Characterization of the intratumoral microbiota in non endemic EBV associated nasopharyngeal carcinoma

Epstein-Barr virus (EBV) related nasopharyngeal carcinoma (NPC) is an epithelial malignancy arising from the nasopharyngeal mucosal lining. A high incidence of EBV related NPC is recorded in endemic areas (EA), East and Southeast Asia. Microbiota resides the upper respiratory tract including nasopharynx. Thus, intratumoral microbiota (ITM) in EA EBV related NPC has been characterized with Corynebacterium and Staphylococcus being the most predominant taxa. Currently, ITM profile in NPC remain unexplored in non endemic area (NEA). We hypothesized that the ITM in NEA NPC is associated to tumor biology. In our previously NEA EBV related NPC study, we stratified patients into three transcriptional clusters (Cl1: immune active; Cl2: defense response; Cl3: proliferative), with Cl3 displaying the poorest prognosis. Hence, the present work investigates the ITM composition according to our clusters' model. Formalin-fixed paraffin embedded (FFPE) samples from 94 NEA NPC EBV related cases were collected. Total RNA was extracted using miRneasy kit (Qiagen) and libraries were generated using QuantSeq 3' mRNA (Lexogen) and sequenced on the NextSeq500 (Illumina). The ITM composition and immune cell deconvolution (xCell) were inferred from RNAseq data. Differential abundance analyses (LEfSe) among clusters and correlation analyses between immune cells populations and differentially abundant taxa were conducted. To validate the RNAseq findings and assess spatial colocalization patterns, spatial and metatranscriptomics (SMT) (Stereoseq OMNI for FFPE, STOmics) was performed on one representative case per cluster. 297 bacterial taxa were detected in our NEA cohort. LEfSe comparisons showed that Bacteroidota and Pseudomonadota were enriched in Cl1 vs Cl2, while Bacillaceae dominated in Cl2. Actinomycetota was more prevalent in Cl1 vs Cl3. Correlation analyses showed that most differentially abundant taxa, found in clusters' comparison, were positively associated with fibroblast and stroma score and inversely correlated with CD8+ and CD4+ T cell populations. SMT profiling confirmed intratumoral bacteria expression: Cl1 and Cl2 exhibited higher bacterial proportion in respect to EBV (the latter expressed in 14.6% and 23.6%, respectively), while Cl3 displayed both bacterial and EBV signals, with a predominant EBV expression (52.5%). SMT analyses is ongoing to disclose co expression patterns between immune cells, intratumoral bacteria and EBV. This study provides the first characterization of the ITM in NEA EBV related NPC. Higher EBV expression was detected in Cl3, the subgroup associated with the poorest prognosis, suggesting a possible synergistic interplay between EBV and specific bacterial taxa in the NPC pathogenesis. These findings highlight the impact of the microbiome on the tumor microenvironment and support its potential as biomarker in NEA EBV related NPC.