Development and validation of a novel hepato-metabolic-renal score nomogram for predicting disease-free survival in head and neck squamous cell carcinoma

Background Head and neck squamous cell carcinoma (HNSCC) exhibits marked heterogeneity in disease-free survival (DFS), highlighting the need for improved prognostic tools. This study aimed to develop and validate an integrative physiological score for surgical HNSCC patients. Methods Using a single-center cohort of 321 surgical HNSCC patients randomly divided into training (n=225) and validation (n=96) cohorts, we developed the Hepato-Metabolic-Renal Score (HMRS) from 11 routine preoperative laboratory parameters via LASSO Cox regression. HMRS was integrated with clinicopathological factors into a prognostic nomogram. Performance was assessed by discrimination (time-dependent ROC), calibration, and clinical utility (decision curve analysis). Results HMRS demonstrated significant prognostic value for DFS in both training (HR = 2.907, 95%CI:1.693–4.991, P = 0.001) and validation (HR = 2.356, 95% CI: 1.082-5.130, P = 0.031) cohorts. The nomogram showed excellent discrimination (1/3/5-year AUCs: 0.913/0.899/0.896 in training, 0.952/0.891/0.923 in validation), good calibration, and positive net benefit on decision curve analysis. Subgroup analyses confirmed consistent performance across clinical subsets without significant interaction effects after multiple comparison correction. Conclusions The Hepato-Metabolic-Renal Score provides a novel, practical prognostic tool for surgical HNSCC, with its nomogram enabling personalized recurrence risk estimation. This integrative approach advances precision surgery strategies.