Geriatric Nutritional Risk Index (GNRI) and Survival in Pancreatic Cancer: A Retrospective Study

Malnutrition is a major contributor to poor treatment and survival outcomes in pancreatic cancer, yet nutritional assessment is not standardized or consistently implemented in the care of oncology patients. The Geriatric Nutritional Risk Index (GNRI), calculated from serum albumin and body weight, may be useful as a practical tool for identifying patients at risk of poor nutritional status. Purpose: To provide a preliminary analysis using a limited selection of variables to examine the association of the GNRI at diagnosis and the GNRI change over time with overall survival in patients with pancreatic cancer.

This retrospective study included 314 patients aged ≥18 years with pancreatic cancer. The GNRI was calculated at diagnosis and ≥30 days later. Patients were categorized by the GNRI at diagnosis (no risk >98, any risk ≤98) and change in the GNRI over time (no change/increase, mild decrease, and severe decrease). Additional variables included were demographics and stage. Comparative analysis included t-tests, chi-square tests, and ANOVA. Survival was analyzed using Kaplan-Meier curves, log-rank tests, and Cox proportional hazards modeling.

Median survival was significantly decreased in patients in the any-nutritional-risk group compared to the no-nutritional-risk group at diagnosis (442 vs. 1105 days), and those experiencing severe decreases in the GNRI scores compared to mild decreases and no change or increases (372.5 vs. 712 vs. 1791 days), respectively. Survival analysis stratified by the GNRI at diagnosis shows that both mild (HR 2.19, 95%, and CI 1.46-3.30) and severe decreases (HR 4.04, 95%, and CI 2.64-6.18) in the GNRI scores were independently associated with decreased survival versus no change or increase in the GNRI group after controlling for stage. Log-rank tests also show patients with any nutritional risk at diagnosis had significantly lower survival than those with no nutritional risk (p = 0.00052).

Lower GNRI scores showing greater nutritional risk at diagnosis and decreasing GNRI scores over time were predictors of decreased survival in pancreatic cancer. Our findings indicate that the GNRI may be valid and effective for the early identification of patients with a high nutritional risk who require nutritional interventions to improve outcomes in pancreatic cancer. However, more research is needed using larger samples and a greater variety of variables to confirm the presence and strength of this relationship, examine the effect of patient factors known to be associated with survival and nutrition, and explore potential influential confounders.