Triple negative breast cancer (TNBC) accounts for 10-20% of all breast cancers, and is one of the most aggressive. Patients with metastatic TNBC are at high risk of death1,2 despite standard-of-care chemotherapies such as eribulin, a halichondrin analogue that inhibits microtubule dynamics. Previously, we found that selinexor (KPT-330), a potent inhibitor of exportin-1, synergizes with eribulin in TNBC cells and patient-derived xenografts (PDXs). Based on these studies, we launched a Phase Ib trial combining selinexor and eribulin, with expansion in TNBC (NCT02419495). Overall response rate was 10.5% in TNBC dose expansion (n = 19); however, two patients had exceptional responses with confirmed durable partial response, maintained long-term after transitioning to single-agent selinexor. We now propose to investigate the mechanism of action of this combination. We hypothesize that integrated analysis of PDXs and patient samples will give insights into the individual contributions of eribulin and selinexor to treatment efficacy, and into potential predictors and pharmacodynamic markers of response that can facilitate the design of follow-up trials using eribulin as chemotherapy backbone, in particular a randomized Phase II trial of eribulin with and without selinexor, to be launched through the Experimental Therapeutics Clinical Trial Network (ETCTN). To this end, we will (a) investigate whether selinexor enhances the efficacy of eribulin in TNBC PDXs established during the eribulin/selinexor trial, and whether efficacy of the combination in PDXs correlate with clinical efficacy; (b) compare baseline molecular features in PDXs and patient samples from responders and nonresponders; (c) compare pharmacodynamic changes in PDXs from responders and nonresponders following single-agent eribulin or selinexor and combination; and (d) compare pharmacodynamic changes in PDXs and in matching pre-treatment and on-treatment patient samples. The proposed study represents a PDXNet/ETCTN collaboration between Drs. Funda Meric-Bernstam and Senthilkumar Damodaran, and is consistent with our long-term goal to leverage PDXs and molecular profiling to identify rational combination therapies that will improve TNBC outcomes, as well as to identify patients who can benefit from such therapies. Taken together, we expect to better understand (a) whether selinexor improves efficacy of eribulin, (b) whether PDX efficacy and pharmacodynamic changes correlate with what is seen in patients, and (c) what tumor features correlate with response.