ABSTRACT Over the past decade, immunotherapy has recruited the patients’ own immune system in the fight against the tumor. In particular, the successful blocking of the interaction between programmed cell death-1 (PD-1) and its ligand PD-L1 has paved the way for the development of new treatments for several of the most devastating cancers. In cancers such as non-small cell lung cancer (NSCLC), upregulation of PD-L1, both in the tumor cells and in macrophages, reduces the host’s immune response, thereby enhancing tumor aggressiveness. While PD-1/PD-L1 checkpoint-blockade immunotherapy has become a real ‘game-changer’ for many cancer patients, response rates remain relatively low, ranging from 15 to 40%, depending on the type of cancer and the stage of the disease. The current assessment of whether a patient should be treated with PD-1/PD-L1 blockade therapy is based on expression levels of PD-L1 and degree of tumor-infiltrating immune cells using immunohistochemistry (IHC). This approach has significant downsides: 1) samples are surgically obtained, posing additional risk, and potentially triggering enhanced metastasis as a result of primary tumor perforation; and 2) these measures are rarely sensitive or specific enough in predicting response efficacy, being prone to generating false-positive results. In addition, IHC lacks a clean and quantitative definition of what threshold defines a PD-L1 positive tumor. It was shown that subsets of patients do not benefit from checkpoint-blockade immunotherapy despite testing positive for PD-L1 expression using IHC, and conversely, a significant number of patients responded well to it, despite testing negative for PD-1/PD-L1 in IHC-based tests. To spare patients from ineffective therapy and limit the number of those exposed to potential autoimmune side effects, the search for a reliable predictive biomarker allowing patient selection of those who would benefit from anti-PD-1/PD-L1 therapy has become more urgent than ever. In this proposal, we will use our ExoView technology to create an assay able to identify and quantify circulating exosomes that carry PD-L1. Exosomes have been shown to play a critical role in many pathologies and more recently in anti-tumor immunity. Furthermore, the vesicular nature of exosomes allows the identification of protein signatures which in turn may reveal valuable information about the parent cell type. Aim 1 will establish an antibody able to immobilize exosomes containing PD-L1 and validate the specificity and sensitivity of such antibodies in human plasma. Aim 2 will quantify PD-L1 positive exosomes originating from immune cells vs. tumor cells. Using various cellular markers will enable identifying the origin of PD-L1-expressing exosomes. Aim 3 will focus on Quantification of PD-L1 in vesicles from immune vs tumor cells in NSCLC plasma samples and correlate to PD-L1 IHC score and outcome of anti-PD-1/PD-L1 therapy.