Evaluation and validation of neutrophil to albumin ratio as a promising prognostic marker for all-cause mortality in patients with cancer: a multicenter cohort study

The practicality and effectiveness of using the prognostic value of the neutrophil-to-albumin ratio (NAR) in evaluating patients with cancer remain unclear, and research is needed to fully understand its potential application in the cancer population. The Kaplan–Meier method was used for survival analysis, and the log-rank test was employed for comparison. Univariate and multivariate Cox proportional hazards models were used to determine the prognostic biomarkers, and Logistic regression analysis was conducted to investigate the relationship between NAR and 90-day outcomes and cachexia. The study included 14,682 patients with cancer, divided into training (6,592 patients), internal validation (2,820 patients), and external validation groups (5,270 patients). Patients with high NAR had higher all-cause mortality than those with low NAR in the discovery (50.15% vs 69.29%, p<0.001), internal validation (54.18% vs 70.91%, p<0.001), and external validation cohorts (40.60% vs 66.68%, p<0.001). In the discovery cohort, high NAR was observed to be independently associated with all-cause mortality in patients (HR 1.16, 95% CI 1.12–1.19; p<0.001). Moreover, we validated the promising prognostic value of NAR as a predictor of survival in patients with cancer through internal validation (HR 1.21, 95% CI 1.16–1.27, p<0.001) and external validation cohorts (HR 1.27, 95% CI 1.21–1.34, p<0.001). Additionally, in the subgroup analysis by tumour type, high NAR was identified as a risk factor for most cancers, except for breast cancer. This study showed that NAR is a feasible and promising biomarker for predicting prognosis and cancer cachexia in cancer patients.