Abstract The current diagnostics landscape for therapeutics that target the PD1/PD-L1 pathway is highly complex. Four different companion or complementary diagnostics have been developed for pembrolizumab in NSCLC, nivolumab in NSCLC and melanoma, and atezolizumab in urothelial carcinoma. The need to reconcile diagnostics for this class of targeted therapies has been recognized by the creation of the FDA-AACR-ASCO “PD-L1 Blueprint” working group to explore means to “harmonize” PD-L1 testing in tissue based IHC assays. The results reported from this working group noted similarities, but also several important discrepancies, between the current assays. Since each test uses a specific interpretation for each assay and indication, this creates a highly complex diagnostic landscape, which is likely to continue to increase in complexity as more PD-1/PD-L1 therapeutics and potentially novel diagnostics continue to be approved in additional indications. To address the need for adaptive, sustainable harmonization for PD-L1 diagnostics, Flagship Biosciences evaluated the utility of image analysis-based methods to harmonize multiple PD-L1 tests. We executed a proof-of-concept study utilizing a cohort of serial tissue sections from the same NSCLC patients, stained with the FDA approved Dako PD-L1 tests (28-8 and 22C3 clones), and our in-house PD-L1 assays (SP142 and E1L3N clones) for comparison. We digitized the tissue slides using a whole slide scanner, and evaluated the samples with our tissue Image Analysis (tIA™) technology. As expected, the patient samples stained with the separate PD-L1 assays yielded differences in staining and, thus, the reported scores for PD-L1 expression based on each test used, despite serial sections being derived from the same patient. To attempt to harmonize the scoring approaches for each test, we leveraged our computational Tissue Analysis (cTA™) platform to create a mathematically-derived “virtual slide score” for each sample, which enabled calibration of the various tests to deliver cross PD-L1 comparative scores. Based on the proof-of-concept demonstrated in this study, the cTA™ approaches could enable harmonization of the various PD-L1 tests through use of a digital pathology platform. The data presented provides a foundation for potential application of the cTA™ platform in the clinical laboratory setting to achieve harmonization of multiple PD-L1 tests. Citation Format: Nathan T. Martin, Joshua C. Black, Zachary Pollack, Famke Aeffner, Joseph Krueger. Evaluating "harmonization" of PD-L1 assays using image analysis [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 661. doi:10.1158/1538-7445.AM2017-661