Proteome-wide Mendelian randomization identifies protein associations and therapeutic targets for B-cell malignancy

Despite therapeutic advances in B-cell malignancies, many patients continue to experience relapse or refractory disease, highlighting an unmet need for novel drug targets. The high attrition rate of drug development programs, however, represents a productivity-limiting step. To prioritize candidate drug targets, we used Mendelian randomization to evaluate causal associations between 2923 circulating proteins and 6 B-cell malignancies (22 922 cases and 388 978 controls). We identified 27 protein-disease associations, including TNFSF13 (APRIL) and TNFRS13B (TACI) in multiple myeloma, CD40 in Hodgkin lymphoma, and FAS in chronic lymphocytic leukemia. All results are interactively accessible via our R/Shiny application (https://software.icr.ac.uk/app/mrcan). Integrating single-cell RNA sequencing from 183 355 hematolymphoid cells, Bayesian colocalization, and clinical trial evidence, we prioritized 23 proteins as candidate drug targets. This study demonstrates the potential of human genetics to guide therapeutic discovery for B-cell neoplasia.