Association between Prognostic Nutritional Index and myelosuppression in gastric cancer patients undergoing chemotherapy: a retrospective cohort study.

Objective: To investigate the association between Prognostic Nutritional Index (PNI) and chemotherapy-induced myelosuppression in gastric cancer patients. Methods: This retrospective cohort study analyzed 562 gastric cancer patients receiving chemotherapy at two Chinese medical centers from January 2022 to December 2024. The exposure variable was PNI, calculated from serum albumin and lymphocyte count. The primary outcome was myelosuppression after the first chemotherapy cycle, defined according to CTCAE 5.0 criteria. Multiple logistic regression models adjusted for demographics, health status, tumor characteristics, treatment factors, and laboratory parameters. Results: Myelosuppression occurred in 75.1% of patients. After full adjustment, each one-unit increase in PNI reduced myelosuppression risk by 13% (OR = 0.87, 95%CI: 0.79-0.96, p = 0.004). Patients with PNI ≤ 48 had a significantly higher risk of myelosuppression (OR = 14.50, 95%CI: 4.93-42.65, p < 0.001). Significant effect modification was observed by sex (interaction p < 0.001), with stronger protective effects in males (OR = 0.71, 95%CI: 0.60-0.84). Conclusion: PNI is an independent predictor of chemotherapy-induced myelosuppression in gastric cancer patients, with a threshold of ≤48 identifying high-risk individuals. This readily available biomarker may guide personalized preventive strategies, particularly for male patients.