Abstract
The prognostic nutritional index (PNI) is a prognostic nutritional-immune marker, but its association with long-term mortality and phenotypic aging among adults with a history of digestive system cancer is uncertain. We analyzed 338 adults with self-reported digestive system cancer in NHANES 1999-2010 and 2015-2018. PNI was modeled continuously and using sample-derived categories (≤48, >48-52, and >52). In the fully adjusted Cox model, each 1-unit higher PNI was associated with lower all-cause mortality (hazard ratio [HR], 0.91; 95% confidence interval [CI], 0.88-0.95), while HRs were 0.47 (0.31-0.70) and 0.42 (0.28-0.62) for the middle and highest categories versus the lowest. Restricted cubic splines indicated nonlinearity, with the main separation occurring between the lowest category and the two higher categories. PNI correlated inversely with PhenoAgeAccel (Spearman ρ = -0.241). Exploratory pathway analysis suggested an indirect-effect component through PhenoAgeAccel; after removing albumin- and lymphocyte-related terms from the aging construct, the estimated indirect-effect proportion decreased from 26.3% to 16.1%. Higher PNI was associated with lower mortality, but the sample-derived thresholds and aging-pathway findings require cautious, noncausal interpretation.