UC San Diego

The cancer microbiome is an emerging research area exploring the relationship between the humanmicrobiome and cancer. Microbes are not static players in the precancer and tumor microenvironment - microorganisms coevolve with host cells and new methods are needed to quantify the composition and dynamics of microbial communities in these ecosystems. The overall goal of my thesis is to demonstrate how microbial communities emerge, persist, or shift throughout cancer initiation, progression, and metastasis.

Advancing this field requires improving the ability to accurately extract microbial reads andinterpret microbial signals within available whole genome sequencing data from large cancer cohorts. To address this, I developed a computational host filtration pipeline that incorporates a variety of human references including those in the Human Pangenome Reference Consortium. This pipeline was validated on various simulated and real human and microbial datasets. The user-friendly tool mitigates artifactual bias, enabling more sensitive host-microbe conclusions in cancer microbiome research.

Secondly, I applied the host filtration technique to esophageal tissue data to examine microbialchanges throughout stages of progression from normal to esophageal adenocarcinoma (EAC). Contrary to prior studies that focused on microbial abundances across single time points, we employed mathematical modeling of microbial population dynamics to understand underlying evolutionary assumptions that explained changes in microbial composition over time. We found that Helicobacter pylori may play a protective role against EAC progression in patients with precancerous Barrett’s esophagus, and that selection pressures likely influence largely niche-based population dynamics in the tumor microenvironment.

Finally, we explored microbial changes across primary and metastatic liver samples from patientswith hepatocellular carcinoma (HCC) and metastatic colorectal cancer (mCRC) to analyze and characterize their tumor microbiomes. We identified unique microbiota associated with HCC and mCRC, which may act as biomarkers for cancer diagnosis, prognosis, and progression. The identification of distinct microbial signatures in the blood opens new possibilities for developing microbiome-driven interventions to reshape the tumor microenvironment.

My thesis establishes a foundational framework for cancer microbiome research, enabling moreprecise microbial analyses and paving the way for future studies to uncover microbiome-based biomarkers, therapeutic targets, and novel insights into cancer progression and treatment response.