129 Characterization of PD-L1 expression in circulating tumor cells (CTCs) of non-small cell lung cancer (NSCLC) patients

( 1% cutoff) using a prototype IHC assay using the 22C3 antibody were randomized to receive pembrolizumab at 10mg/kg every 2 or 3 weeks; 2 pts were never treated. A new tumor biopsy 60 days prior to or, with Amendment 8, any time after the most recent therapy before the first pembrolizumab dose was required. Tumor response was assessed every 9 weeks by RECIST v1.1 by independent central review. All pts will have a minimum of 6 mo of follow-up. PD-L1 expression was subsequently evaluated in the study eligibility tumor samples by a clinical trial IHC assay using the 22C3 antibody for assessment of the relationship with outcomes with a goal of validating the utility of IHC staining in 50% of tumor cells as a biomarker to predict the efficacy of pembrolizumab. Results: After selection of the training set population, over 500 additional pts signed consent between 20 May 2013 and 12 May 2014 and provided tissue for PD-L1 assessment. Based on PD-L1 staining using the prototype IHC assay and other study eligibility criteria, 306 pts received pembrolizumab. Conclusions: At the meeting, we will present data evaluating whether the cutoff of PD-L1 staining by IHC in 50% of tumor cells that was selected based on our training set correctly predicts ORR and PFS in the validation set of 306 pts with advanced NSCLC treated with pembrolizumab.