BACKGROUND The purpose of this study was to (1) perform the first analysis of bone-derived DNA methylation, (2) compare DNA methylation clocks derived from bone with those derived from whole blood, and (3) establish a relationship between DNA methylation age and 1-year mortality within the geriatric hip fracture population. METHODS Patients ≥65 years old who presented to a Level-I trauma center with a hip fracture were prospectively enrolled from 2020 to 2021. Preoperative whole blood and intraoperative bone samples were collected. Following DNA extraction, RRBS (reduced representation bisulfite sequencing) libraries for methylation clock analysis were prepared. Sequencing data were analyzed using computational algorithms previously described by Horvath et al. to build a regression model of methylation (biological) age for each tissue type. Student t tests were used to analyze differences (Δ) in methylation age versus chronological age. Correlation between blood and bone methylation ages was expressed using the Pearson R coefficient. RESULTS Blood and bone samples were collected from 47 patients. DNA extraction, sequencing, and methylation analysis were performed on 24 specimens from 12 subjects. Mean age at presentation was 85.4 ± 8.65 years. There was no difference in DNA extraction yield between the blood and bone samples (p = 0.935). The mean follow-up duration was 12.4 ± 4.3 months. The mortality cohort (4 patients, 33%) showed a mean ΔAgeBone of 18.33 ± 6.47 years and mean ΔAgeBlood of 16.93 ± 4.02 years. In comparison, the survival cohort showed a significantly lower mean ΔAgeBone and ΔAgeBlood (7.86 ± 6.7 and 7.31 ± 7.71 years; p = 0.026 and 0.039, respectively). Bone-derived methylation age was strongly correlated with blood-derived methylation age (R = 0.81; p = 0.0016). CONCLUSIONS Bone-derived DNA methylation clocks were found to be both feasible and strongly correlated with those derived from whole blood within a geriatric hip fracture population. Mortality was independently associated with the DNA methylation age, and that age was approximately 17 years greater than chronological age in the mortality cohort. The results of the present study suggest that prevention of advanced DNA methylation may play a key role in decreasing mortality following hip fracture. LEVEL OF EVIDENCE Prognostic Level I. See Instructions for Authors for a complete description of levels of evidence.

中文翻译：

---

血液和骨源性 DNA 甲基化年龄可预测老年髋部骨折后的死亡率：一项初步研究。


背景 本研究的目的是 （1） 对骨源性 DNA 甲基化进行首次分析，（2） 比较来自骨骼的 DNA 甲基化时钟与来自全血的 DNA 甲基化时钟，以及 （3） 在老年髋部骨折人群中建立 DNA 甲基化年龄与 1 年死亡率之间的关系。方法 前瞻性纳入 2020 年至 2021 年因髋部骨折到 I 级创伤中心就诊的 ≥ 65 岁患者。收集术前全血和术中骨标本。DNA 提取后，制备用于甲基化时钟分析的 RRBS （还原表示亚硫酸氢盐测序） 文库。使用 Horvath 等人之前描述的计算算法分析测序数据，以构建每种组织类型的甲基化（生物）年龄回归模型。学生 t 检验用于分析甲基化年龄与实际年龄的差异 （Δ）。血液和骨甲基化年龄之间的相关性使用 Pearson R 系数表示。结果 收集 47 例患者的血液和骨骼样本。对 12 名受试者的 24 份标本进行 DNA 提取、测序和甲基化分析。平均就诊年龄为 85.4 ± 8.65 岁。血液和骨骼样本之间的 DNA 提取率没有差异 （p = 0.935）。平均随访时间为 12.4 ± 4.3 个月。死亡率队列 （4 例患者，33%） 显示平均 ΔAgeBone 为 18.33 ± 6.47 岁，平均 ΔAgeBlood 为 16.93 ± 4.02 岁。相比之下，生存队列显示平均 ΔAgeBone 和 ΔAgeBlood 显著降低 （7.86 ± 6.7 和 7.31 ± 7.71 岁;p = 0.026 和 0.039）。骨源性甲基化年龄与血源性甲基化年龄密切相关 （R = 0。81;p = 0.0016）。结论 发现骨源性 DNA 甲基化时钟既可行，又与老年髋部骨折人群中来自全血的 DNA 甲基化时钟密切相关。死亡率与 DNA 甲基化年龄独立相关，该年龄比死亡率队列中的实际年龄大约 17 岁。本研究的结果表明，预防晚期 DNA 甲基化可能在降低髋部骨折后死亡率方面发挥关键作用。证据水平 预后 I 级。有关证据级别的完整描述，请参阅作者说明。