Predictive value of composite nutritional indicators geriatric nutritional risk index and controlling nutritional status for mortality risk in early-onset cancer survivors

Background Malnutrition represents a critical determinant of adverse clinical outcomes and substantial disease burden in cancer patients. Despite the established prognostic value of Geriatric Nutritional Risk Index (GNRI) and Controlling Nutritional Status (CONUT) as composite inflammation-immune-nutrition indices in elderly cancer patients, their utility in early-onset (EO) cancer survivors remains unclear. Methods This retrospective study evaluated GNRI and CONUT for predicting mortality in 3,273 early-onset (EO) cancer survivors, with a development cohort ( n = 2,814) from NHANES (1999–2018) and a validation cohort ( n = 459) from the Second Affiliated Hospital of Harbin Medical University (2011–2020). Patients were stratified by GNRI (< 98 vs. ≥ 98) and CONUT (≥ 2 vs. ≤ 1) and grouped into composite risk categories: High-risk (GNRI < 98 + CONUT ≥ 2), Moderate-risk (GNRI < 98 + CONUT ≤ 1 or GNRI ≥ 98 + CONUT ≥ 2), and Low-risk (GNRI ≥ 98 + CONUT ≤ 1). Results In the development cohort, GNRI < 98 and CONUT ≥ 2 independently predicted elevated risks of all-cause mortality (HR = 3.36, 95%CI = 2.69–4.19, P < 0.001), cancer-specific mortality, and non-cancer mortality. High-risk patients exhibited the poorest survival outcomes compared to Low-risk (all-cause mortality HR = 3.36, P < 0.001). Kaplan-Meier analysis confirmed worse prognosis in GNRI < 98, CONUT ≥ 2, and High-risk groups across all mortality endpoints. Validation cohort results aligned with these findings, reinforcing the prognostic significance of composite nutritional risk stratification. Conclusion This study is the first to validate GNRI and CONUT as effective composite inflammation-immune-nutrition indices for identifying high-risk EO cancer survivors. Composite stratification combining both indices enhances multidimensional inflammation-immune-nutrition risk assessment, offering a practical framework for prognostication and personalized care in this population.