Given the increasing evidence that the oral microbiome is involved in obesity, diabetes and cancer risk, we investigated baseline oral microbiota profiles in relation to all-cancer incidence among non-smoking women enrolled in a Texas cohort of first- and second-generation immigrants of Mexican origin. We characterized the 16Sv4 rDNA microbiome in oral mouthwash samples collected at baseline from a representative subset of 305 non-smoking women, aged 20-75 years. We evaluated within (alpha) and between sample (beta) diversity by incident cancer status and applied Linear Discriminant Analysis (LDA) Effect Size analysis to assess differentially abundant taxa. Diversity and candidate taxa in relation to all-cancer incidence was evaluated in multivariable (MV)-adjusted Cox regression models. Over 8.8 median years of follow-up, 31 incident cancer cases were identified and verified. Advanced age, greater acculturation and cardiometabolic risk factors were associated with all-cancer incidence. Higher alpha diversity (age-adjusted Pdiff<0.01) and distinct biological communities (Pdiff=0.002) were observed by incident cancer status. Each unit increase in the Shannon diversity index yielded >8-fold increase in all-cancer and obesity-related cancer risk [MV-adjusted HR (95% CI): 8.11 (3.14, 20.94) and 10.72 (3.30, 34.84), respectively] with similar findings for the inverse Simpson index. Streptococcus was enriched among women who did not develop cancer, while Fusobacterium, Prevotella, Mogibacterium, Campylobacter, Lachnoanaerobaculum, Dialister and Atopobium were higher among women who developed cancer (LDA score≥3, q-value<0.01). This initial study of oral microbiota and overall cancer risk in non-smoking Mexican American women suggests the readily accessible oral microbiota as a promising biomarker.

中文翻译：

---

一组不吸烟的墨西哥裔美国女性的基线口腔微生物组和所有癌症发病率

鉴于越来越多的证据表明口腔微生物群与肥胖、糖尿病和癌症风险有关，我们调查了德克萨斯州第一代和第二代移民队列中非吸烟女性与所有癌症发病率相关的基线口腔微生物群谱。墨西哥原产地。我们在基线时从 305 名 20-75 岁的不吸烟女性中收集到的口腔漱口水样本中的 16Sv4 rDNA 微生物组进行了表征。我们通过事件癌症状态评估样本内（α）和样本间（β）多样性，并应用线性判别分析（LDA）效应大小分析来评估差异丰富的分类群。在多变量 (MV) 调整的 Cox 回归模型中评估了与所有癌症发病率相关的多样性和候选分类群。超过 8.8 年的中位随访时间，确定并核实了 31 起癌症事件。高龄、更高的文化适应和心脏代谢危险因素与所有癌症的发病率相关。通过癌症发病状态观察到更高的α多样性（年龄调整的Pdiff<0.01）和不同的生物群落（Pdiff=0.002）。香农多样性指数每增加一个单位，所有癌症和肥胖相关的癌症风险就会增加 8 倍以上 [MV 调整 HR (95% CI)：分别为 8.11 (3.14, 20.94) 和 10.72 (3.30, 34.84) ] 对逆辛普森指数有类似的发现。链球菌在未患癌症的女性中富集，而梭杆菌属、普氏菌属、莫吉杆菌属、弯曲杆菌属、厌氧杆菌属、Dialister 和 Atopobium 在患癌症的女性中较高（LDA 评分≥3，q 值<0.01）。