Identifying women at high risk of osteoporotic fracture from aromatase inhibitor (AI) therapy for breast cancer is largely based on known risk factors for healthy postmenopausal women, which might not accurately reflect the risk in breast cancer patients post-AI therapy. To determine whether a polygenic score associated with fracture in healthy women is also significant in women treated with AIs for breast cancer, we used data from a prospective observational cohort of 2152 women diagnosed with hormonal receptor positive breast cancer treated with AIs as the initial endocrine therapy and examined a polygenic score of heel quantitative ultrasound speed of sound (gSOS) in relation to incident osteoporotic fracture after AI therapy during a median 6.1 years of follow up after AI initiation. In multivariable models, patients with the second and third highest tertiles (T) versus the lowest tertile of gSOS had significantly lower risk of fracture (T2: adjusted HR = 0.61, 95% CI: 0.46-0.80; T3: adjusted HR = 0.53, 95% CI: 0.40-0.70). The lower risk of fracture in patients with the highest tertile of gSOS remained significant after further adjustment for BMD at the hip (T3: adjusted HR = 0.62, 95% CI: 0.42-0.91). In conclusion, our analysis showed gSOS as a novel genetic predictor for fracture risk independent of BMD among breast cancer patients treated with AIs. Future studies are warranted to evaluate the performance of incorporating gSOS in prediction models for the risk of AI-related fracture in breast cancer patients.

识别乳腺癌芳香酶抑制剂（AI）治疗中骨质疏松性骨折高风险的女性很大程度上是基于健康绝经后女性的已知危险因素，这可能无法准确反映乳腺癌患者接受 AI 治疗后的风险。为了确定健康女性中与骨折相关的多基因评分对于接受 AI 治疗乳腺癌的女性是否也很重要，我们使用了前瞻性观察队列的数据，该队列由 2152 名诊断为激素受体阳性乳腺癌且接受 AI 治疗作为初始内分泌治疗的女性组成。在 AI 开始后平均 6.1 年的随访期间，检查了足跟定量超声声速 (gSOS) 的多基因评分与 AI 治疗后发生的骨质疏松性骨折的关系。在多变量模型中，gSOS 第二和第三最高三分位 (T) 的患者与最低三分位的患者相比，骨折风险显着降低（T2：调整后 HR = 0.61，95% CI：0.46-0.80；T3：调整后 HR = 0.53， 95% CI：0.40-0.70）。在进一步调整髋部 BMD 后，gSOS 最高三分位数的患者骨折风险仍然较低（T3：调整后 HR = 0.62，95% CI：0.42-0.91）。总之，我们的分析表明，gSOS 是接受 AI 治疗的乳腺癌患者骨折风险的一种新型遗传预测因子，与 BMD 无关。未来的研究有必要评估将 gSOS 纳入乳腺癌患者 AI 相关骨折风险预测模型中的性能。