The Future of Genomic Medicine: Day 1

[RockChalkJayhawk]

HI all,

I just wanted to drop some notes from this conference in La Jolla, CA. Rather than posting notes from each of the speakers, I'll just give a brief description of the major points that were discussed at the meeting. For starters, this is the first conference I have been to where I actually took a considerable amount of notes from each speaker. There were a number of leaders in the field here, and as expected, they gave phenomenal talks.

Key point #1: More genomes are being sequenced. Elaine Mardis talked about several of her whole human genome sequencing using non-tumor, tumor, metastatic tumor, and even xenograft tumors from the same patient. She talked about a clinical trial with aromatase inhibitor responsive tumors matched with 25 responsive tumors, breast cancer, AML, etc. How much longer will it be before sequencing one human genome won't be enough to get a paper published?

Key Point #2. Joseph Nadeau showed some very exquisite examples of trans-generational effects of mutations. In other words, a mutation that grandma had (and that you don't) may make you fat. You can sequence all day long, but genetics of the patient will not tell you everything. You may need generations of sequencing data to find genetic predispositions.

Key Point #3. Although this wasn't discussed much in the general session, a few of us talked about the need for some sort of standardization. Getting the sequence data is one thing, but as many of the members in this forum well know, that is the easy part. There are so many algorithms out there and couple that with the weaknesses in each of the technologies (both established and those in the pipeline), it is easy to see that there is no single way to do everything. This, I believe is a huge problem. We need something like the MicroArray Quality Consortium for Next-Gen Sequencing. I know the MAQC is already looking at RNA-Seq, but what about the rest?

There were a lot of really great talks, and I don't want to short-change any of them. So far this has been an excellent conference (although not quite AGBT) for the big picture. Oh yeah, and if you are a student, there is no registration fee. If you're not a student, it's $80, so its cheap enough for many to try it out.

I'll post the info from Day 2 when I get a chance.

[RockChalkJayhawk]

Day 2

Keynote Speaker: Lee Hood
He talked about a lot of things (all of which were very cool), but most of his talk was on some of the work they were doing in prion disease. They infected 8 strains of mice, measured gene expression at several timepooints. They then used these data to correlate with protein-protein interactions, histopathology, cliniclal signs, distribution of disease progression, etc. Finally they found that there were biomarkers they could use to identify perterbations to the system way before any of the clinical signs became apparent. He then went on to discuss organ-specific blood fingerprints to use in diagnostics. Basically, proteins can be secreted from any organ into the serum, but there are differences in those secretions that can tell where the secretion came from. Wow! Another interesting thing out of his talk was the complete genomic sequencing of a family of 4. He used principles of Mendelian genetics to remove 70% of the sequencing errors (which was validated by enriching using Agilent and re-sequencing).

David Flockhart reminded us how difficult it is to take genetic information (let alone genomic information) and affecting medicine. We have known for 20 years the benefit of knowing a genotype that significantly associates with a life-threatening bleeding risk for the drug warfarin. More than 20 clinical trials, GWAS, candidate gene approaches, websites, etc don't seem to be enough to get the FDA to mandate genetic screening. Oh, I also don't want to forget that this drug is used a lot, it is very toxic, and is the posterchild for pharmacogenomics. Are we doing all of this genomic sequencing for nothing?! We have got to come up with new ways of getting this kind of data into the standard of care.

Lastly was Russ Altman, who is in charge of the PharmGKB website. He took Steve Quake's genome (Patient 0), and linked the pharmacogenetic information from PharmGKB to the genotypes identified from his genomic sequence. What this allowed him to do was to generate a table of the SNPs identified in Quake's genome, and raised red flags for well known genetic variants that have demonstrated efficacy in predicting drug response/toxicity, but also included links to the papers from which the data came, the level of confidence associated with the risk (lets face it, some data is just better than others), and a couple other very important bits of information. These data are well curated by PhD-level scientists at PharmGKB. Just awesome!

[Joann]

pharmacogenomics poster child

Keynote Speaker:
David Flockhart reminded us how difficult it is to take genetic information (let alone genomic information) and affecting medicine. We have known for 20 years the benefit of knowing a genotype that significantly associates with a life-threatening bleeding risk for the drug warfarin. More than 20 clinical trials, GWAS, candidate gene approaches, websites, etc don't seem to be enough to get the FDA to mandate genetic screening. Oh, I also don't want to forget that this drug is used a lot, it is very toxic, and is the posterchild for pharmacogenomics. Are we doing all of this genomic sequencing for nothing?! We have got to come up with new ways of getting this kind of data into the standard of care.

To all: The FDA Advisory Committee on Pharmaceutical Science and Clinical Pharmacology meets March 17 in Atlanta, Georgia USA. the Committee will discuss and provide comments on the following topics: (1) General scientific issues related to the application of pharmacogenomics in the early stages of drug development. Pharmacogenomics examines the genetic differences that influence a person’s responses, both beneficial and harmful, to certain drugs; (2) a new patient-centric clinical pharmacology approach to drug safety.

FDA Advisory Committee meetings are open to the public and public comments related to issues before the committee are invited and may be submitted to the FDA and the committee thru Division of Dockets Management at: http://www.regulations.gov/search/Regs/home.html#home
using the indentification [Docket No. FDA–2010–N–0067]

More information about the advisory committee meeting and agenda see http://www.fda.gov/AdvisoryCommittee.../ucm201691.htm

Information on how to apply for membership to an FDA advisory committee can be obtained here:

http://www.fda.gov/AdvisoryCommittees/AboutAdvisoryCommittees/default.htm

[NGSfan]

thanks for sharing all that! Very interesting talks!