[Home] [Headlines] [Latest Articles] [Latest Comments] [Post] [Sign-in] [Mail] [Setup] [Help]
Status: Not Logged In; Sign In
|
Health See other Health Articles Title: Will Whole Genome Sequencing Predict Susceptibility to Illness? Eric Topol, MD: Director, Scripps Translational Science Inst. A recent paper by a group from Johns Hopkins published in Science Translational Medicine evoked considerable interest with the disappointing results...what were they and what does this mean? Below is a transcript of Dr. Topol's post "Will Whole Genome Sequencing Predict Susceptibility to Illness?." We look forward to your feedback. This Genomic Medicine segment is on whether sequencing will be able to predict the susceptibility to illness. I selected this topic because of a May 9th Science Translational Medicine article by Roberts and colleagues [from Johns Hopkins]. It's called "The Predictive Capacity of Personal Genome Sequencing," as you can see here, and with it is an accompanying editorial that is called "What's a Genome Worth?." It's a very provocative paper and editorial, and it led to, in April, a front page New York Times article about the limited power of DNA to predict illness. That was, of course, controversial in some respects because the downplay of the sequencing is occurring just at the time when sequencing is revving up as a technology. But we have known that common variants are not very good at predicting susceptibility to illness overall, although there are some exceptions. The question is when we do whole genome sequencing, will that provide a better window to susceptibility to diseases? This Hopkins paper is very interesting because, even though it talks about the predictive capacity of personal genome sequencing, there is no genome sequencing in the paper. In fact, what it does is collate data from twins, monozygotic twins -- a very large number of them, from many sources, tens of thousands of them -- but we know that even monozygotic twins do not suffer the same illnesses or maladies. Furthermore, without any sequencing data to try to use this as a proxy -- these twins -- with a model that is difficult to interpret, the conclusion is that this is going to be a very limited utility for disease susceptibility. Well, that may be true. In fact, what we have now is just that indeed that may be the case. There is one line and a conclusion of this study that I think is important to note, and I'll read it here: "More than 90% of tested individuals might be alerted to a clinically significant predisposition to at least one disease." Well, that's not so bad actually, and, in fact, if it's a disease that’s actionable, that could be highly prevented. Of course, it could be a whole lot better than that. The accompanying editorial "What's a Genome Worth?," written by Isaac Kohane and Jay Shendure, is really interesting because it goes over the limitations of this Roberts paper, but then goes on to say that whole genome sequencing is likely cost $1000 all in -- that is, not just sequencing, but also the interpretation. And the effect of that, if it is done as a newborn, is particularly striking. We spend $9000 per individual [on healthcare] in the United States and that $1000 per individual over the course of one's lifetime, amortized, comes out to $13, or one-tenth of 1% of current support for each individual’s health. That's a pretty good bargain, even if it only comes up with one illness. This topic is obviously very controversial because the hope has been that when we get a newborn or infant's genome sequencing done, we'd have a lot to go with in terms of preventing many illnesses -- in fact, stop their natural history -- and that yet remains to be seen. When we have data on millions of people -- which is forecasted by 2014 -- sequences with the different critical phenotypes, we'll know a lot more with respect to rare variants and de novo variants and all sorts of structural changes that can occur in the genome. One other thing: this is the hardest thing, predicting illnesses. But what about the use of whole genome sequencing for rare idiopathic conditions, which we have talked about in prior segments? Or for cancer to detect driver mutations? A lot has been done in that area in recent months to understand the branching evolution of how cancer develops. Sequencing appears to identify driver mutations in a number of critical types of cancer, and then, of course, that could be coupled with better therapy. There are other things that will be useful for whole genome sequencing. Perhaps this hardest thing, which is predicting susceptibility to illness, will be the longest it takes to require validation. But until that time, probably it's not safe to conclude the predictive power of sequencing is nil or limited. Let's wait and see. In the meantime, I'd also like to get your comments whether you think there will be value of whole genome sequencing for predicting common diseases. I'll try to respond to your comments, and thank you very much for your continued attention to the Genomic Medicine website. Post Comment Private Reply Ignore Thread
|
||
|
[Home]
[Headlines]
[Latest Articles]
[Latest Comments]
[Post]
[Sign-in]
[Mail]
[Setup]
[Help]
|
||