to rely on the free market — along with the patenting of human gene products and the means by which genetic tests can be done — to develop the future technology in this area. That means that there is essentially no guarantee that scientific data to tell us the truth about how any genetic test performs will become public knowledge. It’s just business.

On that basis, anybody wanna buy a genetic test?

Howard Brody, MD, teaches family practice and medical ethics at Michigan State University. You can reach him at brody@msu.edu.

Care to respond? Send letters to letters@lansingcitypulse.com. View our Letters policy.