Breast Cancer Risk After Salpingo-Oophorectomy in Healthy BRCA1/2 Mutation Carriers: Revisiting the Evidence for Risk Reduction
Breast Cancer Risk After Salpingo-Oophorectomy in Healthy BRCA1/2 Mutation Carriers: Revisiting the Evidence for Risk Reduction
复制标题
DOI：
10.1093/jnci/djv033
复制
发表时间：
2015-05-01
影响因子：
10.3
通讯作者：
Hooning, M. J.
中科院分区：
文献类型：
--
作者：
Heemskerk-Gerritsen, B. A. M.;Seynaeve, C.;Hooning, M. J.
背景：之前的研究报告称，BRCA1/2 突变携带者在进行降低风险的输卵管卵巢切除术 (RRSO) 后，乳腺癌 (BC) 风险降低了约 50%，但可能存在多种类型的偏差。这项全国性队列研究的目的是评估 RRSO 后估计的 BC 风险降低的潜在偏差。方法：我们从正在进行的荷兰遗传性乳腺癌和卵巢癌全国队列研究 (HEBON) 中选择了 BRCA1/2 突变携带者。首先，我们复制了之前在 RRSO 后关于 BC 风险的四项主要研究中应用的分析方法。使用 Cox 比例风险模型计算风险比，使用条件逻辑回归计算优势比。其次，我们分析了修改后的设计中的数据，以便使用扩展的 Cox 模型（以 RRSO 作为时间相关变量来计算风险比）进一步最小化偏差。我们的方法与以前的研究之间最重要的区别是要求在 DNA 诊断之日没有癌症病史，并纳入 RRSO 之前的人次。结果：应用前面描述的四种分析方法和 551 至 934 名 BRCA1/2 突变携带者的数据，中位随访时间为 2.7 至 4.6 年，比值比为 0.61（95% 置信区间 [CI] = 0.35 至1.08），风险比分别为 0.36（95% CI = 0.25 至 0.53）、0.62（95% CI = 0.39 至 0.99）和 0.49（95% CI = 0.33 至 0.71），与之前的研究结果相似。对于修订后的分析，我们纳入了 822 名 BRCA1/2 突变携带者。经过 3.2 年的中位随访期后，我们获得了 1.09 的风险比（95% CI = 0.67 至 1.77）。结论：在之前的研究中，BRCA1/2 突变携带者 RRSO 后 BC 风险的降低可能因偏倚而被高估。使用最大限度地消除偏见的设计，我们没有发现任何保护作用的证据。
Background: Previous studies have reported a breast cancer (BC) risk reduction of approximately 50% after risk-reducing salpingo-oophorectomy (RRSO) in BRCA1/2 mutation carriers, but may have been subject to several types of bias. The purpose of this nationwide cohort study was to assess potential bias in the estimated BC risk reduction after RRSO.Methods: We selected BRCA1/2 mutation carriers from an ongoing nationwide cohort study on Hereditary Breast and Ovarian Cancer in the Netherlands (HEBON). First, we replicated the analytical methods as previously applied in four major studies on BC risk after RRSO. Cox proportional hazards models were used to calculate hazard ratios and conditional logistic regression to calculate odds ratios. Secondly, we analyzed the data in a revised design in order to further minimize bias using an extended Cox model with RRSO as a time-dependent variable to calculate the hazard ratio. The most important differences between our approach and those of previous studies were the requirement of no history of cancer at the date of DNA diagnosis and the inclusion of person-time preceding RRSO.Results: Applying the four previously described analytical methods and the data of 551 to 934 BRCA1/2 mutation carriers with a median follow-up of 2.7 to 4.6 years, the odds ratio was 0.61 (95% confidence interval [CI] = 0.35 to 1.08), and the hazard ratios were 0.36 (95% CI = 0.25 to 0.53), 0.62 (95% CI = 0.39 to 0.99), and 0.49 (95% CI = 0.33 to 0.71), being similar to earlier findings. For the revised analysis, we included 822 BRCA1/2 mutation carriers. After a median follow-up period of 3.2 years, we obtained a hazard ratio of 1.09 (95% CI = 0.67 to 1.77).Conclusion: In previous studies, BC risk reduction after RRSO in BRCA1/2 mutation carriers may have been overestimated because of bias. Using a design that maximally eliminated bias, we found no evidence for a protective effect.