Abstract
Lung cancer is one of the most common and deadly malignancies worldwide, underscoring the need for reliable biomarkers that can inform prognosis and guide postoperative surveillance. This prospective study examined longitudinal changes in 10 tumor-associated autoantibodies in 71 patients with early-stage non-small-cell lung cancer (NSCLC) who underwent surgical resection. Blood samples were collected preoperatively and at 3-, 6-, and 12-month post-surgery. Enzyme-linked immunosorbent assays were used to measure serum autoantibodies against p53, MUC1, NY-ESO-1, APE1, PGP9.5, SOX2, GBU4-5, GAGE7, CAGE, and MAGE1. Logistic regression models assessed associations with 1-year recurrence, while Cox proportional hazards models evaluated overall survival. Substantial reductions in p53, GBU4-5, and CAGE autoantibodies correlated with lower recurrence risk and improved 1-year survival, even after false discovery rate adjustment (p < 0.05). NY-ESO-1 showed borderline significance for recurrence, and SOX2 was borderline for survival but did not remain significant after correction. These findings suggest that monitoring dynamic declines in certain autoantibodies (most notably CAGE) may offer clinically meaningful prognostic information following surgical resection. While further validation in larger, independent cohorts is required, our results highlight the potential of serial autoantibody profiling as a noninvasive tool for personalized postoperative management in early-stage NSCLC patients.