The Revolution Has Not Been Personalized: Genomic Medicine at the Ten-Year Mark
“The future ain’t what it used to be,” as Yogi Berra put it, and the state of genetic medicine arguably bears out that adage. The completion of the Human Genome Project in 2003 seemed to portend a revolution in healthcare. Soon, optimists hoped, we’d be able to stop disease before it started and personalize treatment to fit each individual’s underlying genetic code. But a decade later, genetic medicine remains largely a “desert,” in the words of Reed Pyeritz, professor of medicine and genetics and chief of the Division of Medical Genetics at the Hospital of the University of Pennsylvania.
Pyeritz is a leading voice among geneticists working to translate our immense knowledge of the human genome into clinical practices that actually improve human health. It has proven to be a bigger challenge than many people expected—for reasons that go beyond the inevitable lag between scientific discovery and application. Recently he spoke with the Gazette about the research, ethical, and logistical challenges that will have to be met if genetic medicine is to deliver on its promise.
Why has genetic sequencing not delivered the medical benefits many hoped it would back in 2003?
The expectation was that once we had the sequence, we’d be able to sort everything out very quickly. That has proven not to be the case whatsoever for common diseases. In fact, genome-wide association studies have shown that there are dozens of genetic markers that seem to convey some risk for virtually any common disease you choose.
What kind of research has to happen for genetic information to become more clinically useful?
The amount of information is expanding tremendously. We know that there are about 23,000 genes. We still don’t know what some of them do, or whether they’re associated with disease, and that information continues to emerge on a weekly basis. But those 23,000 genes occupy only about 2 percent of the sequence of the human genome. We used to think the other 98 percent represented “junk DNA,” and we know now that [term] was very pejorative. We know now there’s a tremendous amount of regulatory RNA that is coded in the space between the genes. Until we understand how genetic variation in the non-coding RNA really influences disease, I think we are only looking at the tip of the iceberg.
What do you think is the biggest misconception the public has about genetic medicine?
That there isn’t anything to do if you are identified as having a genetic risk for some common disease. I think that’s a misconception. Genes don’t operate in a vacuum. Obviously there are some disorders inherited in a Mendelian fashion that are almost invariant, but the majority of genetic conditions show a great deal of variation, even in people who have the exact same mutation. Why is that variability present? It’s because you have 23,000 other genes that vary among them, and some of those will affect how that one mutation is expressed. Second, there is the impact of environment and life experiences that do have an impact on gene expression. Third is this whole new area that is only now emerging, what we call epigenetics, where gene regulation is tremendously affected by nutrition, by maternal exposure, by a host of things. All that is by way of saying that we’re dealing with probabilities, not certainties, and that is really at the heart of modern genetics.
You’ve written that the challenge for genetic medicine goes beyond the limited clinically actionable data and also includes a lack of professionals to interpret and analyze genetic data, to provide genetic counseling. Why is that a problem?
Genetic testing is different than having your complete blood count obtained. You sort of have a notion that a blood test will be simple and straightforward—there will be a defined answer that your physician can readily interpret for you.
A genetic test isn’t like that at all. It has all these potential nuances. You may get something that’s confusing. Or in the case of sequencing your whole genome, you may get information that is truly disturbing that pertains not only to you but also to your offspring and other relatives, and so there are all these issues that many of us think should be considered by a patient before the DNA analysis is obtained. But that is time-consuming, and it needs to be done by someone who understands these nuances. Typically that has been a genetic counselor or a geneticist, but these folks are few and far between.
Could you describe the Duty to Recontact problem?
It relates to very practical experiences I have every week in my medical genetics clinic. When you do a genetic test for a specific condition, it’s possible the results will show that you have a variation in a [particular] gene, but it’s never been reported in any other person so we simply don’t know if it’s harmful. We call these variants of unknown significance. Everybody is then sort of stuck—is this an important genetic change or is it just a random variation that is not harmful at all?
We don’t know the answer to that at this point in time, but we may get smarter as time goes on. If I say to the patient, ‘We will get back in touch with you if we get smarter [about this variant],’ then I have established a long-term obligation to do that. I realized that we simply didn’t have the information systems in place to accomplish that. In an ideal situation we would re-contact people, but we weren’t prepared to do it. We’re going to get smarter with respect to an individual person’s results, but I think it’s most appropriate at this point in time to shift the burden to the patient to get back in touch with us if they want to get updated.
What should we look for in the next 10 years from genetic medicine?
I think we should be looking first for the increasing use of genetic sequencing in cancer, because every person’s cancer is genetically based and to a certain degree unique. The individual tumor will be sequenced and compared to the person’s underlying genetic constitution to see what changes have occurred, and that therapy will be based on what that genetic profile of the tumor displays. That’s already happening to a limited degree. A program through the Department of Pathology and Laboratory Medicine was just announced to increase the sequencing of tumors at HUP. That will be, I think, the first true advance based on genomic medicine.