Background: Renal Cell Carcinomas (RCCs) are a heterogeneous group of malignancies. Although patients presenting with locally advanced tumours may be cured by surgery alone, many subsequently relapse and succumb to their disease. Studies of adjuvant tyrosine-kinase-inhibitors (TKIs) did not meet their primary goals but provide a rich data source to refine prognostication in contemporary patients and inform the design of clinical trials testing new therapeutic approaches.
Methods: This study included patients treated surgically with curative intention from two international clinical trials testing the addition of TKIs in intermediate and high-risk patients; SORCE (n=1711) and ASSURE (n=1943). Three questions were addressed; A validation of the 2003 Leibovich prognostic score (a widely used scoring system) in contemporary patients with clear-cell and non-clear-cell RCCs, also comparing this with the Tumour/Nodes/Metastases (TNM) classification. Discrimination and calibration were assessed by comparing data from SORCE to original data used to derive the Leibovich score using Harrell’s concordance-indexes, Kaplan-Meier curves and hazard ratios (HRs). Secondly data from SORCE and ASSURE were combined, generating a large dataset to examine clinical characteristics of higher risk non-clear-cell RCC variants (papillary RCC (pRCC), chromophobe RCC (chRCC) and sarcomatoid RCC (sRCC)). The impact of histology on disease-free-survival and overall survival were presented using Kaplan-Meier curves and adjusted Cox regression models. Finally, a retrospective cohort study examining data from SORCE compared outcomes of those relapsing first at single anatomical sites to those relapsing at multiple sites using Kaplan-Meier methodology. The prognostic impact of organ site and time-to-relapse (TTR), performance status and treatments upon relapse were evaluated using Cox regression models.
Results: The 2003 Leibovich score demonstrated discriminative accuracy in patients with clear-cell (c-index 0.63, 95% CI 0.61 to 0.65) and non–clear-cell RCCs (c-index 0.64, 95% CI 0.59 to 0.69). Discrimination by the 2003 Leibovich score exceeded that of 2002 TNM (c-indexes of 0.67 (SE 0.01) vs 0.56 (SE 0.01)). Distinct patterns of relapse for patients with chRCC, pRCC and sRCCs were shown. Notably, the median TTR for patients with pRCC was five months less than patients with ccRCC; (1.34 years (IQR 0.76, 2.59) vs 1.78 years (IQR 0.96, 3.38, p=0.012)). Those with pRCC relapsing in the abdomen had almost double the risk of death (HR 1.7 (95% CI 1.15-2.5 p