【原】绝经前后女性可预防的乳腺癌风险

　　通过改变现有的不良生活方式，可以预防将来的乳腺癌发生风险。不过，不同生活方式对于绝经前和绝经后女性乳腺癌发生风险的影响缺乏定量分析。

　　2019年2月25日，国际抗癌联盟《国际癌症杂志》在线发表澳大利亚新南威尔士大学、新南威尔士癌症协会、悉尼大学、维多利亚癌症协会、墨尔本大学、澳大利亚国立大学、纽卡斯尔大学、贝克心脏和糖尿病研究所、阿德莱德大学、澳大利亚乳腺癌网络的研究报告，定量分析了个体和人群生活方式改变对于绝经前和绝经后女性乳腺癌发生风险的影响，以及不同人群亚组之间的影响是否不同。

　　作者将六项澳大利亚队列研究21万4536例汇总数据与全国癌症和死亡登记数据库进行关联，并通过校正后的比例风险模型，推算癌症发生及其所致死亡与不良生活方式之间的相关程度。通过抽样健康调查，推算不良生活方式发生率。将这些推算数据结合起来，计算人群归因分数和95%置信区间，并比较人群亚组的人群归因分数。

　　结果，前10年随访期间，发生绝经前女性乳腺癌640例、绝经后女性乳腺癌2632例、任何原因所致8761例。

　　对于绝经前女性，将来发生乳腺癌的风险因素所占比重：

• 经常饮酒无论摄入量：12.6%（95%置信区间： 4.3～20.2%）

• 口服避孕药至少五年：  7.1%（95%置信区间： 0.3～13.5%）

• 上述两个因素都存在：18.8%（95%置信区间： 9.1～27.4%）

　　对于绝经后女性，将来发生乳腺癌的风险因素所占比重：

• 体重指数超过标准值：12.8%（95%置信区间：  7.8～17.5%）

• 目前绝经后激素治疗：  6.9%（95%置信区间：  4.8～ 8.9%）

• 经常饮酒无论摄入量：  6.6%（95%置信区间：  1.5～11.4%）

• 上述三个因素都存在：24.2%（95%置信区间：17.6～30.3%）

　　根据亚组分析，将来发生绝经后乳腺癌的风险因素：

• 绝经后激素治疗女性：体脂、酒精摄入量、社会经济状况

• 身体脂肪量过多女性：酒精摄入量、教育程度

• 酒精摄入量过多女性：母乳喂养史

　　因此，该研究结果为有针对性的人群水平癌症控制行动提供了证据。

Int J Cancer. 2019 Feb 25. [Epub ahead of print]

The preventable burden of breast cancers for premenopausal and postmenopausal women in Australia: a pooled cohort study.

Arriaga ME, Vajdic CM, Canfell K, MacInnis RJ, Banks E, Byles JE, Magliano DJ, Taylor AW, Mitchell P, Giles GG, Shaw JE, Gill TK, Klaes E, Velentzis LS, Cumming RG, Hirani V, Laaksonen MA.

University of New South Wales, Sydney, Australia; Cancer Council New South Wales, Sydney, Australia; University of Sydney, Sydney, Australia; Cancer Council Victoria, Melbourne, Australia; The University of Melbourne, Melbourne, Australia; Australian National University, Canberra, Australia; University of Newcastle, Newcastle, Australia; Baker Heart and Diabetes Institute, Melbourne, Australia; University of Adelaide, Adelaide, Australia; Breast Cancer Network Australia, Melbourne, Australia.

Estimates of the future breast cancer burden preventable through modifications to current behaviours are lacking. We assessed the effect of individual and joint behaviour modifications on breast cancer burden for premenopausal and postmenopausal Australian women, and whether effects differed between population subgroups. We linked pooled data from six Australian cohort studies (N=214,536) to national cancer and death registries, and estimated the strength of the associations between behaviours causally related to cancer incidence and death using adjusted proportional hazards models. We estimated exposure prevalence from representative health surveys. We combined these estimates to calculate Population Attributable Fractions (PAFs) with 95% confidence intervals (CIs), and compared PAFs for population subgroups. During the first 10-years follow-up, there were 640 incident breast cancers for premenopausal women, 2,632 for postmenopausal women, and 8,761 deaths from any cause. Of future breast cancers for premenopausal women, any regular alcohol consumption explains 12.6% (CI=4.3-20.2%), current use of oral contraceptives for ≥5 years 7.1% (CI=0.3-13.5%), and these factors combined 18.8% (CI=9.1-27.4%). Of future breast cancers for postmenopausal women, overweight or obesity (BMI ≥25 kg/m2) explains 12.8% (CI=7.8-17.5%), current use of menopausal hormone therapy (MHT) 6.9% (CI=4.8-8.9%), any regular alcohol consumption 6.6% (CI=1.5-11.4%), and these factors combined 24.2% (CI=17.6-30.3%). The MHT-related postmenopausal breast cancer burden varied by body fatness, alcohol consumption and socio-economic status, the body fatness-related postmenopausal breast cancer burden by alcohol consumption and educational attainment, and the alcohol-related postmenopausal breast cancer burden by breast feeding history. Our results provide evidence to support targeted and population-level cancer control activities.

DOI: 10.1002/ijc.32231