Warfarin for Cancer Prevention

July 17, 2000. Medical researchers have suspected for decades that blood clots may be a prelude to cancer. Not long ago researchers at Dartmouth Medical School found that after a blood clot a person's increased risk of cancer was "very large" and stayed 30% higher than normal.(1) Either some change that leads to cancer "promotes thrombosis," they suggest, "or cancer and thrombosis share common risk factors."

Men with prostate cancer quite often develop clots in the leg or the lung. Clots may arise either in course of the prostate cancer or as a side effect of drugs for treating the disease -- such as estramustine (Emcyt) and PC-SPES -- or coincidentally, on account the patient's age and general health.

Patients who develop clots may be put on the anticoagulant pill warfarin (brand name Coumadin), if necessary after treatment with intravenous heparin. Patients taking a cancer drug known to increase blood clotting may be put on protective, prophylactic doses of warfarin.

Now evidence suggests that oral blood thinners, anticoagulants, or vitamin K antagonists may protect against cancer. (It is not yet known whether these drugs might have such as an effect on prostate cancer once it has developed.) In a study published in the June 29 issue of The New England Journal of Medicine, Sam Schulman, M.D., and colleagues at the Karolinska Hospital in Sweden found (as expected) that risk of cancer in patients treated for a blood clot was higher than for the population at large -- but for patients who took a six-month course of warfarin or dicumarol, it was strikingly lower compared to patients who took only a six-week course.

Of 854 patients who had a blood clot in the leg or lung, 419 received a blood thinner for six weeks and 435 took it for six months. The patients were followed for up to sixteen years (mean of 8.1 years). The rate at which they developed cancer was compared with expected numbers based on national incidence rates.

Of the 854 people on the study (all of whom had had clots), 111 (13 percent) developed cancer for the first time during follow-up. "The risk of newly diagnosed cancer after a first episode of venous thromboembolism is elevated during at least the following two years." the authors state. "Subsequently, the risk seems to be lower among patients treated with oral anticoagulants for six months than among those treated for six weeks."

Patients who took blood thinners for six months were one-and-a-half times less likely to develop cancer than those who took it for only six weeks. Only 45 patients (10.3 percent) developed cancer in the six-month treatment group compared with sixty-six patients (15.8 percent) in the six-week treatment group. "The difference was mainly due to the occurrence of new urogenital cancers," Dr. Schulman says. There were only 12 cases in the six-month group (2.8 percent) compared with 28 cases in the six-week group (6.7 percent).

After six years, those who had six months of warfarin showed only about an 8 percent risk of getting a cancer while those in the six-week-treatment group had a 15 percent risk.

"Our findings strongly support the impression that warfarin has an [anti-cancer] effect," the authors say. They add that this "will remain controversial" until a biochemical reason for this effect can be found.

In an accompanying editorial, Christoph C. Zielinski, M.D. and Michael Hejna, M.D. of University Hospital, Vienna, Austria ask:

"Should the information presented by Schulman and Lindmarker influence the practice of clinical oncology? It should, by raising awareness of the risk of certain cancers in patients with idiopathic thromboembolism. The necessity of appropriate follow-up in these patients is clear. But should patients with cancer or even healthy persons with a high risk of cancer receive anticoagulants, on the basis of these data? For now, the answer must be no."

Anticoagulants lower the risk from clotting but raise the risk of bleeding. Schulman has already stated that six months is best for the average patient: "The optimal duration of anticoagulation ... has been extensively investigated. For the majority of patients a treatment duration of 6 months eliminates the high risk of relatively early recurrences without yielding an increase in the incidence of major haemorrhages."

"Even so," the New England Journal of Medicine editorial writers say, "we must take the current results seriously enough to encourage the study of warfarin in controlled clinical trials, perhaps primarily with the aim of assessing prevention rather than treatment of cancer." "It may very well be," they say, "that we are holding in our hands a drug that could be added to the armamentarium of chemoprotective agents."