A novel in vivo model of Glioblastoma radiation resistance identifies long non-coding RNAs and targetable kinases

Glioblastomas (GBM) are the most common primary CNS tumor. GBMs often recur as highly aggressive, intractable, therapy resistant tumors. Key molecular regulators of acquired radiation resistance in recurrent GBM are largely unknown with a dearth of accurate pre-clinical models. To address this, we generated eight GBM patient-derived xenograft (PDX) models of acquired radiation-therapy selected (RTS) resistance compared with same-patient, treatment naïve (RTU) PDX. A novel bioinformatics pipeline analyzed phenotypic, transcriptomic and kinomic alterations, identifying long non-coding RNAs (lncRNAs) and targetable, PDX-specific kinases. We observed differential transcriptional enrichment of DNA damage repair (DDR) pathways in our RTS models. Multiple molecular routes to acquired radiation-resistance were revealed in our models including PDX-specific kinases that we validated with targeted small molecule inhibitors (SMIs). We identified 184, mostly novel, lncRNAs differentially regulated between RTU and RTS PDX. Several of these lncRNAs were associated with transcriptional changes in DDR, cell cycle progression, stemness, and chromatin remodeling pathways. This study identifies lncRNAs as potential key regulators in recurrent GBM and therapy resistance. We also demonstrate that SMIs aimed at lncRNA-related signaling pathways may represent a novel therapeutic approach for recurrent GBM tumors. Overall design: Examination of baseline differences between patient matched primary and recurrent tumors following serial radiation selection