Predictive and Prognostic Value of Preoperative Thrombocytosis in Upper Tract Urothelial Carcinoma

Micro-Abstract We evaluated the predictive and prognostic role of preoperative thrombocytosis (TC) in upper tract urothelial carcinoma. Records of 2492 patients undergoing radical nephroureterectomy between 1990 and 2008 were retrospectively analyzed. Preoperative TC predicts non–organ-confined disease (P <.001) and lymph node metastases (P <.001) at radical nephroureterectomy. Among other biomarkers, TC could benefit preoperative risk stratification and help guide treatment decisions. Purpose The purpose of this study was to evaluate the predictive and prognostic role of preoperative thrombocytosis (TC) in upper tract urothelial carcinoma (UTUC) after radical nephroureterectomy (RNU) in a large multi-institutional cohort of patients. Methods Records of 2492 patients undergoing RNU for non-metastatic UTUC between 1990 and 2008 were retrospectively analyzed. Preoperative TC was defined as a platelet count > 400 × 109/L, irrespective of gender type. Logistic regression analyses were performed to evaluate its association with pathologic features. Cox proportional hazards regression was used for estimation of recurrence-free survival, cancer-specific survival, and overall survival. Results Preoperative TC was found in 309 (12.4%) patients and was associated with advanced tumor stage and grade, lymph node metastasis, lymphovascular invasion, tumor architecture, necrosis, and concomitant carcinoma in situ (P-values ⠤.027). Preoperative TC independently predicted ⠥ pT2 (P <.001), non–organ-confined (P <.001), and lymph node-positive (P <.001) disease in a preoperative model that adjusted for the effects of age, gender, location, multifocality, and tumor architecture. Within a median follow-up of 45 months, recurrence occurred in 663 (26.6%) patients with 545 (21.9%) dying of UTUC. In univariable Cox proportional hazard regression analysis, TC was significantly associated with recurrence-free survival (hazard ratio [HR], 1.32; P =.015) and overall survival (HR, 1.4; P <.001), but not cancer-specific survival (HR, 1.17; P =.2). In both pre- and postoperative multivariable models, when adjusted for the effects of standard clinicopathologic features, TC did not retain its association with survival outcomes. Conclusions Preoperative TC is associated with adverse clinicopathologic features and predicts worse pathology at RNU. Among other serum biomarkers, TC could benefit preoperative risk stratification and help guide treatment decisions.