Discovery of New Fusion Transcripts in a Cohort of Pediatric Solid Cancers at Relapse and Relevance for Personalized Medicine

Abstract

Fusion detection in cancer is of utmost clinical importance. Dupain et al. validated a reliable tool, ChimComp, to predict their presence and oncogenicity from sequencing data, and they identified new potentially targetable fusions in a cohort of pediatric cancer patients at relapse. This will hopefully contribute to personalized medicine in cancer. © 2018 The American Society of Gene and Cell Therapy; We hypothetized that pediatric cancers would more likely harbor fusion transcripts. To dissect the complexity of the fusions landscape in recurrent solid pediatric cancers, we conducted a study on 48 patients with different relapsing or resistant malignancies. By analyzing RNA sequencing data with a new in-house pipeline for fusions detection named ChimComp, followed by verification by real-time PCR, we identified and classified the most confident fusion transcripts (FTs) according to their potential biological function and druggability. The majority of FTs were predicted to affect key cancer pathways and described to be involved in oncogenesis. Contrary to previous descriptions, we found no significant correlation between the number of fusions and mutations, emphasizing the particularity to study pre-treated pediatric patients. A considerable proportion of FTs containing tumor suppressor genes was detected, reflecting their importance in pediatric cancers. FTs containing non-receptor tyrosine kinases occurred at low incidence and predominantly in brain tumors. Remarkably, more than 30% of patients presented a potentially druggable high-confidence fusion. In conclusion, we detected new oncogenic FTs in relapsing pediatric cancer patients by establishing a robust pipeline that can be applied to other malignancies, to detect and prioritize experimental validation studies leading to the development of new therapeutic options. © 2018 The American Society of Gene and Cell Therapy