Abstract
BACKGROUND: Immunotherapy has significantly changed the treatment prospects of non-small cell lung cancer (NSCLC). However, there is no report based on immune score to predict the overall survival (OS) of lung adenocarcinoma (LUAD) in the stages I, II, and III. Therefore, this study aimed to investigate the immune score and the prognosis-related factors of LUAD and construct a nomogram to predict the prognosis. METHODS: A total of 390 cases with lung adenocarcinoma in the stages I, II, and III were included in the study. The clinicopathological characteristics and immune scores of LUAD patients were downloaded from the TCGA database. Cox proportional hazards regression model was used to estimate hazard ratio (HR) and 95% confidence interval (CI). A Nomogram was composed of the Cox model and internally validated using 1000 bootstrap. The concordance index (c-index) and the calibration curves were used to evaluate the model. The decision curve analysis (DCA) was performed to evaluate the clinical practical value of the model. RESULTS: According to the immune score, the patients were divided into low-, medium-, and high-score groups. This study showed that compared with patients with low and medium immune scores, only patients with high immune scores had significantly improved OS (HR and 95% confidence interval (CI): 0.489 [0.324-0.737]). The C-index for OS prediction was 0.691 (95% CI, 0.646-0.736). The calibration curves for nomogram-predicted probabilities of 3- and 5-year survival have good ability for the calibration and discrimination. CONCLUSION: The high immune score was significantly correlated with better OS of patients with LUAD in the stages I, II, and III. Moreover, the nomogram of predicting prognosis may help assess the survival of LUAD patients.