A scoring algorithm to predict survival for patients with metastatic clear cell renal cell carcinoma: a stratification tool for prospective clinical trials

PURPOSE: We developed a clinically useful scoring algorithm to predict cancer specific survival for patients with clear cell metastatic renal cell carcinoma (RCC). MATERIALS AND METHODS: We studied 727 patients treated with radical nephrectomy for clear cell RCC from 1970 to 2000 who had distant metastases at nephrectomy (285) or in whom metastases subsequently developed (442). A scoring algorithm to predict cancer specific survival was developed using the regression coefficients from a Cox proportional hazards model. RESULTS: There were 606 deaths from clear cell RCC at a median of 1.0 years (range 0 to 14) following metastatic RCC. Constitutional symptoms at nephrectomy (+2), metastases to the bone (+2) or liver (+4), metastases in multiple simultaneous sites (+2), metastases at nephrectomy (+1) or within 2 years of nephrectomy (+3), complete resection of all metastatic sites (-5), tumor thrombus level I to IV (+3), and the primary pathological features of nuclear grade 4 (+3) and histological tumor necrosis (+2) were significantly associated with death from RCC. All patients started with a score of 0 and points were added or subtracted as indicated in parentheses. Cancer specific survival rates at 1 year were 85.1%, 72.1%, 58.8%, 39.0%, and 25.1%, respectively, for patients with scores of -5 to -1, scores of 0 to 2, scores of 3 to 6, scores of 7 or 8, and scores of 9 or more. CONCLUSIONS: This scoring algorithm can be used to predict cancer specific survival for patients with metastatic clear cell RCC.