Venous thromboembolism risk in patients with locoregional urothelial tract tumors

Ramos, Jorge D.; Wingate, Jonathan  T.; Gulati, Roman; Plimack, Elizabeth R.; Harshman, Lauren C.; Powles, Thomas; Crabb, Simon J.; Niegisch, Günter; Bellmunt Molins, Joaquim, 1959-; Ladoire, Sylvain; De Giorgi, Ugo; Hussain, Syed; Alva, Ajjai S.; Baniel, Jack; Agarwal, Neeraj; Rosenberg, Jonathan E.; Vaishampayan, Ulka N.; Galsky, Matthew D.; Yu, Evan Y.

doi:http://dx.doi.org/10.1016/j.clgc.2017.08.001

Venous thromboembolism risk in patients with locoregional urothelial tract tumors

Citation

Ramos JD, Wingate JT, Gulati R, Plimack ER, Harshman LC, Powles T. et al. Venous thromboembolism risk in patients with locoregional urothelial tract tumors. Clin Genitourin Cancer. 2018 Feb; 16(1):e161-7. DOI: 10.1016/j.clgc.2017.08.001

Abstract

BACKGROUND: Venous thromboembolism (VTE) is common in cancer patients, but there is limited data on patients with urothelial tract tumors (UTT). We previously identified several associative factors for increased VTE rates in patients with metastatic UTT. In this study, we assessed the frequency, associative factors, and impact on survival of VTE in patients with locoregional UTT. METHODS: Patients with locoregional bladder, upper urinary tract, or urethral cancer were included in this multi-center study from 29 academic institutions. Patients with < cT2, > N1, or M1 disease at diagnosis were excluded. Patients with incomplete clinical staging or miscoded/missing data were excluded. Cumulative, unadjusted VTE incidence was calculated from time of diagnosis of muscle-invasive disease, excluding VTEs diagnosed in the metastatic setting. χ2 statistics tested differences in VTE rates across baseline and treatment-related factors. Significant covariates were incorporated into a multivariate, logistic regression model. Overall survival stratified by VTE was estimated using Kaplan-Meier methods and evaluated using the log-rank test. RESULTS: A total of 1732 patients were eligible. There were 132 (7.6%) VTEs. On multivariate analysis, non-urothelial histology (P < .001), clinical Nx stage (P < .001), cardiovascular disease (P = .01), and renal dysfunction (P = .04) were statistically significant baseline factors associated with VTE. Using surgery alone as reference, surgery with perioperative chemotherapy (P = .04) and radiation with concurrent chemotherapy (P = .04) also were significant. CONCLUSIONS: The VTE incidence of 7.6% in locoregional disease is comparable with our previously reported rate in the metastatic setting (8.2%). Similar to our findings in metastatic UTT, non-urothelial histology, renal dysfunction, and CVD was associated with increased VTE risk.