Gastrointestinal stromal tumors (GISTs) are mesenchymal neoplasms that generally harbor activating kinase mutations in either KIT or PDGFRA. GISTs respond to front-line treatment with the specific kinase inhibitor imatinib mesylate (IM). However, approximately 15% of adult GISTs lack mutations in these kinases and respond poorly to IM. Activating mutations in the serine-threonine kinase BRAF have been identified in a minority (~7%) of KIT/PDGFRA mutation-negative GISTs, suggesting that somatic mutations in other kinases may be responsible for the onset of a subset of GIST.
In Aim 1 of this proposal we propose deep exome sequencing of paired constitutional and GIST DNA from four adult patients whose tumors lack KIT/PDGFRA/BRAF mutations in order to identify somatic mutations that may be oncogenic. Although we hypothesize that dominant-active kinase mutations in these tumors are responsible for GIST oncogenesis, exome-wide sequencing will ensure that variants in all genes will be examined. We will also examine other mechanisms for oncogenesis, for example by capturing genes with somatic changes in the tumors in addition to novel germline variants in the patients. Variant analysis will be facilitated by comparisons with the existing SNP database, with the database of variants from the 1,000 Genomes project, and with the growing database of exome sequences available within our institution.
Genes with confirmed somatic mutations in the exome-wide analysis will be further examined in a candidate gene analysis as described in Aim 2 of this proposal. The goal is to determine the frequency and spectrum of genetic changes in these genes within a larger panel of GIST tumors, using as comparison constitutional DNA from these patients where available as well as a large panel of constitutional DNA from unaffected individuals. The outcome of these analyses will be a set of novel therapeutic targets for the treatment of GIST refractory to standard therapy.