Untangling the links between obesity and cancer using genomic and phenomic approaches
<p>The pandemics of obesity, cancer and Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) are major global healthcare challenges of this generation. This thesis investigates the interplay between anthropometric measurements, malignancy and, unanticipatedly, SARS-COV-2 infection.<...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
2021
|
| Subjects: |
| _version_ | 1811141102753808384 |
|---|---|
| author | Harris, B |
| author2 | Buffa, F |
| author_facet | Buffa, F Harris, B |
| author_sort | Harris, B |
| collection | OXFORD |
| description | <p>The pandemics of obesity, cancer and Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) are major global healthcare challenges of this generation. This thesis investigates the interplay between anthropometric measurements, malignancy and, unanticipatedly, SARS-COV-2 infection.</p>
<p>Obesity-related cancers account for 40% of the cancer cases observed in the US, and obesity is set to overtake smoking as the most widespread modifiable risk factor for carcinogenesis. My initial literature review uses as a framework the Hallmarks of Cancer as defined by Hanahan and Weinberg. Each Hallmark is shown to be impacted by the physiological state of obesity, particularly through immunomodulation and facilitation of various elements of the metastatic pathway. Most work in this area was carried out in model organisms and cell lines. There are few studies on large prospective human cohorts, but large datasets are now becoming available and one is explored in depth in this thesis.</p>
<p>Previous human studies have shown a reliance on body mass index (BMI) as a surrogate for obesity. Studies in other fields have highlighted how overreliance on BMI can lead to misinterpretation. Now, the opportunity exists to dissect the factors behind BMI links with cancer, looking independently at whole body fat mass (WBFM) and whole body fat-free mass (WBFFM) more directly using bioelectrical impedance in a cohort of 500,000 individuals (the UK Biobank). Further, genomics approaches are used to extract signal from noise in large datasets, but as yet these are not commonly used in epidemiological studies. One such approach, principal component analysis (PCA), is employed here to tease apart the signal coming from the weight component of BMI and its association with cancer.</p>
<p>This work focusses on three cancers, breast (post-menopausal), prostate and colorectal cancers, all of which have been previously shown to associate with obesity. Increasing quintiles of BMI, WBFM and WBFFM are all associated with an increased risk of post-menopausal breast cancer, with the highest quintile of WBFFM giving the highest odds ratio of the three (OR=1.60). PCA suggests that both increased WBFM and WBFFM may both confer increased risk of post menopausal breast cancer. By contrast, obesity is associated with a reduced risk of prostate cancer and the protective effect probably originates from WBFM. In colorectal cancer in men, increased BMI, WBFM and WBFFM all carry increased risk, although there is no obvious association in women. The merit of taking measures of fat mass and fat-free mass when investigating cancer risk is highlighted, raising interesting questions around the roles of adiposity and fat-free mass in relation to cancer. Perhaps genetic pleiotropy might help explain these results.</p>
<p>Anthropometric measures are consistently found to associate with differential cancer risk. Inherited genetic variants in tumour suppressors and proto-oncogenes also increase cancer risk. A further focus of this thesis is to investigate whether/how these two associations might be functionally linked. Through a novel approach, integrating inherited and somatic cancer genetic data, single nucleotide polymorphisms were identified which associate with different cancer types and differential gene expression in at least one tissue type (eSNPs). Potential associations of these SNPs with anthropometric traits and cancers were then explored.</p>
<p>A set of eSNPs related to three important cancer genes, FANCA, MAP3K1 and TP53, were identified as associating with both anthropometric measures and cancer risk. Notably, a previously unrecognised strong association was identified between the rs78378222[C] SNP in the 3'-UTR of TP53 and increased risk of developing non-melanomatous skin cancer (OR=1.36), brain malignancy (OR=3.12) together with increased standing height, fat-free mass and basal metabolic rate. This offers a novel genetic link between these anthropometric traits and cancer risk.</p>
