Department of Gynecology and Obstetrics, University Hospitals Leuven, KU Leuven - University of Leuven, Herestraat 49, 3000, Louvain, Belgium.
Interuniversity Centre for Biostatistics and Statistical Bioinformatics, 3000, Louvain, Belgium.
Breast Cancer Res Treat. 2018 Jun;169(3):481-487. doi: 10.1007/s10549-018-4701-6. Epub 2018 Feb 10.
Pregnancy affects breast cancer risk but how it affects the subtype and prognosis remain controversial. We studied the effect of parity and time since last birth on breast cancer subtype and outcome.
We conducted a retrospective multivariate cohort study including all premenopausal women with early breast cancer aged ≤ 50 years (N = 1306) at diagnosis at the University Hospitals Leuven (Jan. 2000-Dec. 2009). Primary study endpoints were the breast cancer subtype, disease-free survival, and distant disease-free survival by parity and time since last birth. Statistical methods used were baseline-category logits models and Cox proportional hazard models. Multivariable models were used to correct for possible confounders.
Breast cancer subtypes did not differ between nulliparous (N = 266) and parous women (N = 1040) but subtypes differed significantly in parous women by time since last birth (p < 0.001). Tumors within 5 years of last birth were proportionally more likely triple negative and HER-2 like, even when corrected for age at diagnosis. After a mean follow-up period of 10 years, parous women had a better disease-free survival compared to nulliparous women (HR 0.733; CI 0.560-0.961; p = 0.025, HR 0.738; CI 0.559-0.974; p = 0.032 before and after correction for known prognostic factors, respectively). In parous women, a longer time since last birth was correlated with a longer disease-free survival compared to patients with a recent pregnancy (HR 0.976; CI 0.957-0.996; p = 0.018). However, after correction, this association completely disappeared (HR 1.010; CI 0.982-1.040; p = 0.480).
We observed a better disease-free survival for parous than nulliparous women. The influence of recent birth on disease-free survival is probably due to tumor and patient characteristics, as recent birth is associated with more aggressive subtypes.
妊娠会影响乳腺癌的风险，但它如何影响乳腺癌亚型和预后仍存在争议。我们研究了生育次数和上次分娩后时间对乳腺癌亚型和结局的影响。
我们进行了一项回顾性多变量队列研究，纳入了 2000 年 1 月至 2009 年 12 月在鲁汶大学医院诊断为≤50 岁的所有绝经前早期乳腺癌女性（N=1306）。主要研究终点是根据生育次数和上次分娩后时间的乳腺癌亚型、无病生存率和远处无病生存率。使用的统计方法是基线类别对数模型和 Cox 比例风险模型。多变量模型用于校正可能的混杂因素。
在未生育的女性（N=266）和生育的女性（N=1040）中，乳腺癌亚型没有差异，但在生育的女性中，根据上次分娩后的时间，亚型有显著差异（p<0.001）。在最近一次分娩后 5 年内的肿瘤，比例上更可能是三阴性和 HER-2 样，即使在诊断时校正了年龄。平均随访 10 年后，与未生育的女性相比，生育的女性无病生存率更好（HR 0.733；95%CI 0.560-0.961；p=0.025，HR 0.738；95%CI 0.559-0.974；p=0.032，在分别校正已知预后因素前后）。在生育的女性中，与最近怀孕的患者相比，上次分娩后时间较长与无病生存率较长相关（HR 0.976；95%CI 0.957-0.996；p=0.018）。然而，校正后，这种关联完全消失（HR 1.010；95%CI 0.982-1.040；p=0.480）。
我们观察到生育的女性比未生育的女性无病生存率更好。最近分娩对无病生存率的影响可能是由于肿瘤和患者特征，因为最近分娩与更具侵袭性的亚型有关。
