At last year’s meeting of the Association for Molecular Pathology, Marc Ladanyi, MD, spoke on The Cancer Genome Atlas (TCGA), a multi-institution effort funded by the National Cancer Institute and the National Human Genome Research Institute to catalogue genetic mutations responsible for a large number of human cancers. As chief of the molecular diagnostics service at Memorial Sloan-Kettering Cancer Center, Dr. Ladanyi is interested in translating the findings of this undertaking into molecular diagnostics.
He told the plenary audience, “There is a real gap between people who are investigators in this project, most of whom are specialists in cancer research or human genomics, and molecular diagnostics people.” The latter group is “underrepresented” on this panel, he said. In fact, Dr. Ladanyi is the only one.
The rationale for initiating the project, in 2006, was that it was “time to do cancer genomics in a more organized and systematic way,” he said. Three solid tumors were chosen for the pilot project: glioblastoma, squamous lung cancer, and serous cystadenocarcinoma of the ovary. The goal was to sequence up to 500 samples of each tumor. Work on glioblastoma is nearly complete, and the analysis of serous carcinoma of the ovary is well underway. Squamous lung cancer is on track to be completed by the time TCGA’s pilot phase ends late this year. Many additional cancers will be on the list during the project’s extended phase, which was initiated in fall 2009 and which will rely much more heavily on next-generation sequencing instruments from Illumina, ABI, and 454 Life Sciences.
Several known and some novel biomarkers were identified from the initial genomic analyses of 206 glioblastoma specimens, such as EGFR amplification, PTEN loss, and mutations in the genes for TP53, NF1, ERBB2, and PIK3R1 (Cancer Genome Atlas Research Network. Nature. 2008;455:1061–1068). “These results show the value of comprehensive genomic data integrated across a large sample set,” Dr. Ladanyi said.
Early results from the genomic analyses of 188 lung adenocarcinomas by another overlapping set of investigators have also been reported (Ding L, et al. Nature. 2008;455:1069–1075). In addition to the already known mutations in KRAS and EGFR, seven percent of samples had mutations in the NF1 gene, which was largely mutually exclusive of EGFR and KRAS and may identify a distinct subtype, Dr. Ladanyi said. Based on these recent genomic studies, “We will start getting requests for NF1 mutation detection in the near future in lung adenocarcinoma and other cancers,” he predicted.
Further work by an independent group correlated results from the proteomic analyses of 27 surgical samples of glioblastoma multiforme (GBM) to the genomic copy number and expression data from TCGA. “[T]hree subclasses of GBM emerge which appear to be associated with predominance of EGFR activation, PDGFR activation, or loss of the RAS regulator NF1,” these investigators reported (Brennan C, et al. PLoS ONE. 2009;4:e7752). “Each of these clusters has therapeutic implications,” Dr. Ladanyi said. He predicted increased demand for detection of a particular EGFR intragenic deletion, EGFRvIII, often seen in the “EGFR-activated” subset of GBM.
Dr. Ladanyi made one last forecast: “Molecular diagnostics laboratories stand to play a central role in translating these TCGA findings into clinical tests and therapeutic benefit.”
—William Check, PhD