BackgroundNovel biomarkers, such as plasma microRNAs (miRs), are needed to help guide clinical decision-making for the type of chemotherapy to use in patients with advanced pancreatic ductal adenocarcinoma (PDAC). This study assessed the ability of plasma miRs to predict optimal treatment response from FOLFIRINOX or gemcitabine-nab-paclitaxel in these patients. MethodsNext-generation sequencing (NGS) was performed for biomarker discovery in pre-treatment plasma samples from advanced PDAC patients subsequently treated with FOLFIRINOX (n = 12) or gemcitabine-nab-paclitaxel (n = 12). Selected candidate biomarkers were validated in 40 patients with advanced PDAC using RT-qPCR. Cox regression was then used to assess the predictive value of plasma miRs for either FOLFIRINOX or gemcitabine-nab-paclitaxel. ResultsIn the validation cohort, high plasma miR-379 expression was strongly predictive of treatment response (Pinteraction = 0.0004). Overall survival (OS) was significantly better with FOLFIRINOX vs. gemcitabine-nab-paclitaxel in those patients with lower plasma miR-379 expression (hazard ratio, 0.32 [95% confidence interval, 0.08 to 0.98]; P = 0.046). However, gemcitabine-nab-paclitaxel was associated with superior OS in patients with higher plasma miR-379 (hazard ratio, 0.28 [0.10 to 0.86]; P = 0.027). In contrast, miR-127, miR-155, and miR-200 showed no predictive value for treatment response for either chemotherapy regimen (Pinteraction = 0.12, Pinteraction = 0.83 and Pinteraction = 0.12, respectively). ConclusionsPlasma miR-379 appears clinically useful as a predictive biomarker to identify which patients with advanced PDAC benefit most from treatment with FOLFIRINOX or gemcitabine-nab-paclitaxel. Further validation in larger studies and clinical trials is now warranted.