Beyond IMDC: do expanded models improve survival discrimination in metastatic renal cell carcinoma? A real-world cohort study.

Background: The International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) classification remains the cornerstone of prognostic assessment in metastatic renal cell carcinoma (mRCC). However, whether metastatic burden and host-related inflammatory and nutritional biomarkers meaningfully improve prognostic discrimination beyond IMDC has not been fully established. This study aimed to evaluate whether expanded prognostic models incorporating metastatic burden, bone metastasis, neutrophil-to-lymphocyte ratio (NLR), albumin, and lactate dehydrogenase (LDH) improve survival discrimination beyond IMDC in a real-world mRCC cohort. Methods: We retrospectively evaluated 104 patients with mRCC treated at a single tertiary oncology center. Overall survival (OS), defined from the diagnosis of metastatic disease, was analyzed using Kaplan-Meier analysis and Cox proportional hazards regression models. Sequential multivariable Cox models were constructed by progressively adding metastatic burden, NLR, albumin, and LDH to the IMDC model. An alternative model incorporating bone metastasis status was also evaluated. Model performance was assessed using Harrell's concordance index (C-index), -2 Log Likelihood statistics, and likelihood-ratio (LR) tests. Results: The cohort consisted predominantly of male patients (69.2%), with a median age of 60.3 years at metastatic diagnosis. According to IMDC classification, 24.0% of patients were favorable risk, 51.9% intermediate risk, and 24.0% poor risk. Median OS was 85.36 months in the favorable-risk group, 38.11 months in the intermediate-risk group, and 18.27 months in the poor-risk group (log-rank p<0.001). Elevated NLR was associated with significantly shorter OS in Kaplan-Meier analysis and with increased mortality in univariate Cox analysis (HR = 1.86, 95% CI 1.09-3.20, p=0.024), while higher serum albumin was associated with improved OS in univariate analysis (HR = 0.53, 95% CI 0.30-0.91, p=0.021). The fully expanded model demonstrated the highest discrimination (Harrell's C-index 0.720), compared with 0.664 for the IMDC-only model. However, none of the sequential additions beyond IMDC significantly improved model performance according to likelihood-ratio testing, and metastatic burden, bone metastasis, NLR, albumin, and LDH did not retain independent prognostic significance after multivariable adjustment. Conclusions: In this real-world mRCC cohort, IMDC remained the dominant determinant of survival outcomes. Although expanded prognostic models incorporating metastatic burden and host-related biomarkers provided modest numerical improvements in model discrimination, they did not confer statistically significant incremental prognostic value beyond IMDC. These findings support the continued central role of IMDC for prognostic stratification in routine clinical practice.