Using Exome Sequencing to Improve Prediction of FOLFIRINOX First Efficacy for Pancreatic Adenocarcinoma

SIMPLE SUMMARY: Pancreatic ductal adenocarcinoma is one of the deadliest human cancers. The standard treatment for non-resectable tumors is based on usage of FOLFIRINOX (folinic acid, fluorouracil irinotecan oxaliplatin) chemotherapy. The aim of our retrospective study was to identify genomic markers associated with an improved survival in patients treated with FOLFIRINOX using whole exome sequencing (WES). We successfully generated WES analyses in 78 patients. We created and compared several models with clinical or genomic variables and pathways. While the clinical score was associated with overall and progression-free survival, the genomic score and pathways score were associated with the overall survival. The addition of genomic score improved the prediction of prognosis compared to the clinical score alone. Thus, our study showed that WES could provide useful information to predict survival in patients treated with FOLFIRINOX and might be used to select patients who could yield most benefit from FOLFIRINOX treatment. ABSTRACT: Purpose: The first line treatment of advanced pancreatic ductal adenocarcinoma cancer (PDAC) comprises a FOLFIRINOX regimen for most patients with good performance status. However, no biomarker to predict efficacy is currently available. We investigated whether exome sequencing could be used to predict progression-free and overall survival in patients undergoing FOLFIRINOX for PDAC. Methods: In this single-center observational study, we included 78 patients with advanced PDAC who underwent somatic and germline exome analyses during first line therapy with FOLFIRINOX or gemcitabine. Exome-derived variables associated with outcome were then used in Cox regression models to generate a composite biomarker. Results: Performance status, tumor stage, liver metastasis, and lung metastasis were retained to generate a prognostic clinical score associated with overall and progression-free survival. Clonality, ploidy, and copy number variant (CNV) signatures 1 and 5, as well as gene variants in the calcium, non-homologous end-joining (NHEJ), and spliceosome pathways, were retained to generate a genomic prognostic score. The addition of genomic score improved the prediction of prognosis compared to the clinical score alone. Conclusions: This study underlines that structural and pathway genomic features could be used to predict FOLFIRINOX survival in patients with advanced PDAC.