Introduction: Cancer associated thrombosis (CAT) is a common complication of cancer, associated with significant morbidity and mortality. While incidence varies with cancer type, stage, chemotherapy, and other factors, estimates are that up to 20% of cancer patients will experience at least one venous thromboembolic (VTE) episode. VTE consist of deep vein thrombosis (DVT) and/or pulmonary embolism (PE), and there is a spectrum of vascular involvement. DVTs are classified as proximal (popliteal vein or above) and PE may be subsegmental, segmental, lobar, main, or saddle embolism. Our standard approach has been to treat all DVTs and all PEs in cancer patients, regardless of the size of the involved vessel. However, it is not clear if the risk of recurrent VTE in a patient with a "small" subsegmental PE, or a calf vein DVT, is comparable to that of an individual with a larger vascular involvement. In this study, we characterized the largest vessel involved in the initial VTE episode, and the relationship with recurrent VTE.

Methods: All patients at MSKCC with CAT are monitored within an existing Quality Assessment initiative. From 1/1/2014 through 10/31/2016, 1072 patients with CAT were treated with rivaroxaban (Riva). (The overall outcomes of this cohort are the subject of a separate abstract.) In this study we compared the rate of recurrent VTE in patients with a distal (calf) DVT with proximal DVT, and PE. We designated the most proximal, or largest thrombosed vessel. As patients with a PE do not routinely undergo Doppler leg ultrasound, we are unable to differentiate PE with a DVT from those without. We also analyzed if the PE was unilateral or bilateral. We used competing risk endpoints for the purpose of this analysis, including recurrent VTE, major bleeding, clinically relevant non-major bleeding leading to discontinuation of Riva, and death.

Results: In the Table, we present the data on the relationship of the initial VTE and the risk of recurrence. The majority of CAT events (55%) were PE. There were no significant differences in the rates of recurrent VTE between patients with a PE, a distal DVT or proximal DVT. Within patients with a PE as the index VTE event, there was no significant association between the risk of recurrent VTE and the size of the PE. A subsegmental PE as an index event was associated with a comparable rate of recurrent VTE when compared with segmental and more proximal vessel involvement. The only meaningful trend towards a higher rate of recurrent VTE was in patients whose index event was a bilateral PE, compared with unilateral, although this association did not reach statistical significance.

Conclusions: The goal of this Quality Assessment initiative was to evaluate the risk of recurrent VTE in cancer patients to determine if distal DVT's or subsegmental PEs had a significantly lower rate of recurrence than other VTE episodes. Our analysis indicated that the risk of recurrent VTE is not related to the size of the index thrombosed vessel. Within PE, from large, proximal index events through to subsegmental PE, the risk of recurrent VTE is comparable. Similarly, there was only a trend towards lower risk of recurrent VTE in patients with an index distal DVT, versus proximal DVT. But this association was not statistically significant, with overlapping 95% confidence intervals. The one parameter that appeared to have the strongest prediction of recurrent VTE was patients with bilateral PE, versus unilateral. However, this too was only a trend, not statistically significant, and was not one of the parameters within our initial hypotheses.

We were unable to identify any subgroup of index VTE, based on vessels involved, that had a significantly lower rate of recurrent VTE while on anticoagulation. Within cancer patients, a subsegmental PE or a distal, calf vein DVT are associated with a risk of recurrent VTE comparable to thrombosis of larger vessels.


Soff:Janssen: Research Funding; Amgen: Research Funding. Mantha:Janssen: Research Funding; GLG: Consultancy; Heidell, Pittoni, Murphy & Bach, LLP: Consultancy.

Author notes


Asterisk with author names denotes non-ASH members.