Tuesday, April 6, 2010

The efficacy of predicting gene prediction

Francis Collins has a track record of predicting tremendous advances in health from the human genome project --what he and many others said would be a revolution. Writing in the April 1 issue of Nature, which celebrates the 10th Anniversary of the Human Genome, he asks if the revolution that he and others foresaw more than a decade ago has now arrived. He writes that he never deletes a powerpoint file, so he knows exactly what he predicted when he stood on the podium with President Clinton in 2001 and announced the completion of the human genome sequence.
Predictive genetic tests will be available for a dozen conditions
Interventions to reduce risk will be available for several of these
Many primary-care providers will begin to practise genetic medicine
Preimplantation genetic diagnosis will be widely available, and its limits will be fiercely debated
A ban on genetic discrimination will be in place in the United States
Access to genetic medicine will remain inequitable, especially in the developing world
He says that it's fair to say that these predictions have come true. Well, yes, one can always declare victory after the fact. He also adds, "The promise of a revolution in human health remains quite real."  

But let's look at his predictions one at a time.  In fact, there are dozens of predictive genetic tests available -- but how accurate are they? Interventions to reduce risk of, say, obesity or type II diabetes are available, yes, but they've been known for decades, if not centuries -- diet and exercise. What he means by genetic medicine isn't really clear, and preimplantation genetic diagnosis can't be claimed to be due to HGP, but to technological advances, as well as the discovery of diseases for pediatric Mendelian diseases, which was already being done before the human genome sequence. A ban on genetic discrimination is in place (GINA), but how successful that will be is still an open question (and, as our friend, geneticist and tuba player Joe Terwilliger says, the differential rates 18 year old boys and girls pay for car insurance is genetic discrimination, isn't it?). And, well, predicting differential access to anything medical based on income disparities isn't exactly a challenge.

Collins in fact made many other predictions (or promises) for the HGP, particularly in the late 1990s as the human genome was nearing completion. For example, in a lecture he delivered in 1999 he presented a scenario for the year 2010 whereby a hypothetical young man, a smoker, is found to have high cholesterol. Here's a bit of that imagined scenario:
Aided by an interactive computer program that takes John's family history, his physician notes that there is a strong paternal history of myocardial infarction and that John's father died at the age of 48 years.
To obtain more precise information about his risks of contracting coronary artery disease and other illnesses in the future, John agrees to consider a battery of genetic tests that are available in 2010.  After working through an interactive computer program that explains the benefits and risks of such tests, John agrees (and signs informed consent) to undergo 15 genetic tests that provide risk information for illnesses for which preventive strategies are available.
John is please to learn that genetic testing does not always give bad news -- his risks of contracting prostate cancer and Alzheimer's disease are reduced, because he carries low-risk variants of the several genes known in 2010 to contribute to these illnesses. But John is sobered by the evidence of his increased risks of contracting coronary artery disease, colon cancer, and lung cancer. Confronted with the reality of his own genetic data, he arrives at that crucial "teachable moment" when a lifelong change in health-related behavior, focused on reducing specific risks, is possible. And there is much to offer. By 2010, the field of pharmacogenomics has blossomed, and a prophylactic drug regimen based on the knowledge of John's personal genetic data can be precisely prescribed to reduce his cholesterol level and the risk of coronary artery disease to normal levels...
Understandably, Collins was a great cheerleader for the genome project. But that is exactly the problem, and we regularly write about it: the life-as-lobbying worldview. Dr Collins did say that he believed that diseases would turn out to be caused by numerous genes with small effects, interacting with environmental factors, but he clearly believed that these genes would be common and identifiable (hence, the HapMap project, whose scenarios and promises did not materialize except by a lot of post hoc wriggling and redefinitions, such as of 'common').  And he clearly believed that understanding genetic effects would be straightforward -- and counteractable, with some help from the pharmaceutical industry.

In fact, the 'several genes' he predicted would be found by 2010 to be responsible for Alzheimer's disease turn out to be hundreds of genes, each with many alleles and mostly with tiny effects. And, although he did write that lifestyle changes were a component of prevention, and that John should quit smoking, he also clearly believed that designing prophylactic drugs based on what was learned from the human genome would be so easy that by the year 2010 we'd be able to prevent many genetic diseases pharmaceutically. After we easily predicted them.

But in fairness, no one should be strictly held to their predictions (well, except for seers and grant seekers who promise too much) and mostly this hypothetical scenario is interesting as insight into Collins' beliefs about the importance of genes and the power of technology to counteract them -- a set of beliefs that still drives him as director of the NIH, which we have blogged about before. But to us, at least as interesting is something we wrote about last week, the use of modern technology to tell us something we already know. In Collins' hypothetical scenario -- which was completely made up; remember, he could have imagined anything -- John learns from his genetic testing that he was at risk of heart disease and colon cancer. But his family history already told him that! This isn't exactly something that justifies the billions of dollars spent on the HGP!

As director of the National Institutes of Health and past director of the National Human Genome Research Institute, Collins had, and continues to have, tremendous influence over the direction of medical research funding for the last decade, and into the foreseeable future. Indeed, as we've said many times before, his faith-based commitment to improving our health through genetics and technology is taking real money away from real problems that could have real solutions, given equal commitment to solving them. 


Ken Weiss said...

I want to add an afterthought. No endeavor as huge as what Collins has lead and advocated would be devoid of successes. And a scientist, like any human, can be excused for dreaming of glories and greatness, and of commitment to his/her idea.

The problem is that this is major and ongoing public funding based on promised results. While after the fact one can always single out things that worked, that is not the appropriate measure of the worth of the investment.

One more appropriate measure is accountability of results relative to promise. Nobody expects perfect, but one should expect major results from major, long-term funding.

A second measure is difficult to make--but is almost never applied though it should always be applied. That is: what _else_ could have been done with the same funds, and might it have had greater success? Here the answer is at least somewhat clear: most disease and public health problems are not due to genetic variation. They are due to many factors that can be called 'environmental'.

But once vested interests are in place, as they are here, it is unlikely that the ship of interests will be turned. Thus, Dr Collins continues to promise a pot of gold, only now the rainbow he wants us to chase is larger, longer-term, and much more expensive.

That's why a better analysis of what we've done, where we are, and where we should be going--in relation to the source and objective of the funding support--is warranted.

Anne Buchanan said...

Collins says in the Nature piece, "The promise of a revolution in human health remains quite real."

But this is nothing new. Mega-spending on human genetics has long been justified as expenditure that won't be good for us, or for our children, but that will benefit our grandchildren. This is a statement of faith, and we are asked to take it on faith as well.

But, it means essentially that we can expect no accountability for how this money is spent, as the promised pay off can always be said to be far in the future, too far for any of us to see. And that is troubling.