Omics guided small molecule inhibitor screen for the identification of therapeutic vulnerabilities in metastatic lung adenocarcinoma.

Abstract

e21015 Background: Recent studies from our lab and others have demonstrated that double-strand DNA break accumulation, error-prone repair, and genomic instability are strongly linked to increased metastasis to distant organs. Our lab has shown that lung adenocarcinomas overexpressing the transcription factor MYBL2 display chronic replication stress, elevated error-prone repair, and widespread genomic instability. Importantly, this MYBL2-driven phenotype accounts for ̃21% of all lung adenocarcinomas and identifies aggressive disease enriched for metastases to brain, liver, and kidney. This study was performed to identify clinically actionable therapeutic vulnerabilities in MYBL2-driven metastatic lung adenocarcinoma. Methods: RNA-sequencing and proteomic data from the TCGA and ORIEN consortiums were mined to identify highly expressed druggable targets in MYBL2-driven lung adenocarcinomas. Identified targets were used to assemble a custom inhibitor library of 50 small molecules targeting DNA repair effectors, epigenetic factors, protein translation, kinase-signaling, autophagy, and post-translational modifications. Omics data from the Cancer Cell Line Encyclopedia was used to identify human cell line models (H1650 (pleura), H1568 (lymph node), H1299 (lymph node)) of MYBL2-driven metastatic disease. Inhibitors were tested at two doses, 5 uM and 500 nM, in triplicate, across replicate experiments. The PrestoBlue HS viability reagent was used to quantify cell viability in a 96 well plate format. The primary endpoint of this study was statistically significant cancer cell death, compared to vehicle controls. Results: Screen results demonstrated that, at nanomolar doses, inhibitors of protein translation have strong anti-tumor effects in MYBL2-driven disease. Interestingly, small molecules targeting EIF4G1 (SBI-0640756), ribosome biogenesis/RNA export (YM155), and rRNA synthesis (CX5461) were effective in inducing cell death while inhibitors blocking mTOR signaling, EIF2A phosphorylation, and EIF4F complex assembly were not. Bioinformatic analysis revealed that ̃60% of MYBL2-driven transcripts are dependent on EIF4G1 for translation. Importantly, effected transcripts include effectors controlling DNA repair, cell cycle coordination, and cell survival. Conclusions: Our data demonstrates that MYBL2-driven metastatic disease is uniquely sensitive to inhibitors of the protein translation machinery. Importantly, these inhibitors significantly outperform current standard-of-care agents cisplatin and pemetrexed. To our knowledge, our study is the first to demonstrate that blocking EIF4G1 effectively induces widespread cell death in metastatic lung adenocarcinoma models. Collectively, our work supports initiation of clinical trials testing the efficacy of SBI-0640756 in MYBL2-driven metastatic lung adenocarcinoma.