Effects of the Timing and the Type of Venous Thromboembolic Events On Survival in Patients with Pancreatic Cancer

Abstract 1151 Background: Patients (pts) with pancreatic cancer are at high risk for venous thromboembolic events (VTE) and the occurrence of VTE can adversely affect prognosis. However, it is unclear if the type of VTE such as symptomatic vs incidental, deep vein thrombosis (DVT) vs pulmonary embolism (PE), the location of VTE [DVT of extremities vs visceral veins (abdominal/pelvic veins)] or the timing of VTE from diagnosis can influence the survival. The purpose of this study was to evaluate the incidence of different types of VTE, the impact of types and timing of VTE (early vs late) on survival. Methods: Medical records of 260 pts with pancreatic cancer, newly referred to MDACC in 2006, were reviewed for cancer diagnosis, patient demographics (age, gender), presence of metastasis, the date of diagnosis of VTE, timing of VTE, type of VTE, the site of VTE, the incidence of VTE during 2 years of follow up from the date of diagnosis. Clinical and laboratory parameters predictive for survival were also reviewed. All VTE episodes, including symptomatic as well as incidental VTEs were confirmed by the radiological studies using CT ANGIO, CT scan, Doppler compression ultrasound or V/Q perfusion scans. The survival time was calculated from the date of cancer diagnosis to the date of last follow up. Survival analysis was conducted using Kaplan-Meier method and Cox proportional hazard models. The stepwise selection method was employed to build a multivariate model using variables with p<0.15 in univariate analysis. Results: Of the 260 pts referred, 235 were confirmed to have the diagnosis of pancreatic carcinoma. During the 2-year follow-up, 80 pts (34%) had 109 episodes of VTE, including symptomatic and incidental episodes. The median age of the pts with VTE was 59 years (range: 28–86) and 51% were males. Of the 80 pts with VTE, 21 (26%) had PE, 18 (23%) had DVT of extremities, 28 (35%) had DVT of visceral veins and 13 (16%) had concurrent PE/DVT (diagnosed on the same day). Of the 80 pts, 25 (31%) had 29 recurrent episodes. Kaplan-Meier survival analysis, as shown in the table below, indicated that the pts who had early VTE (defined as VTE diagnosed within 30 days from the date of diagnosis of pancreatic cancer) vs late VTE (> 30 days) and pts with metastasis vs no metastasis had statistically poor 1 year survival (log-rank test). Conclusions: These findings suggest that timing of VTE is an important indicator of prognosis, regardless of whether symptomatic or incidental. Patients with VTE within 30 days of diagnosis have shorter survival. Disclosures: No relevant conflicts of interest to declare.