Therapy-Related Oral Toxicity Drug Targets: A Data Mining Analysis of Oral Bacteria Interaction with Human Molecular Networks

Cancer therapy-induced oral mucositis (CTOM) is a common side effect of cancer therapy. Our objective was to identify associations between bacteria and human molecular networks, including drug targets from therapy-related oral toxicities. ClinicalTrials.gov database was used to identify drugs tested for oral mucositis/stomatitis. Gene targets were retrieved from the Drug Gene Interaction database (DGIdbv5.0.13). Secondly, associations between human genes and bacterial species linked with oral mucositis were identified using the MSEAv0.0.2 program. Further, publicly available data from an oral mucositis 3D tissue model were used for differential gene expression (DGE) analysis of everolimus-treated cells targeting the IL6/MTOR pathway. Genes common to all analyses were examined using STRINGv12.0. Of 33 oral mucositis/stomatitis trials, we identified 38 drugs with 89 direct and 438 indirect gene targets. MSEAv0.0.2 identified 548 significant associations (Fisher’s exact test; padj < 0.05) involving 35 bacterial genera. DGE analysis identified 3664 significant genes in everolimus-treated cells (empirical Bayes moderated t-test; padj < 0.05). While two genes (i.e., CFTR and PTGS2) were common between the MSEA and DGE analyses and were direct drug targets identified in clinical trials, twenty-eight genes were in common when including indirect potential drug targets. When the two genes were combined with the twenty-eight genes, sixty-six significant Reactome pathways were identified (hypergeometric FDR-p < 0.05), potentially linked to inflammation and compensatory wound healing in oral mucositis.