Hopper John L, Nguyen Tuong L, Stone Jennifer, Aujard Kelly, Matheson Melanie C, Abramson Michael J, Burgess John A, Walters E Haydn, Dite Gillian S, Bui Minh, Evans Christopher, Makalic Enes, Schmidt Daniel F, Ward Gail, Jenkins Mark A, Giles Graham G, Dharmage Shyamali C, Apicella Carmel
Centre for Epidemiology and Biostatistics, Melbourne School of Population and Global Health, The University of Melbourne, Level 3, 207 Bouverie Street, Carlton, 3053, VIC, Australia.
Seoul Department of Epidemiology and Institute of Health and Environment, School of Public Health, Seoul National University, Seoul, Korea.
Breast Cancer Res Treat. 2016 Feb;156(1):163-70. doi: 10.1007/s10549-016-3719-x. Epub 2016 Feb 23.
The aim of the present study is to determine if body mass index (BMI) during childhood is associated with the breast cancer risk factor 'adult mammographic density adjusted for age and BMI'. In 1968, the Tasmanian Longitudinal Health Study studied every Tasmanian school child born in 1961. We obtained measured heights and weights from annual school medical records across ages 7-15 years and imputed missing values. Between 2009 and 2012, we administered to 490 women a questionnaire that asked current height and weight and digitised at least one mammogram per woman. Absolute and percent mammographic densities were measured using the computer-assisted method CUMULUS. We used linear regression and adjusted for age at interview and log current BMI. The mammographic density measures were negatively associated: with log BMI at each age from 7 to 15 years (all p < 0.05); with the average of standardised log BMIs across ages 7-15 years (p < 0.0005); and more strongly with standardised log BMI measures closer to age 15 years (p < 0.03). Childhood BMI measures explained 7 and 10 % of the variance in absolute and percent mammographic densities, respectively, and 25 and 20 % of the association between current BMI and absolute and percent mammographic densities, respectively. Associations were not altered by adjustment for age at menarche. There is a negative association between BMI in late childhood and the adult mammographic density measures that predict breast cancer risk. This could explain, at least in part, why BMI in adolescence is negatively associated with breast cancer risk.
本研究的目的是确定儿童时期的体重指数（BMI）是否与乳腺癌风险因素“根据年龄和BMI调整后的成人乳房X线密度”相关。1968年，塔斯马尼亚纵向健康研究对1961年出生的每一名塔斯马尼亚学童进行了研究。我们从7至15岁期间的年度学校医疗记录中获取了测量的身高和体重，并对缺失值进行了估算。在2009年至2012年期间，我们对490名女性进行了问卷调查，询问了她们当前的身高和体重，并对每名女性的至少一张乳房X线照片进行了数字化处理。使用计算机辅助方法CUMULUS测量绝对乳房X线密度和乳房X线密度百分比。我们采用线性回归，并对访谈时的年龄和当前BMI的对数进行了调整。乳房X线密度测量结果呈负相关：与7至15岁各年龄的BMI对数相关（所有p<0.05）；与7至15岁标准化BMI对数的平均值相关（p<0.0005）；与更接近15岁的标准化BMI测量值相关性更强（p<0.03）。儿童时期的BMI测量值分别解释了绝对乳房X线密度和乳房X线密度百分比变异的7%和10%，以及当前BMI与绝对乳房X线密度和乳房X线密度百分比之间关联的25%和20%。月经初潮年龄的调整并未改变这种关联。儿童晚期的BMI与预测乳腺癌风险的成人乳房X线密度测量值之间存在负相关。这至少可以部分解释为什么青春期的BMI与乳腺癌风险呈负相关。
