23 Validation of PD-L1 dynamic expression on extracellular vesicles as a predictor of response to immune-checkpoint inhibitors and survival in non-small cell lung cancer patients

BackgroundImmune-checkpoint inhibitors (ICIs) revolutionized the treatment of advanced non-small cell lung cancer (NSCLC).1–3 To date, tissue PD-L1 immunohistochemistry is one of the leading biomarkers for prediction of ICIs response but has several limitations.4 5Extracellular vesicles (EVs) are cell-derived structures involved in cell communication and represent a potential minimally invasive alternative to predicting ICI response.6–9 Based on this and our preliminary results presented at SITC 2020,10 we hypothesize that EV PD-L1 predicts response to ICIs in NSCLC.MethodsThis study evaluates an exploratory cohort of advanced/metastatic NSCLC patients receiving ICIs (cohort A) and a validation cohort receiving Pembrolizumab+docetaxel or docetaxel alone (PROLUNG Phase 2 randomized trial) (cohort B).11 Plasma samples were collected pre-treatment (T1) and at 3 treatment cycles (T2) (figure 1A). Response was assessed by computed-tomography scan at 3 (cohort A) and 6–8 treatment cycles (cohort B) according to mono- or chemotherapy combination therapy. Patients were classified as responders (partial, stable, or complete response) or non-responders (progressive disease) by RECISTv1.1.12 EVs were isolated by serial ultracentrifugation and characterized following ISEV recommendations.13,14 Tissue PD-L1 expression was measured by standardized immunohistochemistry (SP263, 22C3, or 28–8 clones)5 and EV PD-L1 expression by immunoblot and its ratio was calculated as EV PD-L1 T2/T1. Cut-offs from the exploratory cohort were applied to the validation cohort, being EV PD-L1 ratio