Molecular epidemiology of Prostate Cancer progression

Student thesis: Doctoral ThesisDoctor of Philosophy (PhD)

Abstract

Localised Prostate Cancer (PCa) is a slow-growing and heterogenous disease, with both indolent and aggressive phenotypes, which are very difficult to differentiate at diagnosis. Circulating biomarkers measured at diagnosis that distinguish indolent from lethal PCa are yet to be established and could be used clinically to avoid both over- (for indolent disease) or under- (for aggressive cancer) treatment.

This thesis investigated whether circulating metabolites, measured by nuclear magnetic resonance (NMR), are observationally associated with PCa progression (clinical progression, PCa metastases or PCa death) in a cohort of patients with localised disease and eligible to be randomised into a controlled trial (RCT). Using genetic data (Mendelian randomization), the thesis also examined if circulating metabolites are causally linked to PCa mortality. The thesis next explored whether the circulating metabolome of men with localised PCa, who had been randomised to lifestyle (brisk walking) and dietary (lycopene supplementation or increased fruit and vegetable and reduced dairy milk intake) interventions, was altered by these interventions and if, in turn, the altered metabolites were causally linked to PCa death. Finally, the thesis also evaluated the performance of two metabolomic risk scores, previously developed in the general population, at predicting PCa death in men with prostate specific antigen (PSA) detected localised PCa, in a clinical setting.

In the observational analyses, there was suggestive evidence that some metabolites (lipoproteins, cholesterol, glycolysis, fluid balance and inflammation markers) were associated with PCa progression. In the Mendelian randomization genetic analysis, there was evidence that the following metabolites causally increased PCa mortality: total-, free- and esterified-cholesterol, some measures of intermediate-, low- and very low-density lipoprotein cholesterol, sphingomyelins, apolipoprotein B, omega-3 fatty acids, docosahexaenoic acid and valine. Some high-density and very low-density lipoprotein related measures, histidine and the ratio of omega-6 to omega-3 fatty acids were causally associated with decreased PCa mortality. There was some evidence that the randomised lifestyle and dietary interventions may alter lipids, alanine, lactate and pyruvate. In the Mendelian randomization causal analysis, there was evidence that some of the altered metabolites in the brisk walking intervention (cholesterol esters in medium VLDL, phospholipids in chylomicrons and extremely large VLDL) may be linked to increased PCa mortality risk. Lastly, there was evidence that two metabolomic risk scores developed in the general population did not predict PCa or all-cause mortality in a clinical setting.

This thesis supports a role for circulating metabolites in PCa progression, with implications for diagnostic and interventional research. Further studies of localised PCa with larger sample sizes and staging information are however needed to establish the role that metabolites, particular lipids, play in PCa progression and to allow for clinical translation.
Date of Award20 Jun 2023
Original languageEnglish
Awarding Institution
  • University of Bristol
SupervisorRichard M Martin (Supervisor) & Athene Lane (Supervisor)

Cite this

'