ASPO Abstracts
Treatment response and associations with the oral microbiome in treatment naive patients with oropharyngeal cancer
Category: Molecular Epidemiology & Environment
Conference Year: 2019
Abstract Body:
Purpose: Oropharyngeal cancer (OPC) incidence is rising dramatically in the US. Oral microbiota (communities of microorganisms residing in the oral cavity) may influence cancer treatment-related toxicities. We examined the composition and diversity of oral microbiota in OPC patients prior to treatment, and identify associations between oral microbiota and clinical response. Methods: 60 newly diagnosed, treatment naïve, OPC patients were recruited from Moffitt Cancer Center (2015-2017). Patients provided mouthwash-based oral gargles before starting chemoradiation or radiation. DNA was isolated using the QIAGEN DNeasy PowerSoil kit, and 16S rRNA gene sequencing (V1-V3) was performed. An operational taxonomic unit table was generated and Ribosomal Database Project Classifier used to assign taxonomy. Clinical response 3mo after treatment cessation was abstracted from medical records. Clinical responders (cR) experienced complete/partial response, while non-responders (cNR) experienced stable/progressive disease. Comparisons of bacterial relative abundance (Mann Whitney U), alpha diversity (Chao1, Shannon, Simpson), and beta diversity (Bray-Curtis) were performed in R and Phyloseq. Results: Of 60 OPC patients, 65% were stage IV, 48% tonsillar and 42% base of tongue, and 40% never smokers. Pre-treatment levels of phylum Cyanobacteria appeared to be more abundant in the oral cavity of cR (5%) vs. cNR (0.1%),. At the genus level, cR appeared to have a lower abundance of Neisseria (5%) vs. cNR (10%). Anaeroglobus and Propionibacterium genera were significantly more abundant in cR vs. cNR (p<0.04), and Shuttleworthia was significantly less abundant in cR (p=0.03), although these were rare (<1% each). Oral bacterial alpha diversity (intra-subject) appeared to be lower in cR vs. cNR. Analyses of beta diversity (inter-subject) showed that the microbial community structures of cNR appeared to be more similar to one another than cR. Conclusions: Oral microbiome profiling may hold promise as a prognostic biomarker in OPC. Several potentially predictive taxa were identified to be differentially abundant in cR vs. cNR. Future analyses will evaluate the role of oral microbial resilience (the rate of recovery after disturbance) using specimens collected 3mo post-treatment.
Keywords: Microbiome, oropharyngeal cancer, clinical response