PO.TB10.14 · 肿瘤生物学
以口腔病原体为特征的失调性多菌群特征将口腔-胃轴与胃癌发生联系起来——一项智利高危队列研究
A dysbiotic polymicrobial signature characterized by oral pathogens links the oral-gastric axis to gastric carcinogenesis in a high-risk Chilean cohort
该海报暂无可下载的资料
AACR 官方页面
作者与单位 Authors & Affiliations
摘要 Abstract
中文摘要
目的：胃癌（GC）仍是一项严峻的全球健康挑战，尤其是在智利等高危地区。以Helicobacter pylori为中心的传统病因模型不足以解释疾病的全部谱系。本研究旨在鉴定GC组织中一种稳健的多菌群特征，并检验以下假设：口腔-胃轴通过易位的口腔病原体的慢性定植驱动致癌过程。
方法：我们采用高通量16S rRNA扩增子测序分析一个智利队列（15例GC患者[CAN]，15例健康对照[NOC]）组织样本中的胃部微生物组。使用非参数检验（Mann-Whitney U）评估微生物组成、alpha多样性（丰富度、均匀度）和差异丰度，并通过监督式机器学习方法（包括随机森林和惩罚回归模型[Lasso/Ridge]）进行统计学验证，以确保稳健的生物标志物筛选。
结果：GC微生物组表现出深刻的失调性转变，其特征为物种丰富度显著增加（Observed ASVs、ACE、Chao1；p<0.001），但生态均匀度明显失衡。一个由四个属组成的核心特征始终能将CAN与NOC组织区分开来：Fusobacterium（MWU p=0.024）、Lactobacillus（MWU p=0.038）、Neisseria（MWU p=0.005）和Prevotella（MWU p=0.023）。口腔病原体Neisseria（随机森林排名第1）和Fusobacterium（在CAN中富集）被鉴定为最具预测性的特征。机制上的见解提示Fusobacterium通过促进表达PD-L1的肿瘤相关中性粒细胞的募集而促成免疫逃逸，从而将这一特征直接与免疫治疗靶点联系起来。
结论：胃癌与一种独特的、病理上不稳定的、由已知口腔病原体主导的多菌群特征相关。这一稳健的特征提供了有力的定量证据，支持口腔-胃轴假说。所鉴定的微生物群落为新型无创诊断生物标志物提供了有前景的候选者，并提示了新的治疗策略，例如靶向清除促肿瘤发生的口腔微生物，以使胃癌肿瘤对免疫检查点阻断治疗敏感。
查看英文原文 English abstract
Objectives:Gastric cancer (GC) remains a formidable global health challenge, particularly in high-risk regions like Chile. The traditional etiological model centered on Helicobacter pylori is insufficient to explain the full spectrum of disease. This study aimed to identify a robust, polymicrobial signature in GC tissue and test the hypothesis that the Oral-Gastric Axis drives the carcinogenic process via chronic colonization by translocated oral pathogens.
Methods: We employed high-throughput 16S rRNA amplicon sequencing to profile the gastric microbiome in tissue samples from a Chilean cohort (15 GC patients [CAN], 15 healthy controls [NOC]). Microbial composition, alpha diversity (richness, evenness), and differential abundance were assessed using non-parametric tests (Mann-Whitney U) and statistically validated through supervised machine learning approaches, including Random Forest and penalized regression models (Lasso/Ridge), to ensure robust biomarker selection.
ResultsThe GC microbiome showed a profound dysbiotic shift, characterized by a significant increase in species richness (Observed ASVs, ACE, Chao1; p<0.001) but marked ecological unevenness. A core four-genus signature consistently differentiated CAN from NOC tissue: Fusobacterium (MWU p=0.024), Lactobacillus (MWU p=0.038), Neisseria (MWU p=0.005), and Prevotella (MWU p=0.023). The oral pathogens Neisseria (Random Forest Rank 1) and Fusobacterium (enriched in CAN) were identified as the top predictive features. Mechanistic insights suggest Fusobacterium contributes to immune evasion by promoting the recruitment of PD-L1-expressing tumor-associated neutrophils, linking this signature directly to immunotherapy targets.
Conclusions: Gastric cancer is associated with a distinct, pathologically unstable polymicrobial signature dominated by known oral pathogens. This robust signature provides compelling quantitative evidence supporting the Oral-Gastric Axis Hypothesis. The identified microbial community offers promising candidates for novel non-invasive diagnostic biomarkers and suggests new therapeutic strategies, such as targeted depletion of pro-tumorigenic oral microbes to sensitize tumors to immune checkpoint blockade therapy in gastric cancer.
利益披露 Disclosure
J. Figueroa, None.
M. Garrido,
MSD ).
BMS ).
Novartis ).
Roche ).
Astellas ).
Deciphera ).
PPD ).
IQVIA ).
Bayer ).
Principia ).
Covance ).
Daiichi-Sankyo ).
Basilea ).
PRA-Exelisis ).
Syneos ).
Zimeworks ).
A. Riquelme, None.
M. Munoz Medel,
Celerity Clinical Research Employment.
F. Villaroel-Espindola, None..
F. Gomez-Valenzuela, None..
I. Retamal, None.