Moshina Nataliia, Gjesvik Jonas, Hovda Tone, Koch Henrik W, Backmann Heinrich A, Hofvind Solveig
The Cancer Registry of Norway, Department of Screening programs, Norwegian Institute of Public Health, Oslo, Norway.
Department of Radiology, Vestre Viken Hospital Trust, Drammen, Norway.
Breast Cancer Res. 2025 May 16;27(1):83. doi: 10.1186/s13058-025-02037-2.
Mammographic density is known to decrease over time in postmenopausal women. Longitudinal changes in mammographic density prior to breast cancer diagnosis have been widely discussed and less density reduction has been observed for breast developing versus not developing cancer. We aimed to verify these findings among participants of BreastScreen Norway.
In this retrospective cohort study, data from 78,182 women aged 50-69 years who attended three consecutive screening rounds between 2007 and 2020 were included. Among those women, 970 were diagnosed with screen-detected and 308 with interval cancer. Mammographic density data was obtained from an automated software and included absolute (cm) and percent (%) dense volume for each breast and for each woman, per examination. A linear mixed-effects regression model estimating differences in density between the breast developing and not developing cancer was applied to evaluate longitudinal changes, separately for absolute and percent dense volume. The model was adjusted for age at first screening examination, breast volume, follow-up time, history of benign breast disease, body mass index, family history, hormone therapy, use of alcohol and smoking. Results were presented as linear regression coefficient estimates with 95% confidence intervals (CI).
Mean age at the third screening examination for women without breast cancer was 62.5 (standard deviation, SD: 5.1) years, while mean age at diagnosis was 62.3 (SD: 4.4) years for women with screen-detected cancer and 61.9 (SD: 4.8) years for women with interval cancer. In our model, absolute and percent dense volume decreased with follow-up time, estimate=-0.010 (95%CI -0.010; -0.009) and estimate=-0.013 (95%CI -0.014; -0.013), respectively, indicating the overall negative effect of time on mammographic density. The interaction between time and development of breast cancer was positive for absolute and percent dense volume, estimate = 0.009 (95%CI 0.004; 0.014) for both, which implied that mammographic density in breasts developing cancer was stable or slightly decreasing.
Less reduction in longitudinally assessed mammographic density was observed for breasts developing versus not developing cancer in our study. This difference might be used for more precise 4-6 years breast cancer risk prediction and screening personalization.
已知绝经后女性的乳腺X线密度会随时间降低。乳腺癌诊断前乳腺X线密度的纵向变化已得到广泛讨论，与未发生癌症的乳房相比，发生癌症的乳房密度降低较少。我们旨在验证挪威乳腺癌筛查项目参与者中的这些发现。
在这项回顾性队列研究中，纳入了2007年至2020年间连续参加三轮筛查的78182名年龄在50 - 69岁之间的女性的数据。在这些女性中，970人被诊断为筛查发现的癌症，308人被诊断为间期癌。乳腺X线密度数据通过自动化软件获得，包括每次检查时每个乳房和每名女性的绝对（cm）致密体积和百分比（%）致密体积。应用线性混合效应回归模型估计发生癌症和未发生癌症的乳房之间的密度差异，分别针对绝对致密体积和百分比致密体积评估纵向变化。该模型针对首次筛查时的年龄、乳房体积、随访时间、良性乳腺疾病史、体重指数、家族史、激素治疗、饮酒和吸烟情况进行了调整。结果以线性回归系数估计值及95%置信区间（CI）表示。
未患乳腺癌女性在第三次筛查时的平均年龄为62.5（标准差，SD：5.1）岁，而筛查发现癌症的女性诊断时的平均年龄为62.3（SD：4.4）岁，间期癌女性诊断时的平均年龄为61.9（SD：4.8）岁。在我们的模型中，绝对致密体积和百分比致密体积均随随访时间降低，估计值分别为 - 0.010（95%CI - 0.010； - 0.009）和 - 0.013（95%CI - 0.014； - 0.013），表明时间对乳腺X线密度具有总体负面影响。时间与乳腺癌发生之间的相互作用对于绝对致密体积和百分比致密体积均为正向，两者的估计值均为0.009（95%CI 0.004；0.014），这意味着发生癌症的乳房的乳腺X线密度稳定或略有下降。
在我们的研究中，与未发生癌症的乳房相比，发生癌症的乳房在纵向评估的乳腺X线密度上降低较少。这种差异可用于更精确的4 - 6年乳腺癌风险预测和筛查个性化。
