Autoantibody-driven prediction of immunotoxicity from PD-1 treatment in patients with Non-Small-Cell Lung Cancer—a precision proteomics approach

Background: Immunotherapy by Immune Checkpoint Inhibitors (ICIs) has revolutionized the treatment landscape of non-small-cell lung cancer (NSCLC), thereby significantly improving survival outcomes and offering renewed hope to patients with advanced disease. However, patients experience limited long-term benefits from ICIs due to the development of primary or acquired immunotherapy resistance, as well as adverse effects (irAEs). In this study, we investigated the immunological status at baseline and during treatment of NSCLC patients receiving ICI treatment following the prediction of the predictive biomarkers associated with irAEs of an autoimmune nature.

Methods: In the LIMBIO study, NSCLC patients (n=51) received ICI treatment up to X months. Patient samples were collected at baseline and endpoint with or without irAEs and immunologically profiled using biomarker assays, quantitative mass spectrometry-based proteomics, and high-density autoantibody reactive protein arrays to characterize potential irAE biomarker panels.

Results: A total of 76% developed irAEs, whereby a large fraction were autoimmune in nature. Based on the quantitative plasma proteomics data of the NSCLC cohort obtained from investigating the patients, we could not distinguish between the irAE and the non-irAE groups, and the proteins most identified were found from comparisons between endpoint and baseline. Global oncogene-focused protein arrays, however, could identify differentially expressed autoantibodies between the two baselines, suggesting that the patients in the irAE group are predisposed for irAEs due to the high presence of preexisting autoantibodies at baseline.

Conclusions: Our proteomic investigation revealed changes and an association between inflammatory signaling and the dynamic variation of autoantibodies between baseline patients and irAE-affected patients. An elevated level of autoantibodies may serve as predictive biomarker candidates for the prediction of ICI-induced irAEs.