Abstract Many targeted cancer therapies rely on biomarkers, which are assessed by standard pathologist scoring of immunohistochemically stained tissue. However, this process is subjective, semi-quantitative and does not assess expression heterogeneity. A quantitative method to measure IHC markers might therefore significantly improve patient selection particularly of proteins expressed at low levels. To address these challenges, we have developed the Quantitative Continuous Scoring (QCS) that deploys the power of fully supervised Deep Learning (DL) algorithms to provide objective and continuous data of biomarkers in digitized IHC whole slide images (WSI). The two DL-based algorithms, developed using pathologist input as the ground truth, identify areas of invasive tumor and segment each individual tumor cell across the WSI into pixels that represent cell nuclei, cytoplasm and/or membrane. This allows to compute biomarker expression as mean Optical Density (OD) in each of these subcellular compartments based on the Hue-Saturation-Density (HSD) model. Of note, this also allows computation of the spatial distribution of tumor cells across the WSI. The measured OD for each cell is aggregated as a histogram to quantitative continuous readouts for each patient sample. The method’s ability to accurately detect low expression range facilitates selection of antibody clones for IHC assays, has been successfully used to delineate mode of action and PK/PD mechanisms, has provided surrogate markers of spatial expression heterogeneity to predict potential bystander activity and has facilitated marker co-expression analysis to inform rational combination therapies. In retrospective clinical trials analysis, QCS showed superior performance in identifying a patient population gaining maximum treatment benefit. QCS-based quantification of PD-L1 membrane expression was able to stratify anti-PD-L1 treated late-stage non-small cell lung cancer (NSCLC) patients [NCT01693562] with a higher prevalence and more significant log rank p-value (64%, p=0.0001) for OS compared to pathologist TPS (59%, p=0.01). In summary, we describe a computational pathology-based approach for precise biomarker quantification and superior patient selection with broad applicability and the potential to transform pathology, thus addressing one of the key challenges of precision oncology. Citation Format: Hadassah Sade, Ansh Kapil, Philipp Wortmann, Andreas Spitzmueller, Nicolas Triltsch, Lina Meinecke, Susanne Haneder, Anatoliy Shumilov, Jan Lesniak, Valeria Bertani, Tze-Heng Tan, Ana Hidalgo-Sastre, Simon Christ, Andrea Storti, Regina Alleze, Dasa Medrikova, Jessica Chan, Simon Lanzmich, Markus Schick, Guenter Schmidt, J. Carl Barrett. Quantitative assessment of IHC using computational pathology allows superior patient selection for biomarker-informed patients [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 468.