The authors retrospectively reviewed all patients with germ cell tumors treated with platinum-based chemotherapy and followed at the university hospital and cancer center of Bordeaux from June 2007 - January 2015. Several variables were assessed including: age, performance status, body-surface area, histological type, site of primary tumor, stage, prognosis (IGCCCG classification), size of retroperitoneal lymphadenopathy, LDH, type of venous access (peripheral vein, implantable venous access port or PICC-Line), use of thromboembolic prophylaxis, thromboembolic events, and median time to onset. Univariable logistic regression analysis was used to assess factors associated with thromboembolic events.
There were 206 patients meeting inclusion criteria for this study. Thirty-six (17.4%) had one or more venous or arterial thromboembolic events during or within 3 months of the end of chemotherapy, including 26 deep vein thromboses, 13 pulmonary embolisms and 4 arterial accidents (2 acute limb ischemia and 2 acute coronary syndromes). The median time between treatment initiation and occurrence of thromboembolic events was 38.5 days [range 7-248]. Overall, the use of chemoprophylaxis was used in only 16% of patients. Among 12 patients receiving chemotherapy via a PICC-line, six had a thromboembolic event, including four cases of pulmonary embolism. In addition to the presence of a PICC-line (OR 4.62, 95%CI 1.28-15.56), factors associated with an increased thromboembolic event risk in univariate analysis were the use of implantable venous access port (OR 3.97, 95%CI 1.81-9.12), size of retroperitoneal adenopathy ≥ 2 cm ([2-5 cm]: OR 24.39, 95%CI 4.50-454.39; > 5cm: OR 26.25 95%CI 5.0-484.5), stage II disease (OR 10.19 95%CI 1.9- 188.96), stage III disease (OR 27.2 95%CI 5.34-479.9), prognosis group (unfavorable: OR 33.35 95%CI 5.7–637.34), VIP protocol (OR 6.35 95%CI 1.51–28.15) and high LDH level (OR 3.88 95%CI 1.63–10.34). Given the low event rate cited by the authors, a limitation of this study was the lack of multivariable model to assess independent predictors of thromboembolic events.
The authors summarized that the overall thromboembolic event rate was 17.5% in this cohort of germ cell tumor patients receiving chemotherapy. These results suggest a significant association between PICC-line and the occurrence of thromboembolic events in univariate analysis. A PICC-line is not risk-free and should be used judiciously in patients with germ cell tumors whose treatment objective remains a cure at the cost of less toxicity. It seems reasonable to consider evaluating VTE chemoprophylaxis for patients in this setting.
Presented by: Louis Francois, University Hospital Bordeaux, Bordeaux, France
Co-Authors: Marine Gross-Goupil, Gregoire Robert, Alain Ravaud; University Hospital Bordeaux, Bordeaux, France; Centre Hospitalier-Universitaire Saint Andre, Bordeaux, France; Bordeaux University Hospital, Bordeaux, France
Written by: Zachary Klaassen, MD, Urologic Oncology Fellow, University of Toronto, Princess Margaret Cancer Centre, @zklaassen_md at the 2018 American Society of Clinical Oncology Genitourinary (ASCO GU) Cancers Symposium, February 8-10, 2018 - San Francisco, CA