Protective effect of bisphosphonates on endometrial cancer incidence in data from the Prostate, Lung, Colorectal and Ovarian (PLCO) cancer screening trial.
Date: 2015 Feb 1
Major Program(s) or Research Group(s): PLCO
PubMed ID: 25533883
PMC ID: not available
Abstract: BACKGROUND: Preclinical studies have demonstrated antitumor effects of bisphosphonates. The objective of the current study was to determine the effect of exposure to bisphosphonate on the incidence of endometrial cancer. METHODS: The authors used data from the National Cancer Institute's Prostate, Lung, Colorectal and Ovarian (PLCO) Cancer Screening Trial, which collected data on all cancers. In year 5, all participants were asked to complete a self-administered supplemental questionnaire (SQX) that included questions regarding bone medication use. For women without a cancer diagnosis at the time of the SQX, the authors identified whether a woman reported current or former use of a nitrogenous bisphosphonate (NBP), defined as ever-use, and compared them with women never exposed to an NBP. Women with missing information were excluded as were women who reported undergoing a hysterectomy. Incidence rates and rate ratios were calculated with 95% confidence intervals (95% CIs). Cox proportional hazard ratios were also calculated and adjusted for covariates. RESULTS: A total of 29,254 women were included in the current analysis; an additional 77 cases of endometrial cancer have been diagnosed since the SQX. The incidence rate for endometrial cancer among women exposed to NBPs was 8.7 per 10,000 person-years versus 17.7 per 10,000 person-years among never-exposed women (rate ratio, 0.49; 95% CI, 0.30-0.80). The effect was similar after adjusting for all the covariates in the Cox proportional hazards analysis, with a hazard ratio of 0.56 (95% CI, 0.34-0.93). CONCLUSIONS: The results of the current study suggest that use of NBPs may have a protective effect on the incidence of endometrial cancer. However, additional studies are needed that include other potential confounders and a larger sample.