Common nutritional/inflammatory indicators are not effective tools in predicting the overall survival of patients with small cell lung cancer undergoing first-line chemotherapy.

Abstract

Various studies have investigated the predictive significance of numerous peripheral blood biomarkers in patients with small cell lung cancer (SCLC). However, their predictive values have not been validated. This study assessed and evaluated the ability of common nutritional or inflammatory indicators to predict overall survival (OS) in patients with SCLC who received first-line chemotherapy. Between January 2008 and July 2019, 560 patients with SCLC were enrolled at the Sichuan University West China Hospital. Eleven nutritional or inflammatory indices obtained before chemotherapy were evaluated. The cutoff values of continuous peripheral blood indices were confirmed through maximally selected rank statistics. The relationship of peripheral blood indices with OS was investigated through univariate and multivariate Cox regression analyses. Harrell's concordance (C-index) and time-dependent receiver operating characteristic curve were used to evaluate the performance of these indices. A total of 560 patients with SCLC were enrolled in the study. All the patients received first-line chemotherapy. In the univariate Cox analysis, all indices, except the Naples score, were related to OS. In the multivariate analysis, albumin-globulin ratio was an independent factor linked with prognosis. All indices exhibited poor performance in OS prediction, with the area under the curve ranging from 0.500 to 0.700. The lactic dehydrogenase (LDH) and prognostic nutritional index (PNI) were comparatively superior predictors with C-index of 0.568 and 0.550, respectively. The LDH showed incremental predictive values, whereas the PNI showed diminishing values as survival time prolonged, especially for men or smokers. The LDH with highest sensitivity (0.646) and advanced lung cancer inflammation index (ALI) with highest specificity (0.952) were conducive to identifying death and survival at different time points. Common inflammatory or nutritional biomarkers are only marginally useful in predicting outcomes in patients with SCLC receiving first-line chemotherapy. Among them, LDH, PNI, and ALI are relatively promising biomarkers for prognosis evaluation.