<p>This eSNP analysis focused on a small subset of the genome leaving a large proportion unexplored. Therefore, this was extended using genome-wide association studies (GWAS). These identified additional pleiotropic haplotypes showing differential association with BMI, WBFM and WBFFM and the three cancers explored in the epidemiological study (post-menopausal breast cancer, prostate cancer and colorectal cancer). Twenty-two haplotypes associated with prostate cancer and anthropometric traits. Five haplotypes exhibited positive effects on both (i.e. increased prostate cancer risk and increased birth weight). eQTL analysis yielded SNPs within these haplotypes which influence expression of CD81, C11orf21, ASCL2, TH and IGF2. Eighteen haplotypes associated with both post-menopausal breast cancer and anthropometric measures. Three of these haplotypes were positively associated with post-menopausal breast cancer risk and anthropometric traits, and two of these also positively associated with measures of WBFFM. One SNP within these haplotypes, rs537626[C], which falls within the lncRNA CUPID1, has been validated as associated with breast cancer elsewhere. CUPID1 is a regulator of DNA damage repair and rs537626[C] may influence its function.</p>
<p>The SARS-COV-2 pandemic influenced the needs of the world and indeed this work. Redirection of resources and safety of the nation became paramount. I was amongst the first researchers to describe longitudinal immunological responses to SARS-COV-2 infections in two important patient groups (asymptomatic patients and those with Post-COVID-19 Syndrome). More directly related to the body of this thesis, as cancer and obesity both emerged as important risk factors for SARS-COV-2 outcome, common genetic roots became an area of curiosity. Interestingly, two SNPs associated with poor SARS-CoV-2 outcome and WBFFM (rs1265097[A] and rs3130279[A).</p>
<p>This thesis uses a variety of bioinformatics techniques to investigate links between anthropometric measures and cancer, adding several genetic loci to the literature worthy of further investigation.</p> |
| first_indexed | 2024-03-07T04:44:40Z |
| format | Thesis |
| id | oxford-uuid:d2d48fbd-a9d0-40b3-a895-63a2d6f594ea |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-09-25T04:32:33Z |
| publishDate | 2021 |
| record_format | dspace |
| spelling | oxford-uuid:d2d48fbd-a9d0-40b3-a895-63a2d6f594ea2024-09-02T08:40:54ZUntangling the links between obesity and cancer using genomic and phenomic approachesThesishttp://purl.org/coar/resource_type/c_db06uuid:d2d48fbd-a9d0-40b3-a895-63a2d6f594eacancerEnglishHyrax Deposit2021Harris, BBuffa, FLord, SHeather, L<p>The pandemics of obesity, cancer and Severe Acute Respiratory Syndrome Coronavirus 2 (SARS-COV-2) are major global healthcare challenges of this generation. This thesis investigates the interplay between anthropometric measurements, malignancy and, unanticipatedly, SARS-COV-2 infection.</p> <p>Obesity-related cancers account for 40% of the cancer cases observed in the US, and obesity is set to overtake smoking as the most widespread modifiable risk factor for carcinogenesis. My initial literature review uses as a framework the Hallmarks of Cancer as defined by Hanahan and Weinberg. Each Hallmark is shown to be impacted by the physiological state of obesity, particularly through immunomodulation and facilitation of various elements of the metastatic pathway. Most work in this area was carried out in model organisms and cell lines. There are few studies on large prospective human cohorts, but large datasets are now becoming available and one is explored in depth in this thesis.</p> <p>Previous human studies have shown a reliance on body mass index (BMI) as a surrogate for obesity. Studies in other fields have highlighted how overreliance on BMI can lead to misinterpretation. Now, the opportunity exists to dissect the factors behind BMI links with cancer, looking independently at whole body fat mass (WBFM) and whole body fat-free mass (WBFFM) more directly using bioelectrical impedance in a cohort of 500,000 individuals (the UK Biobank). Further, genomics approaches are used to extract signal from noise in large datasets, but as yet these are not commonly used in epidemiological studies. One such approach, principal component analysis (PCA), is employed here to tease apart the signal coming from the weight component of BMI and its association with cancer.