Discovery of efficacy biomarkers for non-small cell lung cancer with first-line anti-PD-1 immunotherapy by data-independent acquisition mass spectrometry

First-line immune checkpoint inhibitors (ICIs) have greatly ameliorated outcomes in non-small cell lung cancer (NSCLC). However, approximately a quarter of patients receiving ICIs demonstrate long-term clinical benefit, and the true responders have not been fully clarified by the existing biomarkers. To discover potential biomarkers treatment-related outcomes in plasma, mass spectrometry assay for the data-independent acquisition was analyzed plasma samples collected before the anti-PD-1 treatment. From July 2019 to January 2020, 15 patients with EGFR/ALK-negative NSCLC receiving first-line anti-programmed cell death protein 1 (PD-1) inhibitors were enrolled, and six healthy individuals have collected the plasma samples as control. We explored plasma proteome profiles and conducted stratified analyses by anti-PD-1 responders and non-responders. To validate the target proteins by ELISA, we recruited 22 additional independent patients and 15 healthy individuals from April 2021 to August 2021. By identifying biomarkers to predict better efficacy, we performed differential expression analysis in 12 responders and three non-responders. Compared with healthy individuals, hierarchical cluster analysis revealed plasma proteome profiles of NSCLC were markedly changed in 170 differentially expressed proteins. Furthermore, we discovered that SAA1, SAA2, S100A8, and S100A9 were noticeably increased among non-responders than responders, which may serve as predictive biomarkers with unfavorable responses. The validated results from all samples via ELISA have confirmed this observation. Identified a set of plasma-derived protein biomarkers (SAA1, SAA2, S100A8, and S100A9) that could potentially predict the efficacy in cohorts of patients with NSCLC treated with first-line anti-PD-1 inhibitors and deserves further prospective study.