Abstract
PURPOSE: Cardiopulmonary toxic effects may reduce the efficacy of postoperative radiation therapy (PORT) in patients with non-small cell lung cancer (NSCLC). However, few studies have examined whether the heart and lung doses affect overall survival (OS). We investigated the correlation of heart and lung doses with OS in patients with NSCLC undergoing PORT. METHODS AND MATERIALS: This retrospective analysis included 307 patients with NSCLC undergoing PORT. The total dose was 50 Gy. Landmark analyses were performed at 36 months, with hazard ratios (HRs) calculated separately for events occurring up to 36 months (early survival) and after 36 months (long-term survival). Stabilized inverse probability of treatment weighting (sIPTW) was performed to balance the characteristics of the high- and low-dose groups. We performed sensitivity analyses at 24 and 48 months. RESULTS: The median follow-up period was 67.42 months. Heart doses were significantly correlated with long-term survival (HR, 1.14; P = .015) but not with early survival (HR, 0.97; P = .41) or whole survival (HR, 1.02; P = .58). Lung doses were marginally significantly correlated with early survival (HR, 1.03; P = .07) but not with long-term survival (HR, 1.00; P = .85) or whole survival (HR, 1.02; P = .12). Higher heart and lung doses were associated with decreased long-term and early survival, respectively, before and after sIPTW. Landmark analyses at 24 and 48 months showed consistent results. CONCLUSIONS: For patients with NSCLC undergoing PORT, a higher heart dose was associated with decreased long-term survival, whereas a higher lung dose was associated with decreased early survival.