LAUSR

Background Several observational studies have suggested a protective effect of oral bisphosphonates (BP) on the risk of breast cancer, but no such association has been seen in randomized control trials. The role of oral BP in breast cancer prevention remains unclear. Aim To investigate the association between different levels of BP exposure and breast cancer incidence in a cohort of osteoporotic post-menopausal women. Subjects and methods This historical prospective study was conducted using the computerized databases of Maccabi Healthcare Services (MHS) in Israel. Included in the study were osteopenic and osteoporotic women aged 55–75 years who started BP therapy between 1998 and 2012. The subjects were enrolled in MHS for at least 3 years before therapy initiation, and had a minimum follow-up of 5 years in MHS. Women with a previous cancer, and women treated with selective estrogen receptor modulators (SERMs) were excluded. BP exposure was expressed in quintiles of proportion of days covered (PDC) with BP during follow-up period and cancer incidence was ascertained by the Israel National Tumor Registry. Person-years of follow-up began on January 1st, 1998 and ended at the date of cancer diagnosis, death, or December 31st, 2012, whichever occurred first. Results A total of 11,717 patients (mean age = 66.87 ± 4.38) were eligible for the analysis. During a total of 130,252 person-years of follow-up, (mean 7.2 years) 173 incident cases of breast cancer were diagnosed. Compared to women with a PDC with BP of 20% or lower, the adjusted hazard ratio for breast cancer were HR = 0.81 (95%CI: 0.48–1.39), HR = 0.82 (95%CI: 0.50–1.33), HR = 0.72 (95%CI:0.45–1.15) and HR = 1.14 (95%CI:0.76–1.70) among women with 20–40%, 40–60%, 60%–80%, and 80% or higher, PDC, respectively. Conclusion In this study, we did not find a significant association between oral BP therapy for osteoporosis and the risk of breast cancer in postmenopausal women. The discrepancy between our results and the reports of such an association in observational studies might originate from an indication bias.