Division of Pediatric Cardiothoracic Surgery, Department of Surgery, Washington University School of Medicine, St. Louis, MO, United States.
Department of Immunology, Sidra Medicine, Doha, Qatar.
Front Immunol. 2024 May 16;15:1281111. doi: 10.3389/fimmu.2024.1281111. eCollection 2024.
Sepsis remains a major cause of mortality and morbidity in infants. In recent years, several gene marker strategies for the early identification of sepsis have been proposed but only a few have been independently validated for adult cohorts and applicability to infant sepsis remains unclear. Biomarkers to assess disease severity and risks of shock also represent an important unmet need.
To elucidate characteristics driving sepsis in infants, we assembled a multi-transcriptomic dataset from public microarray datasets originating from five independent studies pertaining to bacterial sepsis in infant < 6-months of age (total n=335). We utilized a COmbat co-normalization strategy to enable comparative evaluation across multiple studies while preserving the relationship between cases and controls.
We found good concordance with only two out of seven of the published adult sepsis gene signatures (accuracy > 80%), highlighting the narrow utility of adult-derived signatures for infant diagnosis. Pseudotime analysis of individual subjects' gene expression profiles showed a continuum of molecular changes forming tight clusters concurrent with disease progression between healthy controls and septic shock cases. In depth gene expression analyses between bacteremia, septic shock, and healthy controls characterized lymphocyte activity, hemostatic processes, and heightened innate immunity during the molecular transition toward a state of shock.
Our analysis revealed the presence of multiple significant transcriptomic perturbations that occur during the progression to septic shock in infants that are characterized by late-stage induction of clotting factors, in parallel with a heightened innate immune response and a suppression of adaptive cell functionality.
败血症仍然是婴儿死亡和发病的主要原因。近年来,已经提出了几种基因标志物策略来早期识别败血症,但只有少数已被独立验证适用于成人队列,其在婴儿败血症中的适用性仍不清楚。评估疾病严重程度和休克风险的生物标志物也代表了一个重要的未满足需求。
为了阐明导致婴儿败血症的特征,我们从五个独立的研究中收集了公共微阵列数据集,这些研究涉及 6 个月以下婴儿的细菌性败血症(总 n=335)。我们利用 COmbat 共归一化策略来实现跨多个研究的比较评估,同时保留病例和对照之间的关系。
我们发现只有两个成人败血症基因特征中的七个(准确性>80%)具有良好的一致性,这突出表明成人来源的特征对于婴儿诊断的应用范围狭窄。对个体患者基因表达谱的伪时间分析显示,分子变化的连续性形成了紧密的簇,与健康对照组和败血症性休克组之间的疾病进展同步。对菌血症、败血症性休克和健康对照组之间的深入基因表达分析表明,在向休克状态的分子转变过程中,淋巴细胞活性、止血过程和先天免疫增强。
我们的分析揭示了在婴儿发生败血症性休克的进展过程中存在多种显著的转录组扰动,这些扰动的特征是凝血因子的晚期诱导,与先天免疫反应增强和适应性细胞功能抑制同时发生。