Aspirin use in relation to risk of prostate cancer.

  • 1 October 2002
    • journal article
    • Vol. 11, 1108-11
Abstract
Experimental studies have shown inhibitory effects of nonsteroidal anti-inflammatory drugs on prostate cancer cell proliferation and reduction of prostate cancer metastasis, suggesting their possible preventive role for prostate cancer. We examined the association between regular aspirin use and the risk of prostate cancer among participants in the Health Professionals Follow-up Study, a prospective cohort of 47,882 United States men who were 40-75 years of age and without a history of prostate cancer in 1986. Biennial self-administered questionnaires were used to assess regular aspirin use from 1986 to 1996. We confirmed and staged incident cases of prostate cancer according to medical records and pathology reports. During 518,072 person-years of follow-up, 2,479 new cases of prostate cancer were ascertained. Of these, 608 were diagnosed as advanced (extraprostatic) prostate cancer and 258 as metastatic prostate cancer. We found no association between aspirin use and total prostate cancer. After accounting for prostate-specific antigen examinations and other potentially confounding variables, the relative risk of total prostate cancer for aspirin users compared with nonusers was 1.05 (95% confidence interval, 0.96-1.14). For metastatic prostate cancer, we observed a suggestive decrease in risk among men reporting greater frequency of aspirin use. The multivariate relative risk of metastatic prostate cancer among men using aspirin 22 or more days/month was 0.73 (95% confidence interval, 0.39-1.38) compared with nonusers. We noted no evidence of a linear dose-response relationship (P for trend = 0.40). The results from this cohort indicate that regular aspirin use is not likely to prevent the incidence of total prostate cancer, but we cannot exclude a possible benefit of frequent aspirin use on risk of developing metastatic prostate cancer.

This publication has 0 references indexed in Scilit: