From the Division of Neonatology, Children's Hospital of Philadelphia, Philadelphia, PA (Dr Burris and Ms Ledyard); Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA (Dr Burris); Department of Neonatology, Beth Israel Deaconess Medical Center, Boston, MA (Dr Burris).
Department of Obstetrics and Gynecology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA (Dr Gerson); Department of Microbiology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA (Dr Gerson).
Am J Obstet Gynecol MFM. 2023 Jan;5(1):100783. doi: 10.1016/j.ajogmf.2022.100783. Epub 2022 Oct 22.
Preterm birth remains a major public health issue affecting 10% of all pregnancies and increases risks of neonatal morbidity and mortality. Approximately 50% to 60% of preterm births are spontaneous, resulting from preterm premature rupture of membranes or preterm labor. The pathogenesis of spontaneous preterm birth is incompletely understood, and prediction of preterm birth remains elusive. Accurate prediction of preterm birth would reduce infant morbidity and mortality through targeted patient referral to hospitals equipped to care for preterm infants. Two previous studies have analyzed cervical microRNAs in association with spontaneous preterm birth and the length of gestation, but the extent to which microRNAs serve as predictive biomarkers remains unknown.
This study aimed to examine associations between cervical microRNA expression and spontaneous preterm birth, with the specific goal of identifying a subset of microRNAs that predict spontaneous preterm birth.
We performed a prospective, nested, case-control study of 25 cases with spontaneous preterm birth and 49 term controls. Controls were matched to cases in a 2:1 ratio on the basis of age, parity, and self-identified race. Cervical swabs were collected at a mean gestational age of 17.1 (4.8) weeks of gestation, and microRNAs were analyzed using a quantitative polymerase chain reaction array. Normalized microRNA expression was compared between cases and controls, and a false discovery rate of 0.2 was applied to account for multiple comparisons. Histopathologic analysis of slides of cervical swab samples was performed to quantify leukocyte burden for adjustment in conditional regression models. We explored the use of Relief-based unsupervised identification of top microRNAs and support vector machines to predict spontaneous preterm birth. We performed microRNA enrichment analysis to explore potential biologic targets and pathways in which up-regulated microRNAs might be involved.
Of the 754 microRNAs on the polymerase chain reaction array, 346 were detected in ≥75% of participants' cervical swabs. Average cervical microRNA expression was significantly higher in cases of spontaneous preterm birth than in controls (P=.01). There were 95 significantly up-regulated individual microRNAs (>2-fold change) in cases of subsequent spontaneous preterm birth compared with term controls (P<.05; q<0.2). Notably, miR-143, miR-30e-3p, and miR-199b were all significantly up-regulated, which is consistent with the 1 previous study of cervical microRNA and spontaneous preterm birth. A Relief-based, novel variable (feature) selection machine learning approach had low-to-moderate prediction accuracy, with an area under the receiver operating curve of 0.71. Enrichment analysis revealed that identified microRNAs may modulate inflammatory cell signaling.
In this prospective nested case-control study of cervical microRNA expression and spontaneous preterm birth, we identified a global increase in microRNA expression and up-regulation of 95 distinct microRNAs in association with subsequent spontaneous preterm birth. Larger and more diverse studies are required to determine the ability of microRNAs to accurately predict spontaneous preterm birth, and mechanistic work to facilitate development of novel therapeutic interventions to prevent spontaneous preterm birth is warranted.
早产仍然是一个主要的公共卫生问题，影响了所有妊娠的 10%，并增加了新生儿发病率和死亡率的风险。大约 50%至 60%的早产是自发性的，由胎膜早破或早产引起。自发性早产的发病机制尚未完全了解，早产的预测仍然难以捉摸。早产的准确预测将通过将患者转介到有能力照顾早产儿的医院，从而减少婴儿的发病率和死亡率。先前有两项研究分析了与自发性早产和妊娠时间相关的宫颈微 RNA，但微 RNA 是否可以作为预测生物标志物尚不清楚。
本研究旨在探讨宫颈微 RNA 表达与自发性早产之间的关系，并确定一组能够预测自发性早产的微 RNA。
我们进行了一项前瞻性、嵌套、病例对照研究，纳入了 25 例自发性早产病例和 49 例足月对照组。对照组根据年龄、产次和自我认定的种族，以 2:1 的比例与病例相匹配。宫颈拭子在平均妊娠 17.1(4.8)周时采集，使用定量聚合酶链反应阵列分析微 RNA。比较病例组和对照组之间的微 RNA 表达差异，并应用 False Discovery Rate (FDR) 0.2 进行多重比较校正。对宫颈拭子样本的病理切片进行白细胞负荷的组织病理学分析，以进行条件回归模型的调整。我们探索了基于 Relief 的无监督识别顶级微 RNA 和支持向量机的方法，以预测自发性早产。我们进行了微 RNA 富集分析，以探索潜在的生物学靶标和途径，其中上调的微 RNA 可能参与其中。
在聚合酶链反应阵列上的 754 个微 RNA 中，346 个在≥75%的参与者的宫颈拭子中被检测到。自发性早产病例的宫颈微 RNA 表达明显高于对照组(P=.01)。与足月对照组相比，自发性早产病例中有 95 个个体微 RNA 显著上调(>2 倍变化)(P<.05; q<0.2)。值得注意的是，miR-143、miR-30e-3p 和 miR-199b 均显著上调，这与之前的一项关于宫颈微 RNA 和自发性早产的研究一致。基于 Relief 的新型变量(特征)选择机器学习方法具有较低到中等的预测准确性，其接受者操作特征曲线下面积为 0.71。富集分析显示，鉴定出的微 RNA 可能调节炎症细胞信号。
在这项关于宫颈微 RNA 表达与自发性早产的前瞻性嵌套病例对照研究中，我们发现与随后自发性早产相关的微 RNA 表达普遍增加，95 种不同的微 RNA 上调。需要更大和更多样化的研究来确定微 RNA 准确预测自发性早产的能力，并需要进行机制研究，以促进预防自发性早产的新治疗干预措施的发展。
