Aspirin could cut the risk of dying of prostate cancer by almost 50% for men diagnosed with localized disease, a study has shown.
The investigators also found that other anticoagulants decreased the risk of prostate cancer–specific death in these men, but aspirin seemed to have the biggest positive effect, Dr. Kevin Choe said at a press briefing, held in advance of the annual meeting of the American Society for Radiation Oncology.
“Evidence has shown that anticoagulants may interfere with cancer growth and spread,” said Dr. Choe, a radiation oncologist at the University of Texas at Dallas. “If the major effect of anticoagulants is preventing metastasis, this may be why previous clinical trials with anticoagulation medications produced mixed results, since most patients in these trials already had metastasis. If the cancer had already metastasized, then anticoagulants may not be as beneficial.”
Dr. Choe and his colleagues based their retrospective study on data extracted from the Cancer of the Prostate Strategic Urologic Research Endeavor (CaPSURE). They used the database to investigate the effect of anticoagulation medications, including aspirin, warfarin, clopidogrel, and enoxaparin, on the risk of dying of prostate cancer that has not spread.
The study group comprised 5,295 men with localized adenocarcinoma of the prostate. Patients were treated with radical prostatectomy (3,523) or radiotherapy (1,772). Of the entire group, 1,932 men (37%) were using at least one anticoagulant at the time of their diagnosis or during follow-up (warfarin 428, clopidogrel 287, enoxaparin 26, aspirin 1,649). Almost a third of the men (28%) were taking androgen deprivation therapy.
The patients’ mean age was 65 years, and the mean prostate-specific antigen, 6 ng/mL. Based on the National Comprehensive Cancer Network risk stratification, 41% of the patients had low-risk disease, 37% intermediate risk, and 22% high risk.
Men who were not taking any anticoagulants were significantly younger than those taking the drugs (64 vs. 66 years), and more likely to have undergone radiotherapy (69% vs. 62%). The PSA was slightly, but not significantly, lower in the anticoagulant group, but the Gleason score and T stages were similar.
The median follow-up time was 59 months. At 7 and 10 years, prostate cancer–specific mortality was significantly lower in the anticoagulant group (1% vs. 4% and 4% vs. 10%, respectively; P less than .01).
The investigators also performed a subgroup analysis to determine the survival advantage associated with each risk category and with each type of anticoagulant. The survival benefit was greatest in patients with high-risk disease (2% vs. 8% mortality at 7 years and 4% vs. 22% at 10 years; P = less than .01). The benefit was seen whether the men had received radiation or hormone therapy.
A multivariate analysis found that the use of any anticoagulant was an independent predictor of prostate cancer–specific survival, reducing the risk of death by 47% (hazard ratio, 0.53; P less than .01). A medication analysis suggested that aspirin conferred most of the survival benefit.
Because the database didn’t contain much information on the duration of anticoagulant treatment, Dr. Choe couldn’t draw any conclusion about an association between length of treatment and risk of metastasis and death. Nor, he said, could he recommend that men without prostate cancer take aspirin as a prostate cancer preventative, or that men with newly diagnosed prostate cancer begin taking it.
“There are risks associated with anticoagulants, and in fact they can increase the risk of bleeding from the rectal wall during prostate cancer treatment,” he said. “Further studies are necessary before the addition of aspirin to prostate cancer therapy could become a standard treatment.”
However, he added, “for those who are already taking an anticoagulant for some other reason, we might expect to see an additional benefit in prostate cancer.”
The mechanistic relationship between metastatic disease and anticoagulants remains unclear, Dr. Choe said. However, in vitro and in vivo experiments suggest that normal platelet aggregation may protect circulating tumor cells. “Once the cells break away from the original site into the bloodstream, they are coated by platelets that prevent immune cells from attacking and seem to aid the tumor cells in sticking somewhere and beginning to grow,” Dr. Choe said. “Any interference with this clotting function might decrease the protective effect of platelets.”
Neither Dr. Choe nor his coinvestigators reported any relevant financial conflicts.