复方与仅孕激素激素避孕药与乳腺癌风险:英国巢式病例对照研究及荟萃分析
该研究基于英国初级保健数据库,对9498例50岁以下乳腺癌患者与18171例匹配对照进行巢式病例对照分析,并结合12项观察性研究进行荟萃分析。结果显示,当前或近期使用复方口服、仅孕激素口服、注射或宫内节育器避孕均与乳腺癌风险轻微升高相关,且不同给药方式间风险相似;口服避孕药使用5年的15年绝对超额风险在高收入国家随年龄增长而升高。
要点
- 当前或近期使用仅孕激素避孕药与乳腺癌风险轻微升高相关,不同给药方式间风险相似。
- 复方口服避孕药与仅孕激素避孕药的乳腺癌风险升高幅度相近。
- 口服避孕药使用5年的15年绝对超额风险在较年轻女性中较小,在较年长女性中较大。
核心结论
- 当前或近期使用仅孕激素避孕药与乳腺癌风险轻微升高相关,且风险似乎不随给药方式变化,与复方激素避孕药的风险幅度相似。
- 鉴于乳腺癌基础风险随年龄增长而增加,口服避孕药相关的绝对超额风险估计在较年轻使用者中比较年长使用者中更小。
- 此类风险需与育龄期使用避孕药的获益相权衡。
研究设计与样本
- 英国初级保健数据库(CPRD)巢式病例对照研究,纳入9498例1996至2017年诊断的50岁以下新发浸润性乳腺癌女性及18171例密切匹配对照。
- 每个病例及其匹配对照在诊断日期前平均有7.3年(标准差4.6)的临床记录。
- 采用条件logistic回归估计乳腺癌比值比(OR)及95%置信区间(CI),控制年龄、全科诊所、体重指数、记录生育次数、末次生育时间及酒精摄入。
- 检索MEDLINE和Embase中1995年1月1日至2022年11月1日发表的观察性研究,采用固定效应荟萃分析将CPRD结果与12项已发表观察性研究结果合并。
主要发现
- 总体上,44%(4195/9498)乳腺癌女性与39%(7092/18171)匹配对照在乳腺癌诊断(或对照相应日期)前平均3.1年(标准差3.7)有过激素避孕药处方。
- 约半数处方为仅孕激素制剂。
- 末次激素避孕药处方为口服复方、口服仅孕激素、注射孕激素或孕激素释放宫内节育器(IUD)时,乳腺癌OR均相似且显著升高:OR分别为1.23(95%CI 1.14至1.32;p<0.001)、1.26(95%CI 1.16至1.37;p<0.001)、1.25(95%CI 1.07至1.45;p=0.004)和1.32(95%CI 1.17至1.49;p<0.001)。
- 荟萃分析显示当前或近期使用仅孕激素避孕药的相对风险(RR)显著升高:口服=1.29(95%CI 1.21至1.37;异质性χ25=6.7;p=0.2),注射=1.18(95%CI 1.07至1.30;异质性χ28=22.5;p=0.004),植入=1.28(95%CI 1.08至1.51;异质性χ23=7.3;p=0.06),IUD=1.21(95%CI 1.14至1.28;异质性χ24=7.9;p=0.1)。
- 将CPRD结果与先前发表结果(包括更宽年龄范围女性)合并后,高收入国家口服复方或仅孕激素避孕药使用5年相关的15年绝对超额风险估计为:16至20岁每10万使用者8例,35至39岁每10万使用者265例。
临床怎么用
- 原文未据此给出推荐。
- 原文结论指出,当前或近期使用仅孕激素避孕药与乳腺癌风险轻微升高相关,风险似乎不随给药方式变化,与复方激素避孕药的风险幅度相似。
- 原文结论指出,鉴于乳腺癌基础风险随年龄增长而增加,口服避孕药相关的绝对超额风险估计在较年轻使用者中比较年长使用者中更小,此类风险需与育龄期使用避孕药的获益相权衡。
局限性
- 研究设计的主要局限性在于,由于CPRD数据及多数其他处方数据库的性质,避孕药使用信息仅在限定时期内记录,进入数据库前的信息通常不可获得。
- 这意味着尽管研究发现提供了激素避孕药与乳腺癌风险短期关联的证据,但未提供关于长期关联或避孕药总使用时长对乳腺癌风险影响的信息。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Current or recent use of combined oral contraceptives (containing oestrogen+progestagen) has been associated with a small increase in breast cancer risk. Progestagen-only contraceptive use is increasing, but information on associated risks is limited. We aimed to assess breast cancer risk associated with current or recent use of different types of hormonal contraceptives in premenopausal women, with particular emphasis on progestagen-only preparations. METHODS AND FINDINGS: Hormonal contraceptive prescriptions recorded prospectively in a UK primary care database (Clinical Practice Research Datalink [CPRD]) were compared in a nested case-control study for 9,498 women aged <50 years with incident invasive breast cancer diagnosed in 1996 to 2017, and for 18,171 closely matched controls. On average, 7.3 (standard deviation [SD] 4.6) years of clinical records were available for each case and their matched controls prior to the date of diagnosis. Conditional logistic regression yielded odds ratios (ORs) and 95% confidence intervals (CIs) of breast cancer by the hormonal contraceptive type last prescribed, controlled for age, GP practice, body mass index, number of recorded births, time since last birth, and alcohol intake. MEDLINE and Embase were searched for observational studies published between 01 January 1995 and 01 November 2022 that reported on the association between current or recent progestagen-only contraceptive use and breast cancer risk in premenopausal women. Fixed