背景:在子宫颈癌的不同阶段,阴道微生物群落在遗传水平上的适应性进化变化的探索仍然有限。这项研究旨在阐明整个宫颈疾病进展过程中阴道微生物群的突变谱,并随后建立诊断模型。
方法:这项研究利用了由151名受试者组成的宏基因组数据集,这些受试者分为四类:浸润性宫颈癌(CC)(n=42),宫颈上皮内瘤变(CIN)(n=43),HPV感染(HPVi)患者无宫颈病变(n=34),和健康对照(n=32)。分析的重点是微生物组丰度的变化,并提取了有关遗传变异的信息。因此,与CC相关的综合多峰微生物特征,包括分类学改变,突变特征,丰富的代谢功能途径,已确定。考虑基于单核苷酸变体(SNV)的基因特征,建立用于预测CC的诊断模型。
结果:在这项研究中,我们筛选并分析了CC进展过程中18个关键微生物菌株的丰度。此外,鉴定出71,6358个非冗余突变,主要由进一步注释到25,773个基因中的SNV组成。在四组中观察到改变的SNV丰度和突变类型。具体来说,HPV感染组中有9847个SNV,CC组中有14,892个SNV.此外,我们确定了良性和恶性组对应的两种不同的突变特征.富集的代谢途径显示出有限的相似性,四组中仅有两个重叠的途径。HPVi患者表现出活跃的核苷酸生物合成,而CC患者的信号传导和细胞相关蛋白家族的丰度明显更高。相比之下,健康对照显示糖代谢明显富集。此外,基于微生物SNV丰度的生物标志物显示出较强的诊断能力(cc.AUC=0.87)比物种水平的生物标志物(cc。AUC=0.78)。最终,多模式生物标志物的整合显示了准确识别不同宫颈状态的最佳性能(cc.AUC=0.86),在外部测试集中具有可接受的性能(AUC=0.79)。
结论:阴道微生物组表现出与CC进展相关的特定SNV进化,作为区分不同宫颈疾病状态的特异性生物标志物。
BACKGROUND: Exploration of adaptive evolutionary changes at the genetic level in vaginal microbial communities during different stages of cervical cancer remains limited. This study aimed to elucidate the mutational profile of the vaginal microbiota throughout the progression of cervical disease and subsequently establish diagnostic models.
METHODS: This study utilized a metagenomic dataset consisting of 151 subjects classified into four categories: invasive cervical cancer (CC) (n = 42), cervical intraepithelial neoplasia (CIN) (n = 43), HPV-infected (HPVi) patients without cervical lesions (n = 34), and healthy controls (n = 32). The analysis focused on changes in microbiome abundance and extracted information on genetic variation. Consequently, comprehensive multimodal microbial signatures associated with CC, encompassing taxonomic alterations, mutation signatures, and enriched metabolic functional pathways, were identified. Diagnostic models for predicting CC were established considering gene characteristics based on single nucleotide variants (SNVs).
RESULTS: In this study, we screened and analyzed the abundances of 18 key microbial strains during CC progression. Additionally, 71,6358 non-redundant mutations were identified, predominantly consisting of SNVs that were further annotated into 25,773 genes. Altered abundances of SNVs and mutation types were observed across the four groups. Specifically, there were 9847 SNVs in the HPV-infected group and 14,892 in the CC group. Furthermore, two distinct mutation signatures corresponding to the benign and malignant groups were identified. The enriched metabolic pathways showed limited similarity with only two overlapping pathways among the four groups. HPVi patients exhibited active nucleotide biosynthesis, whereas patients with CC demonstrated a significantly higher abundance of signaling and cellular-associated protein families. In contrast, healthy controls showed a distinct enrichment in sugar metabolism. Moreover, biomarkers based on microbial SNV abundance displayed stronger diagnostic capability (cc.AUC = 0.87) than the species-level biomarkers (cc.AUC = 0.78). Ultimately, the integration of multimodal biomarkers demonstrated optimal performance for accurately identifying different cervical statuses (cc.AUC = 0.86), with an acceptable performance (AUC = 0.79) in the external testing set.
CONCLUSIONS: The vaginal microbiome exhibits specific SNV evolution in conjunction with the progression of CC, and serves as a specific biomarker for distinguishing between different statuses of cervical disease.