Association of the global inflammatory-nutritional index (GINI) with time to next treatment and overall survival in patients with metastatic colorectal cancer receiving third-line therapy: a real-world retrospective study

Background: Inflammation and immuno-nutritional status are key determinants of tumor progression and therapeutic resistance in metastatic colorectal cancer (mCRC). This study aimed to assess the prognostic performance of composite biomarker global inflammatory-nutritional index (GINI), which integrates multiple inflammatory, immune, and nutritional parameters, for predicting time to next treatment (TTNT) and overall survival (OS) in real-world mCRC patients receiving third-line therapy.

Methods: A comprehensive analysis was performed on 320 mCRC patients hospitalized at Shanghai Municipal Hospital of Traditional Chinese Medicine. Demographic data, vital signs, laboratory results, and diagnostic information were collected. Clinical characteristics were compared across GINI tertiles high GINI group (H group), moderate GINI group (M group), and low GINI group (L group). Kaplan-Meier survival analysis, Cox proportional hazards regression, time-dependent receiver operating characteristic (ROC) analysis, and restricted cubic spline (RCS) modeling were applied to systematically characterize the association between GINI and mortality, with subgroup analyses conducted accordingly.

Results: Significant differences were identified in 1-, 3-, and 5-year mortality rates across the three groups (P<0.05). Cox proportional hazards modeling indicated that GINI was a significant predictor of OS, with a statistically significant survival difference between the H and L groups (P<0.05). RCS modeling demonstrated a significant linear association between GINI and OS (P_Overall<0.01, P_Nonlinear≥0.05), and time-dependent ROC analysis showed favorable predictive performance at multiple follow-up time points. For TTNT following third-line therapy, significant differences were also observed among groups (P<0.05). Cox proportional hazards analysis showed that TTNT was significantly shorter in the H group than in the M and L groups (P<0.05). A nonlinear association was observed between GINI and TTNT following third-line therapy (P_Overall<0.05, P_Nonlinear<0.05), with better predictive performance at 6, 12, and 18 months. Subgroup analyses further demonstrated significant associations between clinical characteristics, including non-obese status and metastatic site, and prognostic outcomes, particularly supporting the relationship between GINI and TTNT following third-line therapy in patients with lung and bone metastases.

Conclusions: GINI is significantly associated with overall mortality risk and TTNT following third-line therapy in mCRC and may serve as a clinically informative indicator for prognostic assessment.