RESUMEN
BACKGROUND: Immune checkpoint inhibitors (ICIs) have achieved important outcomes in cancer treatment. However, current clinical biomarker tests are not suitable for some patients because they require tumor tissues and have poor predictive value for treatment responses. Therefore, the identification of biomarkers that enable screening tests in all patients is necessary. METHODS: We performed an immune complexome analysis of non-small cell lung cancer patients treated with nivolumab to comprehensively identify and compare antigens incorporated into immune complexes (IC-antigens) in serum samples from the responders (n = 15) and non-responders (n = 20). Additionally, combinations of IC-antigens characteristic to the responder group were evaluated by logistic regression analysis and receiver operating characteristics curves to examine their predictiveness for ICI treatment responses. RESULTS: The combination of predictive biomarkers detected before treatment was profilin-1, purine nucleoside phosphorylase, alpha-enolase, and nucleoside diphosphate kinase A [p = 0.0043, odds ratio = 2.26, 95% confidence interval (CI) = 1.19-4.28, area under the curve = 0.76]. The combination of predictive biomarkers detected after treatment was peptidyl-prolyl cis-trans isomerase A, ubiquitin-like modifier-activating enzyme 1, complement component C8 beta chain, and apolipoprotein L1 (p = 0.0039, odds ratio = 2.56, 95% CI = 1.25-5.23, area under the curve = 0.77). CONCLUSION: Combinations of serum IC-antigens may predict the therapeutic effect of nivolumab in non-small cell lung cancer patients.