Incidence of cancer after prophylaxis with warfarin against recurrent venous thromboembolism

Background: The length of time after an episode of venous thromboembolism d uring which the risk of newly diagnosed cancer is increased is not known, a nd whether vitamin K antagonists have an antineoplastic effect is controver sial. Methods: In a prospective, randomized study of the duration of oral anticoa gulation (six weeks or six months) after a first episode of venous thromboe mbolism, patients were questioned annually about any newly diagnosed cancer . After a mean follow-up of 8.1 years, we used the Swedish Cancer Registry to identify all diagnoses of cancer and causes of death in the study popula tion. The observed numbers of cases of cancer were compared with expected n umbers based on national incidence rates, and the standardized incidence ra tios were calculated. Results: A first cancer was diagnosed in 111 of 854 patients (13.0 percent) during follow-up. The standardized incidence ratio for newly diagnosed can cer was 3.4 (95 percent confidence interval, 2.2 to 4.6) during the first y ear after the thromboembolic event and remained between 1.3 and 2.2 for the following five years. Cancer was diagnosed in 66 of 419 patients (15.8 per cent) who were treated for six weeks with oral anticoagulants, as compared with 45 of 435 patients (10.3 percent) who were treated for six months (odd s ratio, 1.6; 95 percent confidence interval, 1.1 to 2.4). The difference w as mainly due to the occurrence of new urogenital cancers, of which there w ere 28 cases in the six-week group (6.7 percent) and 12 cases in the six-mo nth group (2.8 percent) (odds ratio, 2.5; 95 percent confidence interval, 1 .3 to 5.0). The difference in the incidence of cancer between the treatment groups became evident only after two years of follow-up, and it remained s ignificant after adjustment for sex, age, and whether the thromboembolism w as idiopathic or nonidiopathic. Older age at the time of the venous thrombo sis and an idiopathic thromboembolism were also independent risk factors fo r a diagnosis of cancer. No difference in the incidence of cancer-related d eaths was detected. Conclusions: The risk of newly diagnosed cancer after a first episode of ve nous thromboembolism is elevated during at least the following two years. S ubsequently, the risk seems to be lower among patients treated with oral an ticoagulants for six months than among those treated for six weeks. (N Engl J Med 2000;342:1953-8.) (C)2000, Massachusetts Medical Society.