Establishment of a Novel Renal Immune Prognostic Index (RIPI) to Predict Clinical Outcomes in Renal Cell Cancer Patients who received Surgery

Yu Zhang^*

• Department of Nephrology, Second Affiliated Hospital of Harbin Medical University, Harbin, China

Objective: The aim of this study is to establish a novel Renal Immune Prognostic Index (RIPI) and investigate its predictive ability for the clinical outcomes of renal cell cancer (RCC) patients. Methods: This multicenter retrospective study included 259 RCC patients who underwent surgical resection at the Second Affiliated Hospital of Harbin Medical University (January 2016–December 2017) as the training cohort, and 350 patients from Harbin Medical University Cancer Hospital during the same period as the external validation cohort. The RIPI was developed using Cox regression with multicollinearity addressed by Lasso regression. The optimal cutoff was determined by Receiver Operating Characteristic (ROC) curve analysis. Survival differences were evaluated with Kaplan–Meier curves, and potential confounding factors were adjusted using Propensity Score Matching (PSM). Model performance and clinical utility were assessed using the concordance index (C-index), calibration curves, time-dependent ROC curves, and decision curve analysis (DCA). Results: Lasso regression identified prealbumin (PALB), lymphocyte count (LYM), and immunoglobulin M (IgM) as key hematological prognostic parameters. RIPI was constructed as: RIPI = 0.005 × PALB (g/L) + 0.248 × LYM (10⁹/L) + 0.372 × IgM (g/L). The optimal cutoff value of 4.96 stratified patients into low and high RIPI groups. In the training cohort, RIPI showed strong discriminatory ability with an AUC of 0.750, outperforming individual markers and conventional indices. Time-dependent ROC analysis demonstrated consistently higher predictive performance of RIPI across all time points. Kaplan–Meier survival analysis revealed that patients in the low RIPI group had significantly shorter progression-free survival (PFS) and overall survival (OS) (all P < 0.001), and RIPI remained an independent prognostic factor alongside tumor size and TNM stage. After PSM, RIPI continued to demonstrate significant associations with both PFS and OS. In the validation cohort, similar results were observed, with RIPI maintaining robust prognostic value (AUC = 0.723). Nomograms incorporating RIPI achieved good calibration and C-index values, while DCA confirmed its clinical utility. Conclusion: This multicenter retrospective study demonstrated that RIPI, integrating PALB, LYM, and IgM, provides robust and reproducible prognostic value in RCC patients. RIPI represents a reliable and clinically applicable tool for individualized risk stratification and outcome prediction.