With so many great speakers presenting, it is hard to choose which sessions to attend at the AACR meeting this year. Even my underlying interest in the application of Illumina technologies for cancer research is sometimes not enough to help choose between so many sessions. I guess the take-home message is that more and more researchers are using Illumina next-generation sequencing (NGS) and array technologies to advance their studies.
Some themes from this meeting are definitely emerging. There is a very real interest in personalized medicine or, as John Carpten from TGen called it, “genomic-enabled medicine.” John gave an inspiring talk on performing complete tumor/normal sequencing studies in a single experiment using the HiSeq 2500 system, achieving 100× coverage with whole-exome sequencing along with 5–8× coverage of long-insert reads and RNA-Seq to assess structural variants and gene fusions. Getting this kind of coverage is impressive in itself, but doing it in 27 hours is, in John’s own words, “awesome.” Arul Chinnaiyan from the University of Michigan reiterated this sentiment. The MI-ONCOSEQ project aims to provide individualized therapy based on complete tumor molecular profiling. Using the HiSeq 2500 system should enable them to reduce the turnaround time from four weeks to two—a crucial improvement for patient treatment decisions that await these results.
David Solit from Memorial Sloan-Kettering Cancer Center is using a reverse approach, identifying individuals with an exceptional treatment response (usually defined as complete sustained remission) and sequencing their tumors to understand why they responded well. Using this non-biased approach, he hopes to identify biomarkers for clinical trials.
Elaine Mardis from the Washington University School of Medicine prefers what she dubbed the “Maserati approach,” or performing whole-genome tumor/normal sequencing (90×/30×), exome sequencing for maximum read depth to assess heterogeneity and clonality, and transcriptome sequencing for gene expression and fusions. On the other end of the scale, Raju Kucherlapati from Harvard extolled the benefits of low-pass, whole-genome sequencing. In his experience, 5–8× coverage provides sufficient information on somatic CNVs and structural aberrations. Dr. Kucherlapati’s data indicate that this technique may provide important prognostic insight for certain cases.
Regardless of the approach taken, the common goal of each of these presenters is to use the genomic information to make informed decisions rather than relying on the standard-of-care model that is followed today. The results reported so far certainly show the value of this approach, although there has been much discussion on how to standardize and implement these approaches on a large scale.