Abstract 3692: Computational modeling of human microbiome evolution in the context of esophageal adenocarcinoma progression

Le résumé fourni par la source

Abstract Elucidating the role of microbes in precancer evolution could transform clinical risk prediction and management strategies. In the case of esophageal adenocarcinoma (EAC) progression, microbial communities have been detected across precancerous and neoplastic stages. However, unlike prior studies that statistically compare microbial abundances at single time-points, we employed mathematical modeling of microbial population dynamics within human hosts to determine likely modes of evolution that drive changes in microbial composition over time. We first aggregated whole genome sequencing (WGS) datasets from esophageal tissue samples taken across stages of EAC progression, which included healthy esophagus, gastroesophageal reflux disease (GERD), Barrett’s esophagus (BE), and EAC. In particular, we analyzed baseline WGS data from a case-control study of patients with BE who later progressed to EAC (“progressors”) versus those patients with BE who did not progress to cancer within 5 years (“non-progressors”). We employed a robust bioinformatics host filtration pipeline to remove human DNA reads, mapped remaining reads to a human-scrubbed Web of Life microbial database, and filtered out taxa with low genomic coverage. After extracting occurrence and abundance data for taxa in each disease stage, we applied mathematical models of both neutral and non-neutral microbial evolution. We found evidence of neutral dynamics across stages of normal to pre-cancer (goodness-of-fit to the neutral expectation R2 > 0.85 for normal esophagus and GERD, R2 > 0.65 for BE non-progressors and BE progressors) but not for EAC (R2 = 0.14 in EACs from the International Cancer Genome Consortium, and R2 = 0.34 in a validation set of 19 EACs who also had surrounding BE). We also found that Helicobacter pylori (H. pylori), a bacterium inversely associated with EAC risk, deviated significantly from the expected neutral model in BE non-progressor patients but not in BE progressor patients. To test hypotheses that accounted for potential selection effects, we performed simulations for both neutral and non-neutral processes. Using simulated data for the EAC cohorts, we found that assuming widely varying growth rates best reflected the data, implying that selection pressures likely influence largely niche-based population dynamics in the tumor microenvironment. For the BE non-progressor cohort, we found that assuming a lower death rate for H. pylori compared to other taxa recapitulated the data. This implies that H. pylori may have a selective advantage in non-progressing BE (as it does when present in the stomach), and further studies are needed to understand the impact on EAC progression. Overall, considering metagenomic data as the result of a dynamic process within a human host will enhance our understanding of the human microbiome's role in precancer and cancer evolution. Citation Format: Caitlin E. Guccione, Igor Sfiligoi, Antonio Gonzalez, Justin Shaffer, Mariya Kazachkova, Yuhan Weng, Daniel McDonald, Shailja Shah, Samuel S. Minot, Thomas Paulson, Ludmil Alexandrov, Rob Knight, Kit Curtius. Computational modeling of human microbiome evolution in the context of esophageal adenocarcinoma progression [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 3692.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.