Prognostic value of prognostic nutritional index in lung cancer: a meta-analysis

Background: Previous studies reported pretreatment prognostic nutritional index (PNI) was associated with clinical outcome of lung cancer. However, the result was not conclusive. We conducted a comprehensive meta-analysis to clarify the impact of PNI in lung cancer prognosis. Methods: We identified eligible studies by searching PubMed, EMBASE, and Web of Science, up to August 15, 2017. Overall survival (OS) and clinicopathological characteristics were collected from included studies. Pooled hazard ratios (HR) and corresponding 95% confidence intervals (CIs) were used to estimate clinical and prognostic value of PNI in lung cancer. Results: Ten studies including 5,085 patients were enrolled in the meta-analysis. The result demonstrated that low PNI was correlated with unfavorable OS in lung cancer (HR =1.72; 95% CI, 1.43-2.06; P=0.000), especially among non-small cell lung cancer patients (HR =1.93; 95% CI, 1.56-2.37; P=0.000). As for clinical characteristics, low PNI status was found related to gender (female versus male, HR =0.68; 95% CI, 0.5540.857; P=0.001) and histology (adenocarcinoma versus non-adenocarcinoma, HR =0.59; 95% CI, 0.47-0.74, P=0.000), but not smoking status (smoker versus never smoker, HR =1.49; 95% CI, 0.99-2.25, P=0.056). No significant publication bias was found (P=0.210). Conclusions: PNI was an independent prognostic indicator for lung cancer, and can serve as a novel biomarker to help guide clinical practice and promote clinical outcomes of lung cancer patients.