Prognostic nutritional index as a predictor of survival in metastatic gastroesophageal cancer patients treated with immunotherapy.

439 Background: The prognostic value of Prognostic Nutritional Index (PNI), which reflects both the nutritional and immunological status, has been shown in several cancers including gastroesophageal cancer (GEC). Its predictive value for survival in GEC patients treated with immunotherapy (IO) still needs be validated. Methods: A total of 130 metastatic GEC patients whose first-line treatments included IO from April 2017 to May 2025 were included. Pre-treatment serum albumin and total lymphocyte count were used to calculate PNI. The optimal cutoff value was determined with time-dependent receiver operating characteristic (ROC) curves. The area under the curve (AUC) was estimated using the Kaplan-Meier method and 95% CIs were derived via bootstrapping. The association between PNI and overall survival (OS) was assessed by the Kaplan-Meier method and Cox proportional hazards models, adjusting for age, sex, race (Asian vs non-Asian), primary tumor, de novo vs recurrence and histology. Results: This study included 31 patients with esophageal cancer, 47 with gastroesophageal junction cancer, and 52 with gastric cancer. Among these patients, 86 (66%) had de novo metastatic disease and 44 (34%) had recurrent metastatic disease. The median age was 60 years (IQR: 17). The overall median follow-up time was 26.5 months and the median OS was 16.4 months. ROC analysis using continuous PNI yielded the highest discriminative performance at 6 months (AUC: 0.81, 95% CI: 0.70-0.91). The optimal cutoff value for PNI was identified as 44, with an AUC of 0.75 (95% CI: 0.66-0.84) at 6 months. Patients were categorized into high-PNI (≥44, 62%) and low-PNI groups (<44, 38%). The low-PNI group had significantly worse median OS compared to the high-PNI group (7.7 vs 22.1 months; adjusted HR: 2.4, 95% CI: 1.5-3.8, p<0.001). Conclusions: PNI demonstrates potential as an early survival discriminator in GEC patients treated with IO, and may inform targeted interventions. Future studies should evaluate longitudinal changes in PNI in larger cohorts and assess the effectiveness of interventions aimed at improving PNI.