Comparative analysis of comorbidity and performance indices for prediction of oncological outcomes in patients with upper tract urothelial carcinoma who were treated with radical nephroureterectomy

OBJECTIVE: Comorbidity and performance indices (CPIs) are useful tools to evaluate patient׳s risk of comorbidities and thus may guide clinical decision making regarding surgery or multimodal therapy approaches. Hence, the aim of the current study was to assess the predictive capacity of CPIs comprising the American Society of Anaesthesiologists (ASA)-score, the Charlson comorbidity index (CCI), the age-adjusted CCI (ACCI), and the Eastern Cooperative Oncology Group performance status (ECOG-PS) in patients with upper tract urothelial carcinoma (UTUC) who were treated with radical nephroureterectomy (RNU).

METHODS AND MATERIALS: A total of 242 patients with UTUC underwent RNU without neoadjuvant chemotherapy between 1992 and 2012 at 3 German academic centers. Patients were stratified according to the pre-RNU CPIs dichotomized as ASA 1/2 vs.≥3, CCI 0 to 2 vs.>2, ACCI 0 to 5 vs. >5, and ECOG-PS 0 to 1 vs. >1. We assessed the associations of CPIs with clinicopathologic features, as well as the prognostic effect on recurrence-free survival, cancer-specific survival (CSS), overall survival, and cancer-independent mortality (CIM), using univariable and multivariable Cox regression analyses.

RESULTS: Sixty-two patients (25.6%) had an ASA-score≥3, 71 patients (29.3%) a CCI>2, 50 patients (20.7%) an ACCI>5, and 122 (50.4%) patients an ECOG-PS>1. The ASA-score (P = 0.001), CCI (P = 0.029), and the ECOG-PS (P<0.001) were significantly associated with age. In addition, the ECOG-PS was associated with pelvicalyceal tumors (P = 0.012), and the CCI with preoperative hydronephrosis (P = 0.026). The median follow-up was 30 months. In Kaplan-Meier analyses, ACCI>5 (P≤0.025) and ECOG-PS>1 (P≤0.042) were associated with recurrence-free survival, CSS, and overall survival, and ASA-score≥3 (P = 0.011) and ACCI>5 (P = 0.006) with CIM. In multivariable analysis that adjusted for standard clinicopathologic parameters, an ECOG-PS>1 was an independent predictor for CSS (hazard ratio = 1.89, P = 0.019), and an ASA-score≥3 (hazard ratio = 1.86, P = 0.026) was a predictor for CIM.

CONCLUSION: CPIs are easy assessable predictors for outcome in patients with UTUC who were treated with RNU. CPIs have carefully to be taken into account in patient counseling regarding operative decision making and multimodal treatment.