Topol: Cancer Mutation Screening 'Grossly Insufficient'


November 25, 2013

This feature requires the newest version of Flash. You can download it here.

Hello. I'm Dr. Eric Topol, Editor-in-Chief of Medscape. I'm here with you for a commentary on cancer genomic medicine.

This is turning out to be an ultra-hot topic today, because so many things are moving at the same time. On the one hand, for a long time we've had separate mutation screening tools; if someone has colon cancer, we screen for KRAS mutations, or in breast cancer, we screen for HER2 mutations. But what's changing quickly is our recognition that that is grossly insufficient; the question is where we go from that predecessor.

One tool that's been recently validated, as published in Nature Biotechnology, is a limited exome panel of 287 genes, all of which have been implicated in the development of cancer.[1] It's from a company called Foundation Medicine, and it involves sending formalin-fixed, paraffin-embedded blocks of tissue from the tumor biopsy to [their facility in] Cambridge, Massachusetts, where the sequencing is done at very high coverage -- well over 1000×, on average. They then sequence this limited swath of 287 genes, which is only about 15% of the human genome. They are not addressing anything beyond these coding elements; as we know, there is a lot more to cancer than just this little piece of the genome.

Nonetheless, the results were interesting: They reported a rate of actionability of finding a driver mutation deemed to be causative of cancer in almost 80% of individuals. This doesn't particularly bode well for centers all around the country that are not used to sending out their samples for sequencing.

Of course, there are other solutions to this, not only at the local level in health systems, but more in the ability to go well beyond just 287 genes. How about the entire exome of 20,000 genes? How about the whole genome? We recently learned about such things as TERT promoters, a long, noncoding segment that is involved in aggressive prostate cancer, and that more than 100 noncoding variants well removed from genes were implicated as drivers for cancer by a new informatics tool called FunSeq.[2,3,4]

We have a ways to go to catch up with the knowledge that we're learning about the oncogenes and tumor suppressor genes that are at the root of cancer. Over the past 2 years, there have been 20 new drugs that the US Food and Drug Administration (FDA) has approved that are mutation-guided drugs, but unfortunately, we're not broadly screening people yet and we have just 1 commercial new entry to this space. We need to really gear up and come up with better solutions that are more comprehensive for patients, given the extraordinary body of knowledge that we've learned about cancer genomics and about what is driving disease.

One last point: Today we can pick a drug, often a very expensive drug, that's specific to a mutation, and we can have an extraordinary effect that can last for 9 months or 12 months. But then, typically, as is the case with BRAF mutations, there's a recurrence.

We have to be much more intelligent about up-front combination therapy, sequencing at very high coverage, and figuring out whether we can use cell-free tumor DNA or other methods to take into account this heterogeneity of tumor, to be able to anticipate resistance and recurrence. That is the next phase of cancer treatment that we need to prospectively assess.

Thanks very much for your attention. You can see that there's a lot of unrest in coming up with better ways to approach and use the knowledge that we have in cancer genomics for the daily practice of medicine.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: