Abstract Background: Treatment of the mesenchymal-derived gastrointestinal stromal tumor (GIST) is based on targeting alterations in c-KIT and PDGFRA, present in up to 85% of these cancers. Immunohistochemistry (IHC) to assay expression of c-KIT is the current standard of care for defining GIST. IHC does not identify all potential cases with driver mutations in c-KIT and PDGFRA, and will miss other potentially actionable drivers such as BRAF mutations. Moreover, alterations in c-KIT and PDGFRA outside of “hot spot” mutations may alter sensitivity to tyrosine kinase inhibitors (TKIs), requiring alternative therapeutic approaches. Even c-KIT-mutant GIST may harbor other potentially actionable mutations. NCCN consensus guidelines do not yet fully endorse upfront genomic analysis nor reevaluation of the genomic profile upon disease progression, potentially leaving a subset of GIST patients with suboptimal treatment options. Methods: Comprehensive genomic profiling was performed on formalin-fixed, paraffin embedded tissue from patients diagnosed with GIST by histology using the CLIA-certified FoundationOne® platform (Foundation Medicine, Cambridge, MA) for targeted sequencing of the entire coding sequence of 236 genes and 47 introns of 19 genes involved in fusions. A subset was profiled on the expanded panel of 315 genes and introns of 28 genes involved in fusions. Clinical history and profiling data were presented at a multidisciplinary molecular tumor board for development of therapeutic recommendations. Results: The primary site of histologically-defined GIST included three gastric, five intestinal, and one pelvic. Five of eight cases (62.5%) had driver mutations in c-KIT but none had identifiable PDGFRA mutations. An average of 2.8 mutations were seen in the c-KIT mutant GIST and 2 mutations in the c-KIT wild type GIST. All cases, except for an intestinal atypical GIST, were positive for c-KIT expression by IHC regardless of profiling results. The atypical GIST was reclassified as ovarian carcinosarcoma after genomic profiling of the synchronous intestinal “GIST” and ovarian squamous cell carcinoma revealed six identical alterations and lack of traditional GIST mutations. In the remaining two cases lacking c-KIT or PDGFRA mutations, both had potentially actionable alterations in the MAPK pathway: one had BRAF V600E and one had an NF1 truncating mutation (NF1 Q315*), the latter in a patient with clinical diagnosis of neurofibromatosis. Both arose in the small bowel. CDKN2A/B/C alterations were found in 29%. Additional genomic alterations were observed in genes in the PI3K/mTOR, chromatin remodeling, and signal transduction pathways. Six of 8 patients initially received imatinib therapy. For cases found to be wild type for both c-KIT and PDFGRA on subsequent genomic profiling or deemed non-GIST, imatinib was discontinued after review at the Rutgers Cancer Inst. of NJ Molecular Tumor Board. Three patients (37.5%) had progressive disease, two with c-KIT mutation and one with BRAF V600E. c-KIT-altered cases with recurrent and progressive disease despite TKI therapy displayed alterations in CDKN2A/B/C, suggesting that this and BRAF alteration may be markers for poor prognosis. Conclusion: Our series of pathologically defined GIST underscores the role of genomic profiling for understanding GIST pathogenesis and developing a therapeutic approach. CDKN2A/B/C alteration was found in a subset of recurrent tumors, and suggests that it may be both a marker of resistance and an additional potential target for therapy. Our data suggest that molecular profiling should be performed upon initial diagnosis to improve clinical outcomes, identify those patients at highest risk of relapse, and to reduce potential side effects in the setting of ineffective therapy, particularly for imatinib-resistant or IHC-positive/c-KIT wild type GIST. Citation Format: Aleksander Chojecki, Rebecca A. Moss, Hua Zhong, Elizabeth Poplin, Ellen Ronnen, Shridar Ganesan, Lorna Rodriguez Rodriguez, Kim M. Hirshfield. Role of upfront genomic analysis of gastrointestinal stromal tumors to identify imatinib-insensitive or resistant cases. [abstract]. In: Proceedings of the AACR Precision Medicine Series: Integrating Clinical Genomics and Cancer Therapy; Jun 13-16, 2015; Salt Lake City, UT. Philadelphia (PA): AACR; Clin Cancer Res 2016;22(1_Suppl):Abstract nr 08.
Read full abstract