5-alpha Reductase inhibitors and risk of male breast cancer: a systematic review and meta-analysis

Abstract Objective: To assess the relationship between 5α-reductase inhibitors (5ARIs) and the risk of male breast cancer (MBC). Material and Methods: We systematically searched Medline via PubMed, Embase and the Cochrane Library Central Register up to May 2017 to identify published articles related to 5ARIs and the risk of MBC. Results: Summary effect estimates were calculated by a random-effect model, and tests for multivariable-unadjusted pooled risk ratios (RR) and heterogeneity, as well as the sensitivity analyses were conducted to assess publication bias. All four studies were conducted in a quality assessment according to the Newcastle Ottawa Scale system. The strength of association between 5ARIs and the prevalence of MBC was evaluated by using summarized unadjusted pooled RR with a 95% confidence interval [CI]. Four studies involving 595.776 participants, mean age range from 60 to 73.2 years old, were included in a meta-analysis, which produced a summary unadjusted RR of the risk of MBC for the treatment of 5ARIs of 1.16 (95% CI 0.85-1.58, P=0.36) and the multivariable-adjusted RR is 1.03, (95% CI 0.75-1.41, p=0.86). There was no heterogeneity among included studies (I2=0%, P=0.49). Estimates of total effects were generally consistent with the sensitivity. Conclusion: We did not observe a positive association between the use of 5ARIs and MBC. The small number of breast cancer cases exposed to 5ARIs and the lack of an association in our study suggest that the development of breast cancer should not influence the prescribing of 5ARIs therapy.

摘要 目的：评估5α-还原酶抑制剂（5α-reductase inhibitors，5ARIs）与男性乳腺癌（male breast cancer，MBC）发病风险的相关性。
材料与方法：本研究通过PubMed检索Medline、Embase及Cochrane图书馆中心注册库进行系统文献检索，检索时限截至2017年5月，旨在筛选与5ARIs及MBC发病风险相关的已发表研究。
结果：本研究采用随机效应模型（random-effect model）计算合并效应量，同时开展未经多变量校正的合并风险比（risk ratios，RR）检验、异质性检验及敏感性分析，以评估发表偏倚。依照纽卡斯尔渥太华量表（Newcastle Ottawa Scale）对纳入的4项研究进行质量评价。本研究以未校正合并RR及95%置信区间（95% confidence interval，CI）评估5ARIs与MBC患病率的关联强度。本次荟萃分析（meta-analysis）共纳入4项研究，涉及595776名受试者，平均年龄范围为60~73.2岁；结果显示，5ARIs治疗人群的MBC发病风险未校正合并RR为1.16（95%CI 0.85~1.58，P=0.36），经多变量校正后的合并RR为1.03（95%CI 0.75~1.41，P=0.86）。纳入研究间无明显异质性（I²=0%，P=0.49），总效应估计值与敏感性分析结果基本一致。
结论：本研究未观察到5ARIs的使用与MBC发病风险存在正相关。本研究中暴露于5ARIs的乳腺癌病例数较少，且未发现二者存在关联，提示乳腺癌的发生不应影响5ARIs治疗方案的临床开具。