Abstract Introduction: The oral microbiome is intimately involved in inflammatory processes in the oral cavity and may be involved in metabolism of carcinogens in tobacco smoke. Our hypothesis is that oral microbiota may promote head and neck carcinogenesis, possibly related to smoking. To test this hypothesis, we conducted a nested case-control study to prospectively examine the role of the oral microbiome in head and neck cancer, in the Prostate, Lung, Colorectal, and Ovarian (PLCO) Cancer Screening Trial cohort. Methods: Subjects in the PLCO Trial completed risk factor questionnaires and provided oral wash samples. Trial participants were followed for cancer incidence by annual follow-up, medical chart review and search of mortality databases. During follow-up, 87 histologically confirmed head and neck cancer cases occurred subsequent to oral wash sample collection. We matched these cases to 174 cohort-nested controls (matched by age, gender, and race). DNA was extracted from oral wash samples, 16S rRNA genes were amplified by universal primers, and product was bar-coded and sequenced by 454 FLX technology. 16S rRNA bacterial gene sequences were binned into operational taxonomic units (based on 97% identity) and further assigned to taxonomy in the QIIME pipeline. We performed nonparametric Wilcoxon signed-rank tests and conditional logistic regression to compare differences between head and neck cancer cases and controls, with respect to overall bacterial community structure and relative abundance of specific taxa. Results: From 261 oral samples, we obtained 2,548,067 high quality 16S rRNA gene sequence reads. Overall community structure, based on a weighted UniFrac phylogenetic distance index, did not differ significantly between cases and controls (p>0.05). The major phyla in controls were Firmicutes (59.5%), Actinobacteria (15.5%), Bacteroidetes (11.3%), Proteobacteria (10.3%), and Fusobacteria (2.8%). Decreased abundance of Proteobacteria was observed in cases (7.8%) compared to controls (Wilcoxon test, p = 0.008, FDR controlled p=0.041). After consideration of age, race and sex in conditional logistic regression analysis, an inverse association of Proteobacteria with head and neck cancer risk was observed in smokers (OR comparing extreme tertiles =0.34, 95% CI 0.15-0.79; ptrend = 0.009 [FDR controlled p=0.065]), but not in non-smokers (ptrend=0.62). We also noted a relative increase in abundance of Bifidobacteriales order (multivariate p=0.002, FDR controlled p=0.045) and Lactobacillaceae family (multivariate p=0.005, FDR controlled p>0.05) in cases compared with controls. Conclusion: This first prospective study of the oral microbiome and head and neck cancer suggests that Proteobacteria and Bifidobacteriales are related to risk of head and neck cancer. Replication of these first observations is needed. Citation Format: Jiyoung Ahn, Yingfei Ma, Mark P. Purdue, Neal D. Freedman, Susan M. Gapstur, Liying Yang, Richard B. Hayes, Zhiheng Pei. Oral microbiome and risk of head and neck cancer, a nested case-control study. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 4143. doi:10.1158/1538-7445.AM2014-4143