PO.PS01.06 · 人群科学
体育活动与绝经后乳腺癌女性的总死亡率及癌症特异性死亡率：一项随访19年的竞争风险分析
Physical activity and overall and cancer-specific mortality in postmenopausal women with breast cancer: A competing risks analysis with 19-year follow-up
该海报暂无可下载的资料
AACR 官方页面
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
背景：随着乳腺癌幸存者寿命延长，非癌症死亡与癌症死亡形成竞争。然而，许多评估体育活动的研究未考虑竞争风险，也未充分检验绝经状态的效应修饰作用。因此，我们旨在评估体育活动与乳腺癌女性死亡率之间的关联，同时考虑绝经状态和竞争风险。
方法：我们分析了墨西哥多中心CAMA研究中的1000例乳腺癌病例，随访19年。乳腺癌诊断或症状出现前通常每周从事的中至高强度休闲体育活动时长以2.5小时/周为界二分。主要结局为总死亡率、癌症死亡率和非癌症死亡率。我们采用两阶段分析：（1）针对总死亡率的Cox模型，检验绝经状态作为效应修饰因子；（2）在具有显著效应的分层中采用Fine-Gray竞争风险模型。模型校正了源自DAG（有向无环图）的临床、生活方式和社会经济混杂因素。缺失数据采用基于随机森林的链式方程多重插补处理。
结果：绝经状态显著修饰了体育活动与总死亡率之间的关联（p=0.035）。在绝经后女性（n=584）中，体育活动与死亡率降低相关（见表），而在绝经前女性中无显著关联。
结论：在绝经后乳腺癌女性中，诊断或症状出现前的体育活动与长期随访时较低的总死亡率相关。这一益处反映了在考虑竞争性非癌症死亡时癌症特异性死亡率的显著降低。这些发现强调了绝经后体育活动的重要性，并突显了在生存期研究中需要采用能同时考虑绝经状态和竞争风险的方法学途径。
绝经后乳腺癌女性中体育活动与死亡结局的关联：结局 HR/sHR（95% CI） p值；总死亡率 0.59（0.41—0.84） 0.004；癌症死亡率 0.63（0.40—0.98） 0.042；非癌症死亡率 0.63（0.35—1.14） 0.125。模型校正了年龄、临床分期、合并症、膳食炎症指数、饮酒、吸烟、教育、居住地和医疗可及性。HR = 风险比（Cox模型）；sHR = 亚分布风险比（Fine-Gray模型）。
查看英文原文 English abstract
Background: As breast cancer survivors live longer, non-cancer mortality competes with cancer mortality. However, many studies evaluating physical activity have not accounted for competing risks or adequately tested effect modification by menopausal status. Therefore, we aimed to assess the association between physical activity and mortality in women with breast cancer, accounting for menopausal status and competing risks.
Methods: We analyzed 1000 breast cancer cases from the multicenter CAMA study in Mexico with 19-year follow-up. Usual hours of leisure-time moderate-to-vigorous physical activity before breast cancer diagnosis or symptoms were dichotomized at 2.5 h/week. Primary outcomes were overall, cancer, and non-cancer mortality. We used a two-stage analysis: (1) Cox model for overall mortality testing menopausal status as an effect modifier, and (2) Fine-Gray competing risks models in strata with significant effects. Models were adjusted for DAG-derived clinical, lifestyle, and socioeconomic confounders. Missing data were handled using Multiple Imputation by Chained Equations with Random Forest.
Results: Menopausal status significantly modified the association between physical activity and overall mortality (p=0.035). In postmenopausal women (n=584), physical activity was associated with reduced mortality (Table), with no significant association in premenopausal women.
Conclusions: In postmenopausal women with breast cancer, physical activity before diagnosis or symptoms was associated with lower overall mortality at long-term follow-up. This benefit reflects significantly lower cancer-specific mortality when accounting for competing non-cancer mortality. These findings underscore the importance of physical activity after menopause and highlight the need for methodological approaches that jointly consider menopausal status and competing risks in survivorship research.
Association of physical activity and mortality outcomes in postmenopausal women with breast cancer Outcome HR/sHR (95% CI) p-value Overall mortality 0.59 (0.41-0.84) 0.004 Cancer mortality 0.63 (0.40-0.98) 0.042 Non-cancer mortality 0.63 (0.35-1.14) 0.125 Models adjusted for age, clinical stage, comorbidities, Diet Inflammatory Index, alcohol, smoking, education, residence, and healthcare access. HR = hazard ratio (Cox model); sHR = subdistribution hazard ratio (Fine-Gray models).
利益披露 Disclosure
J. I. Castillo Silva, None..
L. S. Palacio Mejia, None..
J. E. Hernandez Avila, None..
S. Zamora Muñoz, None..
R. Rodriguez Valentin, None..
A. Angeles Llerenas, None..
A. Jiménez Avila, None..
G. Torres Mejía, None.