Abstract 570: High-resolution microbiome profiling of 16SrRNA sequencing data, Microbiome Arrays, and expression arrays identifies differential bacterial communities in head and neck cancer patients treated with surgery, chemo-radiation, and PD-1 checkpoint blockade therapy

Abstract

Abstract We have previously shown altered microbiota in saliva from head and neck squamous cell carcinoma (HNSCC) patients, when compared to controls. We now applied high-resolution microbiome profiling (Resphera Insight) to analyze 16S rRNA sequencing and Axiom Microbiome Array (Affymetrix) data in tissue samples from HNSCC patients and healthy controls. DNA from HNSCC (n=43) and controls (n=12) tissue samples was processed for 16SrRNA sequencing and microarray analyses. Raw sequences were processed for quality and length, screened for chimeras and filtered for contaminant human and chloroplast DNA. High-quality passing sequences were submitted to for species-level taxonomic assignments, followed by differential abundance analysis. Compositional arrangement was obtained with the R microbiome package, using a 25% compositional filtration of differentially significant Resphera OTUs, obtained with the Negative Binomial Wald test (p value <0.05). DESeq2 Variance Stabilizing Transformation algorithm was used for normalization after samples underwent the 25% prevalence threshold filtration. We found five assigned phyla dominating across the samples after 16S rRNA sequencing: Firmicutes, Proteobacteria, Bacteroidetes, Fusobacteria and Actinobacteria. We found three assigned phyla dominating across the samples hybridized to the microarray: Enterobacteria, Fusobacteria, and Prevotella. Alpha Rarefaction curves using the Chao1 index revealed that species richness did not differ between tumor and normal oral mucosa. However, there were differences by anatomic site and tumor staging when comparing HNSCC tissue samples. Differential abundance analysis between tumors and normals at the species level using the negative binomial test revealed that Lactobacillus salivarius, Lactobacillus gasseri, Lactobacillus johnsonii and Prevotella nanceiencis had the highest tumor/normal ratio (FDR-adjusted p values <0.05). We also found that Fusobacterium nucleatum (F.nucleatum), an oral cavity flora commensal bacterium linked to colon cancer, is enriched 51x higher in tissue in a subset of HNSCC patients with advanced tumors (T3 or above). F. nucleatum was detected in samples obtained before and after treatment with chemo-radiation, but not with surgery alone. Interestingly, we identified upregulation of the oncogenic Wnt/beta catenin pathway (Wnt7B, FZD6, SFRP4) and downregulation of immune system pathways (TLR10, IRF8) with genome-wide mRNA arrays (Affymetrix) in HNSCC tissue samples enriched for F. nucleatum. Together these data suggest the existence of complex interactions between the tissue microbiome in oral cavity mucosa and HNSCC tumor tissue, which may be modulating. Citation Format: Rafael E. Guerrero-Preston, Kelvin Navarro, Filipa Godoy-Vitorino, James White, Anne Jedlicka, Gustavo Rivera-Alvarez, Barbara Mora-Lagos, Winston Timp, William Westra, Wayne Koch, Luigi Marchionni, Young Kim, David Sidransky. High-resolution microbiome profiling of 16SrRNA sequencing data, Microbiome Arrays, and expression arrays identifies differential bacterial communities in head and neck cancer patients treated with surgery, chemo-radiation, and PD-1 checkpoint blockade therapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2018; 2018 Apr 14-18; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2018;78(13 Suppl):Abstract nr 570.