Proteomic profiling of early-stage non-small cell lung cancer identifies a high-performance protein signature associated with postoperative recurrence.

[BACKGROUND] The 5-year recurrence rate remains significantly high (∼30 %) in patients with early-stage Non-Small Cell Lung Cancer (NSCLC), even after complete tumor resection. Recurrence prediction primarily relies on pathological assessment and genomic abnormalities. However, proteins - the functional executors of genetic information - may offer additional prognostic value. In this study, we aimed to develop a model integrating proteomic and clinical features to improve recurrence prediction in early-stage NSCLC.

[METHODS] We recruited 351 stage Ⅰ NSCLC patients who underwent radical surgery in discovery corhort. An additional 103 participants from external prospective cohort were used for validation. Clinical data and follow-up outcomes were retrospectively collected. Tumor proteomics profiling was performed using liquid chromatography-mass spectrometry (LC-MS/MS). The proteomics data were acquired using a data-independent acquisition mode with a 150-minute gradient method and analyzed against the human UniProt database using DIA-NN (v1.8.1). We assessed the association between proteomic and clinicopathologic factors and disease-free survival (DFS) using Cox proportional hazards regression. A receiver operating characteristic (ROC) curve analysis was used to construct the predictive model.

[RESULTS] Of the 351 patients analyzed, 4260 differentially expressed proteins (DEPs) were identified as being associated with tumor recurrence. A nine-protein prediction model outperformed the clinicopathologic-based model (AUC, 0.898 vs. 0.742; P < 0.001) in predicting DFS. A combined model incorporating nine proteins and clinicopathological features demonstrated excellent predictive value for 5-year recurrence in the discovery cohort (AUC = 0.896). Nine proteins combined with clinicopathological features showed an AUC of 0.810 in the external validation cohort and an AUC of 0.844 in the combined cohort.

[CONCLUSION] Integrating tumor proteomics with clinicopathologic features enhances risk stratification and improves recurrence prediction after surgical resection of early-stage NSCLC. This approach may enable more personalized postoperative management through refined surveillance intervals and potential adjuvant therapies.