13 Whole exome sequence analysis of canine transitional cell carcinoma of the bladder

Abstract

associated drug-response relationships. One specimen was tested in vitro and in vivo using a matched, patient-derived xenograft model. Results: NGS for the 11 tumor panel found ~30% of cases with EGFR amplification, p16, and/or PTEN deletion. Novel potentially actionable targets included a TRIM54-FGFR3 fusion, STAG2 mutation, MDM2 amplification, KIF11 and KIF15 mutation, and BRAF amplification. GBM150 is derived from a recurrent tumor and harbors the TRIM54-FGFR3 fusion, and this line is the first to be analyzed for chemovulnerability. In a Cyquant proliferation assay, GBM150 were relatively resistant to TMZ (IC50 300 uM) but significantly more sensitive to pan-FGFR1−3 inhibitors as compared to GBM108: ponatinib IC50 0.3 uM vs. 1.2 uM; AZD4547 IC50 3 uM vs. >10 uM. Similarly, in a flank tumor regrowth study, established GBM150 tumors were randomized to therapy with placebo, temozolomide, ponatinib or AZD4547. While TMZ-treated tumors were marginally smaller than placebo treated tumors (mean tumor volume 1312±208 vs. 992±183mm3, 97 days after therapy initiation), both ponatinib or AZD4547 resulted in sustained tumor stasis with a mean tumor volume of 240±35 and 199±40mm3, respectively, at the same time-point. Conclusions: These results support the concept that NGS can be used to individualize treatment of GBM and highlight how corresponding patient derived xenograft models can be used to validate the accuracy of potential therapeutic predictions.