Divergent Microbial Profiles in Tumor and Adjacent Normal Tissue across Cancer Types
Divergent Microbial Profiles in Tumor and Adjacent Normal Tissue across Cancer Types
| dc.contributor.advisor | Hernandez, Brenda Y. | |
| dc.contributor.advisor | Deng, Youping | |
| dc.contributor.author | Rodriguez, Rebecca Maria | |
| dc.contributor.department | Biomedical Sciences | |
| dc.date.accessioned | 2019-07-02T17:42:33Z | |
| dc.date.issued | 2019 | |
| dc.description.degree | Ph.D. | |
| dc.embargo.liftdate | 2020-07-01 | |
| dc.identifier.uri | http://hdl.handle.net/10125/63103 | |
| dc.subject | Epidemiology | |
| dc.subject | Bioinformatics | |
| dc.subject | bacteria in cancer | |
| dc.subject | computational subtraction | |
| dc.subject | racial differences | |
| dc.subject | TCGA | |
| dc.subject | tumor tissue microbiome | |
| dc.subject | whole exome sequencing | |
| dc.title | Divergent Microbial Profiles in Tumor and Adjacent Normal Tissue across Cancer Types | |
| dc.type | Thesis | |
| dcterms.abstract | Background: There is growing evidence that microbial variation can influence cancer development, progression, response to therapy, and outcomes. We wanted to examine the microbial composition of paired tumor and adjacent normal tissue across various cancer types in order to provide an improved understanding of microbial diversity and abundance patterns of the tumor microenvironment and their influence on clinical presentation and survival. Methods: Using raw whole exome sequencing data from 22 cancer types from the The Cancer Genome Atlas (TCGA) network, we examined differential relative abundance and diversity data in primary tumor and adjacent solid tissue normal (adjacent normal), nine of which are presented in this work. Data were processed through a bioinformatics pipeline designed to extract microbial profiles from human sequencing data based on PathoScope 2.0. Differential abundance and diversity metrics were calculated using R-package tools to compare primary tumor and adjacent normal within cancer types and across cancer cohorts correlating to clinical features including histologic and pathologic features and survival data. Analyses were controlled for demographic, exposures and batch effects. Findings were then validated by qPCR for selected cancer types with tissue from the Hawaii Tumor Registry RTR. Results: As part of a pilot project we have created microbial composition and diversity profiles for a subset of solid tumors within TCGA cancers building a platform for ongoing and future studies. We screened over 10,000 files encompassing a total of 1,838 paired cases from which 813 are discussed here. From those, 767 tumors and 753 adjacent normal samples had positive microbial (viral and bacterial) sequence reads detection. Microbial composition and diversity (richness, within sample alpha diversity, evenness and beta diversity) varied across cohorts with similar patterns at the phylum level in compositional structures. Bacterial shifts were evident in tumor compared to adjacent normal. Proteobacteria phyla was observed to be increased in tumors of all cohorts except for STAD, where Proteobacteria species were reduced and Firmicutes levels increased. Differences between patient samples were evident at the higher taxonomic levels. Differential abundance analyses revealed significant differences in stomach adenocarcinoma (STAD) and colon adenocarcinoma (COAD). Compared to adjacent normal, tumor samples were found to have lower number of species present overall and lower diversity indices. We found significant association between microbial relative abundance and diversity to clinicopathological presentation and survival dependent on race in some cancers, particularly those of infectious origin like STAD and LIHC. Conclusion. This project demonstrates the feasibility of the utilization of exome sequencing data to derive complex microbial data with easy to interpret results. This project facilitates the understanding of the role of bacteria play in cancer pathogenesis across different race groups as demonstrated in LIHC and STAD cancer cohorts. In these cancers, relative abundance was associated with tumor stage and overall survival days and within sample diversity was associated with race in fully adjusted models. | |
| dcterms.description | Ph.D. Thesis. University of Hawaiʻi at Mānoa 2019 | |
| dcterms.extent | 177 pages | |
| dcterms.language | eng | |
| dcterms.publisher | University of Hawai'i at Manoa | |
| dcterms.rights | All UHM dissertations and theses are protected by copyright. They may be viewed from this source for any purpose, but reproduction or distribution in any format is prohibited without written permission from the copyright owner. | |
| dcterms.type | Text | |
| local.identifier.alturi | http://dissertations.umi.com/hawii:10150 |
Files
Original bundle
1 - 1 of 1
No Thumbnail Available
- Name:
- Rodriguez_hawii_0085A_10150.pdf
- Size:
- 5.07 MB
- Format:
- Adobe Portable Document Format
- Description: