Effectiveness of warfarin among patients with cancer.

BACKGROUND: Among patients treated with warfarin for venous thromboembolism (VTE), cancer patients have more thrombotic and hemorrhagic events than patients without cancer. Is this also the case when cancer patients are anticoagulated for other indications? OBJECTIVE: The objective of the study is to evaluate the effectiveness of warfarin, given for any indication, among patients with cancer in a community setting. METHODS: We identified patients with cancer from a larger prospective cohort of 6,761 patients from 101 clinical sites in the United States, matched to controls without cancer. The proportion of time spent in the therapeutic range, international normalized ration (INR) variability, and the rate of thromboembolic and major hemorrhagic events were compared between the two groups. RESULTS: Ninety-five patients undergoing treatment for cancer were matched to 283 patients without cancer. The cancer group spent less time in the target INR range (54 vs 66%, P < .001) and had more variable INR values (standard deviation around the mean INR value 1.30 vs 0.71, P < .001). There were more thrombotic events in the cancer group than in the control group (5 vs 0 events, P < .001). These analyses were repeated after excluding all of the patients anticoagulated for VTE; the results were unchanged. CONCLUSIONS: Compared to matched controls, cancer patients receiving warfarin spend less time in the target INR range, have more variable INR values, and have more thrombotic events. These effects are not dependent on whether the patient is anticoagulated for VTE or another indication.