Addition of Primary Metastatic Site on Bone, Brain, and Liver to IMDC Criteria in Patients With Metastatic Renal Cell Carcinoma: A Validation Study.

Abstract

The International Metastatic Renal Cell Carcinoma Database Consortium (IMDC) criteria have been largely adopted in clinical practice. In a recent retrospective study, we assessed that the addition of the first site of metastatic disease to brain, bone, and liver improves prognostic stratification of patients with metastatic renal cell carcinoma (mRCC). Here, we performed an external validation in patients with mRCC. Our aim was to evaluate if the addition of a new independent variable could improve IMDC prognosis prediction and reduce heterogeneity within risk categories. We selected all 1073 patients treated at a single institution for mRCC and included in the Institute Gustave Roussy Renal Cell Carcinoma database. All patients included received at least 1 line of targeted therapy or immune checkpoint inhibitors. Univariate and multivariate analyses (Cox regression model) were performed. Bootstrap validation of the final model was also carried out for internal validation. The IMDC modified classification was defined by the addition of the seventh variable, and we defined the modified IMDC good-risk criteria as 0 risks, intermediate-risk as 1 to 2 risks, and poor-risk as 3 or more risks. The presence of brain, bone, and/or liver as the first site of metastatic disease plus the other variables included in the IMDC score were statistically significant variables associated with overall survival (OS) after univariate and multivariate analysis and bootstrap validation. Finally, 122 (15%) patients had a modification of their initial risk category. The median OS in the poor-, intermediate-, and favorable-risk groups was 10, 26, and 52 months, respectively (P< .001). The bias-corrected concordance index in patients receiving immune checkpoint inhibitors (n= 241) was 0.71. The addition of brain, bone, and/or liver metastases as an additional variable to the other IMDC variables improves the prognostic predictive power of the model.