Abstract 3664: A workflow to evaluate PD-L1 protein expression on circulating tumor cells (CTCs) from non-small cell lung cancer (NSCLC)

• Published in: Cancer research (Chicago, Ill.) Ill.), 2018-07, Vol.78 (13_Supplement), p.3664-3664
• Main Authors: Renier, Corinne, Wilkerson, Charles L., Hur, SJ Claire, Park, Da Eun Rachel, Lemaire, Clementine A., Matsumoto, Melissa, Carroll, James, Crouse, Steve, Goldman, Jonathan W., Garon, Edward B., Carlo, Dino Di, Christen-Sollier, Elodie
• Format: Article
• Language: English
• ISSN: 0008-5472; 1538-7445
• DOI: 10.1158/1538-7445.AM2018-3664

Introduction Many tumors evade immune surveillance by deploying immunosuppressive mechanisms and co-opting immune checkpoint pathways, such as PD-1/PD-L1. PD-1/PD-L1 pathway blockade is a highly promising therapy and has elicited durable antitumor responses in a variety of cancers. Challenges in advancing immunotherapies lie in patient stratification and monitoring therapy. Evaluation of PD-L1 expression is generally regarded as an inclusion criterion for clinical trials. Current methods rely on the analysis of excisional biopsies, which are invasive and carry associated medical risks. Isolation and analysis of CTCs is a noninvasive alternative that may provide, in real-time, more representative information on tumor biomarkers, and allow the monitoring of treatment effectiveness over time. Here, we present an integrated workflow for the analysis of PD-L1 protein expression on CTCs from metastatic NSCLC patients isolated with the Vortex Biosciences VTX-1 liquid biopsy system. Methods Cell lines expressing various level of PD-L1 and white blood cells prepared from healthy donors were used to optimize the PD-L1 immunostaining assay. Several PD-L1 antibodies were titrated, and their specificity estimated with isotype controls. PD-L1 expression quantification was evaluated using two independent methods i) Zen 2 imaging (Zeiss) and ii) CellProfiler software. Clinical validation was performed on blood samples collected from a cohort of metastatic NSCLC patients (stage IV) with known PD-L1 tumor status. CTCs were isolated using the VTX-1 Liquid Biopsy System, stained (PD-L1, CK, CD45, CD11b, DAPI) and classified, then PD-L1 expression was quantified. Results We have developed a PD-L1 immunofluorescent assay that shows specificity, sensitivity, and very good intra-assay repeatability. The two methods evaluated for the quantification of PD-L1 expression provided equivalent data and could be used interchangeably. The PD-L1 assay was further validated on CTCs isolated from a small cohort of metastatic NSCLC patients. Preliminary data shows heterogeneity in PD-L1 expression levels across CTCs from the same patient. Interestingly, among PD-L1 positive CTCs, both traditional CK+/PD-L1+ and CK-/PD-L1+ cells were observed. To move toward a more automated workflow for immunostaining, imaging, and fluorescence signal analysis, methods to collect CTC on glass slides were evaluated, with cell recovery ranging from 33 to 98%. Conclusion We demonstrated the feasibility of evalua