Variability of the prognostic nutritional index within 3 months after curative resection for non-small cell lung cancer and 2-year recurrence-free survival

Early recurrence after curative resection remains common in non-small cell lung cancer (NSCLC). We evaluated whether early postoperative trajectories of the prognostic nutritional index (PNI) improve risk stratification. In a single-center prospective cohort of patients undergoing R0 resection for stage I–III NSCLC (2021–2025), 119 were enrolled; 94 who were alive and recurrence-free at 3 months with complete PNI data formed the landmark cohort. PNI was measured preoperatively and during the first 3 postoperative months and calculated as 10 × serum albumin (g/dL) + 0.005 × total lymphocyte count (/mm 3 ). The primary exposure was 3-month PNI recovery, and the primary endpoint was 2-year recurrence-free survival (RFS) from the 3-month landmark, estimated with Cox models adjusted for stage and adjuvant therapy. Among 94 patients, 51 (54.3%) showed recovered PNI and 43 (45.7%) non-recovered PNI. During follow-up, 27 RFS events occurred, with higher event rates in the non-recovered group (41.9% vs. 17.6%; 2-year RFS 59.2% vs. 82.3%). Non-recovered PNI was independently associated with worse RFS (adjusted hazard ratio 2.3) and improved a clinical model C-index from 0.66 to 0.73. Sensitivity analyses using alternative cutoffs, continuous PNI change, complication adjustment, treatment-timing exclusions, and conservative Cox models were directionally consistent. Failure to recover PNI within 3 months may identify patients at higher risk of recurrence after curative NSCLC resection. Multicenter validation is required before clinical implementation.