effects meta-analyses combined the CPRD results with previously published results from 12 observational studies for progestagen-only preparations. Overall, 44% (4,195/9,498) of women with breast cancer and 39% (7,092/18,171) of matched controls had a hormonal contraceptive prescription an average of 3.1 (SD 3.7) years before breast cancer diagnosis (or equivalent date for controls). About half the prescriptions were for progestagen-only preparations. Breast cancer ORs were similarly and significantly raised if the last hormonal contraceptive prescription was for oral combined, oral progestagen-only, injected progestagen, or progestagen-releasing intrauterine devices (IUDs): ORs = 1.23 (95% CI [1.14 to 1.32]; p < 0.001), 1.26 (95% CI [1.16 to 1.37]; p < 0.001), 1.25 (95% CI [1.07 to 1.45]; p = 0.004), and 1.32 (95% CI [1.17 to 1.49]; p < 0.001), respectively. Our meta-analyses yielded significantly raised relative risks (RRs) for current or recent use of progestagen-only contraceptives: oral = 1.29 (95% CI [1.21 to 1.37]; heterogeneity χ25 = 6.7; p = 0.2), injected = 1.18 (95% CI [1.07 to 1.30]; heterogeneity χ28 = 22.5; p = 0.004), implanted = 1.28 (95% CI [1.08 to 1.51]; heterogeneity χ23 = 7.3; p = 0.06), and IUDs = 1.21 (95% CI [1.14 to 1.28]; heterogeneity χ24 = 7.9; p = 0.1). When the CPRD results were combined with those from previous published findings (which included women from a wider age range), the resulting 15-year absolute excess risk associated with 5 years use of oral combined or progestagen-only contraceptives in high-income countries was estimated at: 8 per 100,000 users from age 16 to 20 years and 265 per 100,000 users from age 35 to 39 years. The main limitation of the study design was that, due to the nature of the CPRD data and most other prescription databases, information on contraceptive use was recorded during a defined period only, with information before entry into the database generally being unavailable. This means that although our findings provide evidence about the short-term associations between hormonal contraceptives and breast cancer risk, they do not provide information regarding longer-term associations, or the impact of total duration of contraceptive use on breast cancer risk. CONCLUSIONS: This study provides important new evidence that current or recent use of progestagen-only contraceptives is associated with a slight increase in breast cancer risk, which does not appear to vary by mode of delivery, and is similar in magnitude to that associated with combined hormonal contraceptives. Given that the underlying risk of breast cancer increases with advancing age, the absolute excess risk associated with use of either type of oral contraceptive is estimated to be smaller in women who use it at younger rather than at older ages. Such risks need be balanced against the benefits of using contraceptives during the childbearing years.
引用
- Fitzpatrick D, Pirie K, Reeves G, et al. Combined and progestagen-only hormonal contraceptives and breast cancer risk: A UK nested case-control study and meta-analysis. PLoS Med. 2023. PMID: 36943819. doi:10.1371/journal.pmed.1004188
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。