Targeting the Cancer Genome
Levi Garraway, M.D., Ph.D., assistant professor of medicine at Dana-Farber Cancer Institute, presented information from his group’s work to characterize the genomic alterations in melanoma. Statistically significant amplifications, deletions, and mutations identified from the study of 100 genomes are likely to be nonrandom, Dr. Garraway told the attendees, and several have been described in recognized oncogenes and tumor suppressor genes associated with melanoma. These include amplification of BRAF and MITF and alterations in the ETV1 transcription factor.
Knockdown of ETV1 has been shown to reduce melanoma cell proliferation, and ETV1 expression has been associated with substantial accumulation of the MITF oncogene. In the setting of MAP kinase activation, the addition of ETV1 can promote melanocyte transformation.
Dr. Garraway described ongoing research to identify small molecules that can target this signaling pathway at multiple levels. This work involves the use of microarrays to identify ligands that bind target transcription factors, and the use of SPR technology to measure small molecule/target protein binding events. The results suggest that ligand binding can reduce ETV1 expression. Dr. Garraway envisions the development of a “multipronged approach to attack a pathway and intercept it fully.”
Massively parallel next-generation sequencing is enabling whole-genome analysis and is replacing capillary- and array-based technologies, noted Dr. Garraway. It is helping researchers “characterize the RNA world in melanoma” and identify chromosomal rearrangements and novel fusion genes. Pilot studies have demonstrated the presence of highly chimeric transcripts and evidence of alternative splicing in fusion genes.
“Most gene fusions we see in melanoma are not recurrent,” said Dr. Garraway, emphasizing the importance of distinguishing between the genomic alterations that drive oncogenic transformation versus those that are merely passengers, or perhaps even “chauffeurs” or “alternative drivers,” which he described as alterations present in a particular tumor that cannot be generalized across tumors of that type.
Dr. Garraway underscored the need to weave together genetic and functional information, combining the results of cancer genome analysis with functional/phenotypic alterations identified in tumor cells using techniques such as pooled RNAi screening.
Ben Ho Park, M.D., Ph.D., associate professor of oncology at Johns Hopkins Hospital, described the development of models of somatic cell knock-ins of mutant PIK3CA at the meeting. PIK encodes the catalytic component of P13 kinase, and “hotspot mutations account for most alterations.” The P13 kinase pathway is involved in downstream AKT phosphorylation events.
To study these mutations and begin to search for targeted therapeutic interventions, Dr. Park’s group created somatic knock-in cell lines, in which they introduced the mutant form of the gene, yielding 50–50 expression of the mutant and normal gene, rather than overexpressing the gene of interest. Studies of these oncogenic PIK3CA mutant knock-in models have led to the identification of downstream signaling pathways that might contain targets for therapeutic intervention.
These model systems can be used to screen compound libraries to potential drug candidates. Dr. Park presented data showing that lithium, for example, selectively targets mutant PIK3CA knock-in cells.
Alex Adjei, M.D., Ph.D., senior vp of clinical research at Roswell Park Cancer Center, presented study data for a small molecule inhibitor of a mutant BRAF gene found in melanoma. The treatment led to 85% tumor shrinkage in visceral lesions that contained the BRAF mutation and produced no benefit in patients with tumors expressing wild-type BRAF.
Dr. Adjei provided other examples in which agents designed to target driver genes are generating positive responses. While these types of results are promising and demonstrate the ability to target specific genes and mutations, he echoed the views of the other speakers, highlighting the need to broaden this approach and target multiple points along disease-associated pathway. “Pathway redundancy” in therapeutic interventions will be crucial, he noted, as tumor cells are resilient, and when one pathway is blocked, they will use another.