Name: Mark Guyer
Position: Director of the division of extramural research at the US National Human Genome Research Institute
Background: Staff, various capacities, Office of Human Genome Research, 1988 – the present
Education: PhD, bacteriology and immunology, University of California, Berkeley, 1974
The National Institutes of Health this week said it will map genomic changes that occur in lung, brain, and ovarian cancers as part of its mammoth Cancer Genome Atlas project.
Pharmacogenomics Reporter this week spoke with Mark Guyer, director of the division of extramural research at the US National Human Genome Research Institute, to get an update on the Atlas project’s schedule and to learn more about how findings from a recent study at Johns Hopkins University may impact the effort.
Hopkins researchers last week published an analysis of 11 breast cancer tumors and 11 colorectal cancer tumors in the Sept. 7 issue of Science Express that validates the concept behind the Cancer Genome Atlas Project [see related story, this issue].
According to the Hopkins researchers, the results “support the notion that large-scale mutational analysis of other tumor types will prove useful for identifying genes not previously known to be linked to cancer.”
What are the next steps for the Cancer Genome Atlas project to announce?
Some of the sample collection will begin immediately. Another component is the Cancer Genome Characterization Centers. They will be announced by the end of the month; those decisions haven’t been made yet. Another component is the Cancer Genome Sequencing Centers, and for those, too, the decisions haven’t been made yet, and so they’ll be announced sometime after September.
Can you tell me how the Johns Hopkins study is relevant to the Cancer Genome Atlas project?
What the Hopkins group did is a small version of what the [Cancer Genome Atlas] project wants to do in the sense that the finding from the Hopkins group is that there are many more genes in which variants or mutations can be identified that seem to be linked to cancer.
It basically provides a demonstration that the premise on which the Cancer Genome Atlas project is predicated is a good assumption, and that is that a thorough analysis of the genomic structure of tumor types will reveal many more genes which are candidates for involvement in cancer somehow.
Does it tell you anything about the pharmacogenomic potential of the Cancer Genome Atlas project?
This finding? Not directly, but it gives indication that we will be able to do the same kind of analysis by doing the kind of whole-genome scanning analysis that is also possible now to look at patient response.
Ultimately, we believe that there is promise that individual response to therapeutic activity and sensitivity to side effects — to the extent that those are genetically determined, we should be able to obtain that information.
Have the Johns Hopkins results helped to clarify how much of a problem cancer heterogeneity is going to be in characterizing cancer types?
The concern that we’ve had is that heterogeneity would make it difficult even to do the analysis. In the sense that the Hopkins results came from two tumor types that are generally reasonably heterogeneous, it suggests that the analysis can be done.
I will put the caveat on that statement that I don’t know anything about the actual tumor samples that they investigated. So, I don’t know whether they did any pre-treatment to reduce the heterogeneity, or whether they took particularly homogeneous examples of the breast and colon tumor.
I think there are a number of good strategies that will allow us to determine what degree of heterogeneity can be tolerated. And some of the new sequencing technologies — 454’s and Solexa’s, both — are likely to be better at identifying variant bases that are present at low frequency in the population than the ABI [Sanger sequencing] technology is.
So, we’re reasonably optimistic about dealing with the heterogeneity problem to a significant extent. I don’t claim that we’re going to be able to solve it, but we’ll be able to better than we can do now.
Do the Hopkins findings have implications for personalized medicine?
The Hopkins results seemed to a first reading as if they reinforced the idea that the development of approaches to personalized medicine is going to be necessary. If the mutational spectrum in each tumor — even of the same type — is so different from individual to individual, that is, I think, a good indication that there will be a lot of personalization of tumor types.
The implication of that, I hope, is that we can use that either for diagnosis or treatment at follow up. So, I think there was nothing in the Hopkins results that discouraged anybody’s enthusiasm for personalized medicine.
Is it safe to say the results were encouraging to the Cancer Atlas project?
I think that’s right.