</p> <p>This work focusses on three cancers, breast (post-menopausal), prostate and colorectal cancers, all of which have been previously shown to associate with obesity. Increasing quintiles of BMI, WBFM and WBFFM are all associated with an increased risk of post-menopausal breast cancer, with the highest quintile of WBFFM giving the highest odds ratio of the three (OR=1.60). PCA suggests that both increased WBFM and WBFFM may both confer increased risk of post menopausal breast cancer. By contrast, obesity is associated with a reduced risk of prostate cancer and the protective effect probably originates from WBFM. In colorectal cancer in men, increased BMI, WBFM and WBFFM all carry increased risk, although there is no obvious association in women. The merit of taking measures of fat mass and fat-free mass when investigating cancer risk is highlighted, raising interesting questions around the roles of adiposity and fat-free mass in relation to cancer. Perhaps genetic pleiotropy might help explain these results.</p> <p>Anthropometric measures are consistently found to associate with differential cancer risk. Inherited genetic variants in tumour suppressors and proto-oncogenes also increase cancer risk. A further focus of this thesis is to investigate whether/how these two associations might be functionally linked. Through a novel approach, integrating inherited and somatic cancer genetic data, single nucleotide polymorphisms were identified which associate with different cancer types and differential gene expression in at least one tissue type (eSNPs). Potential associations of these SNPs with anthropometric traits and cancers were then explored.</p> <p>A set of eSNPs related to three important cancer genes, FANCA, MAP3K1 and TP53, were identified as associating with both anthropometric measures and cancer risk. Notably, a previously unrecognised strong association was identified between the rs78378222[C] SNP in the 3'-UTR of TP53 and increased risk of developing non-melanomatous skin cancer (OR=1.36), brain malignancy (OR=3.12) together with increased standing height, fat-free mass and basal metabolic rate. This offers a novel genetic link between these anthropometric traits and cancer risk.</p> <p>This eSNP analysis focused on a small subset of the genome leaving a large proportion unexplored. Therefore, this was extended using genome-wide association studies (GWAS). These identified additional pleiotropic haplotypes showing differential association with BMI, WBFM and WBFFM and the three cancers explored in the epidemiological study (post-menopausal breast cancer, prostate cancer and colorectal cancer). Twenty-two haplotypes associated with prostate cancer and anthropometric traits. Five haplotypes exhibited positive effects on both (i.e. increased prostate cancer risk and increased birth weight). eQTL analysis yielded SNPs within these haplotypes which influence expression of CD81, C11orf21, ASCL2, TH and IGF2. Eighteen haplotypes associated with both post-menopausal breast cancer and anthropometric measures. Three of these haplotypes were positively associated with post-menopausal breast cancer risk and anthropometric traits, and two of these also positively associated with measures of WBFFM. One SNP within these haplotypes, rs537626[C], which falls within the lncRNA CUPID1, has been validated as associated with breast cancer elsewhere. CUPID1 is a regulator of DNA damage repair and rs537626[C] may influence its function.</p> <p>The SARS-COV-2 pandemic influenced the needs of the world and indeed this work. Redirection of resources and safety of the nation became paramount. I was amongst the first researchers to describe longitudinal immunological responses to SARS-COV-2 infections in two important patient groups (asymptomatic patients and those with Post-COVID-19 Syndrome). More directly related to the body of this thesis, as cancer and obesity both emerged as important risk factors for SARS-COV-2 outcome, common genetic roots became an area of curiosity. Interestingly, two SNPs associated with poor SARS-CoV-2 outcome and WBFFM (rs1265097[A] and rs3130279[A).</p> <p>This thesis uses a variety of bioinformatics techniques to investigate links between anthropometric measures and cancer, adding several genetic loci to the literature worthy of further investigation.</p> |
| spellingShingle | cancer Harris, B Untangling the links between obesity and cancer using genomic and phenomic approaches |
| title | Untangling the links between obesity and cancer using genomic and phenomic approaches |
| title_full | Untangling the links between obesity and cancer using genomic and phenomic approaches |
| title_fullStr | Untangling the links between obesity and cancer using genomic and phenomic approaches |
| title_full_unstemmed | Untangling the links between obesity and cancer using genomic and phenomic approaches |
| title_short | Untangling the links between obesity and cancer using genomic and phenomic approaches |
| title_sort | untangling the links between obesity and cancer using genomic and phenomic approaches |
| topic | cancer |
| work_keys_str_mv | AT harrisb untanglingthelinksbetweenobesityandcancerusinggenomicandphenomicapproaches |