PO.BCS02.04 · 生物信息与计算
结合乳腺癌多基因风险评分的MIRAI 5年风险模型的开发与评估
Development and evaluation of a MIRAI 5-year risk model with a breast cancer polygenic risk score
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
从数字乳腺X线摄影估计的人工智能（AI）评分可预测未来乳腺癌（BC）风险。MIRAI是一种深度学习BC风险模型，从四张筛查全视野数字乳腺X线摄影提供连续的5年总体BC风险。以多基因风险评分（BC-PRS）形式存在的常见种系遗传变异与BC风险增加相关，并可能改善MIRAI的5年风险预测。我们的目标是开发一个纳入BC-PRS的更新版MIRAI 5年风险模型（MIRAI+PRS），并评估其在总体和浸润性BC方面的判别准确性和校准度，与单独的MIRAI进行比较。我们通过将每位女性的5年MIRAI风险估计值乘以其基于BC-PRS相对于人群均值的相对风险来开发MIRAI+PRS模型。我们在Mayo Clinic Biobank乳腺X线摄影队列中评估了这些模型，该队列包含12,307名既往无BC病史的女性；5年内诊断出176例浸润性和250例总体BC。MIRAI在最接近入组但至少在BC诊断前6个月的筛查乳腺X线摄影上进行估计。通过C统计量评估的判别准确性对于总体BC和浸润性BC，MIRAI+PRS与MIRAI模型均较高且相似（见表）。通过观察值与期望值（O/E）比评估的校准度，MIRAI+PRS与MIRAI预测在BC结局方面也相似（见表），尽管MIRAI+PRS在最低风险十分位数（<1.67%）的十分位数特异性O/E比方面有所改善。无论是否包含PRS，MIRAI对浸润性癌症的校准都较差。总之，与MIRAI模型相比，MIRAI+PRS风险模型在判别准确性或总体校准方面未产生显著差异，但有证据表明对于5年风险低于1.67%的女性校准有所改善。对于浸润性BC，无论是否纳入PRS，该模型的校准都较差，这凸显了针对与临床干预相关的BC结局训练AI模型的重要性。
MIRAI和MIRAI+PRS 5年风险模型在Mayo Clinic Biobank乳腺X线摄影队列中的表现 模型 每SD的HR（95% CI） C指数（95% CI） 观察/期望比（95% CI） 总体BC（浸润性+DCIS），N=250 MIRAI 5年风险（对数）1.69（1.55, 1.84）0.71（0.68, 0.74）0.96（0.84, 1.09） MIRAI+PRS 5年风险（对数）1.97（1.78, 2.19）0.72（0.69, 0.75）0.93（0.82, 1.05） 浸润性BC，N=176 MIRAI 5年风险（对数）1.68（1.51, 1.86）0.71（0.67, 0.75）0.68（0.58, 0.78） MIRAI+PRS 5年风险（对数）1.99（1.76, 2.24）0.73（0.69, 0.76）0.66（0.56, 0.76）
查看英文原文 English abstract
Artificial intelligence (AI)-scores estimated from digital mammograms predict future breast cancer (BC) risk. MIRAI is a deep learning BC risk model that provides a continuous 5-year risk of overall BC from four screening full field digital mammograms. Common germline genetic variation in the form of a polygenic risk score (BC-PRS) is associated with increased BC risk and may improve MIRAI's 5-year risk prediction. Our goal was to develop an updated MIRAI 5-year risk model that incorporates the BC-PRS (MIRAI+PRS) and to evaluate its discriminatory accuracy and calibration for both overall and invasive BC compared to MIRAI alone. We developed the MIRAI+PRS model by multiplying each woman's 5-year MIRAI risk estimate by their relative risk based on their BC-PRS relative to the population mean. We evaluated the models within the Mayo Clinic Biobank mammography cohort, comprised of 12,307 women without a prior history of BC; 176 invasive and 250 overall BC were diagnosed within 5 years. MIRAI was estimated on screening mammograms closest to enrollment but at least 6 months prior to BC. Discriminatory accuracy, assessed by C-statistic, was high and similar for MIRAI+PRS vs. MIRAI models, for overall BC and invasive BC (Table). Calibration assessed by observed to expected (O/E) ratios was also similar for MIRAI+PRS compared to MIRAI predictions for BC outcomes (Table), although there was improvement in decile-specific O/E ratios across the lowest risk deciles (<1.67%) for MIRAI+PRS. Calibration for invasive cancer was poor for MIRAI with or without PRS. In summary, the MIRAI+PRS risk model did not result in significant difference of discriminatory accuracy or overall calibration compared to the MIRAI model, but there was evidence for improved calibration for women with 5-year risk below 1.67%. For invasive BC, the model had poor calibration regardless of whether PRS was included, underscoring the importance of training AI models for BC outcomes that are associated with a clinical intervention.
MIRAI and MIRAI+PRS 5-year risk model performance within the Mayo Clinic Biobank mammography cohort Model HR per SD (95% CI) C-index (95% CI) Obs/Exp Ratio (95% CI) Overall BC (Invasive + DCIS), N=250 MIRAI 5 yr Risk (log) 1.69 (1.55, 1.84) 0.71 (0.68, 0.74) 0.96 (0.84, 1.09) MIRAI+PRS 5 yr Risk (log) 1.97 (1.78, 2.19) 0.72 (0.69, 0.75) 0.93 (0.82, 1.05) Invasive BC, N=176 MIRAI 5 yr Risk (log) 1.68 (1.51, 1.86) 0.71 (0.67, 0.75) 0.68 (0.58, 0.78) MIRAI+PRS 5 yr Risk (log) 1.99 (1.76, 2.24) 0.73 (0.69, 0.76) 0.66 (0.56, 0.76)
利益披露 Disclosure
C. G. Scott, None..
P. Kraft, None..
I. Banerjee, None..
R. Correa Medero, None..
F. J. Couch, None..
K. Kerlikowske, None..
S. J. Winham, None..
C. M. Vachon, None.