Comparative value of postoperative inflammatory indices in predicting recurrence in T3 stage renal cell carcinoma: superiority of the Cally index

Abstract

Background This study evaluated the comparative efficacy of postoperative inflammatory indices (C-reactive protein-Albumin-Lymphocyte Index [Cally Index], Neutrophil-to-Lymphocyte Ratio [NLR], Platelet-to-Lymphocyte Ratio [PLR] and Systemic Inflammation Response Index [SIRI] ) in predicting disease-free survival in patients undergoing radical nephrectomy for T3 stage renal cell carcinoma (RCC). Methods We retrospectively analyzed 142 patients who underwent radical nephrectomy for T3 stage RCC between January 2015 and December 2023. Inflammatory indices measured on postoperative days 7-14 were calculated. Optimal cut-off values were determined using ROC curve analysis. Univariate and multivariate Cox regression analyses were performed for disease-free survival assessment. Results During a median follow-up of 36 months, 33.8% of patients developed recurrence. ROC analysis showed that the Cally Index (Area Under the Curve [AUC]: 0.73) demonstrated superior predictive accuracy compared to SIRI (AUC: 0.69), NLR (AUC: 0.70), and PLR (AUC: 0.66) (p < 0.001). In multivariate analysis, low Cally Index (< 4.5) emerged as the strongest independent prognostic factor (Hazard Ratio [HR]: 2.86, 95% CI: 1.92-4.26, p < 0.001), followed by Fuhrman grade 3-4 (HR: 2.14), SIRI (HR: 1.98), tumor size > 7 cm (HR: 1.82), NLR (HR: 1.76), PLR (HR: 1.65), and lymphovascular invasion (HR: 1.58). Harrell's C-index values for the Cally Index-based model were significantly higher than other index models. Conclusion Postoperative Cally Index shows superior prognostic value in predicting disease recurrence in T3 stage RCC compared to other inflammatory indices. This finding suggests that an integrated approach evaluating both inflammation and nutritional status may provide clinical benefit in postoperative risk stratification and individualized follow-up strategies. Trial registration Not applicable.