The School of Public Health, Sun Yat-sen University, 74 Zhongshan 2nd Road, Guangzhou, 510080, People's Republic of China.
Mailman School of Public Health, Columbia University, New York, USA.
BMC Cancer. 2018 Oct 17;18(1):989. doi: 10.1186/s12885-018-4884-6.
DNA methylation (DNAm) age was found to be an indicator for all-cause mortality, cancer incidence, and longevity, but no study has involved in the associations of DNAm age with the prognosis of breast cancer.
We retrieved information of 1076 breast cancer patients from Genomic Data Commons (GDC) data portal on March 30, 2017, including breast cancer DNAm profiling, demographic features, clinicopathological parameters, recurrence, and all-cause fatality. Horvath's method was applied to calculate the DNAm age. Cox proportional hazards regression models were used to test the associations between DNAm age of the cancerous tissues and the prognosis (recurrence of breast cancer and all-cause fatality) with or without adjusting for chronological age and clinicopathological parameters.
The DNAm age was markedly decelerated in the patients who were premenopausal, ER or PR negative, HER2-enriched or basal-like than their counterparts. In the first five-year follow-up dataset for survival, every ten-year increase in DNAm age was associated with a 15% decrease in fatality; subjects with DNAm age in the second (HR: 0.52; 95%CI: 0.29-0.92), the third (HR: 0.49; 95%CI: 0.27-0.87) and the fourth quartile (HR: 0.38; 95%CI: 0.20-0.72) had significant longer survival time than those in the first quartile. In the first five-year follow-up dataset for recurrence, every ten-year increase in DNAm age was associated with a 14% decrease of the recurrence; in the categorical analysis, a clear dose-response was shown (P for trend =0.02) and the fourth quartile was associated with a longer recurrence free survival (HR: 0.32; 95%CI: 0.14-0.74). In the full follow-up dataset, similar results were obtained.
DNAm age of breast cancer tissue, which associated with menopausal status and pathological features, was a strong independent predictor of the prognosis. It was suggested that the prognosis of breast cancer was related to intrinsic biological changes and specific molecular targets for treatment of breast cancer may be implicit.
DNA 甲基化（DNAm）年龄被发现是全因死亡率、癌症发病率和长寿的指标，但尚无研究涉及 DNAm 年龄与乳腺癌预后的关系。
我们于 2017 年 3 月 30 日从基因组数据共享（GDC）数据门户中检索了 1076 名乳腺癌患者的信息，包括乳腺癌 DNAm 分析、人口统计学特征、临床病理参数、复发和全因死亡率。应用 Horvath 方法计算 DNAm 年龄。Cox 比例风险回归模型用于测试癌症组织的 DNAm 年龄与（乳腺癌复发和全因死亡率）预后之间的关系，无论是否调整了年龄和临床病理参数。
与相应患者相比，绝经前、ER 或 PR 阴性、HER2 富集或基底样的患者 DNAm 年龄明显减慢。在生存的前五随访数据集，DNAm 年龄每增加十年，死亡率降低 15%；DNAm 年龄处于第二（HR：0.52；95%CI：0.29-0.92）、第三（HR：0.49；95%CI：0.27-0.87）和第四四分位数（HR：0.38；95%CI：0.20-0.72）的患者比处于第一四分位数的患者具有显著更长的生存时间。在复发的前五随访数据集，DNAm 年龄每增加十年，复发率降低 14%；在分类分析中，显示出明显的剂量反应（趋势 P=0.02），第四四分位数与无复发生存时间较长相关（HR：0.32；95%CI：0.14-0.74）。在全随访数据集也得到了类似的结果。
与绝经状态和病理特征相关的乳腺癌组织的 DNAm 年龄是预后的强独立预测因子。这表明乳腺癌的预后与内在的生物学变化有关，并且乳腺癌的治疗可能隐含特定的分子靶标。
