Subtype specific metabolic vulnerabilities in pancreatic cancer

Cunningham, Richard (2020) Subtype specific metabolic vulnerabilities in pancreatic cancer. PhD thesis, University of Glasgow.

Abstract

Pancreatic cancer has historically been characterised by its poor prognosis, with very little increase in 5-year expectancy relative to other, comparable cancertypes. This clinical observation is largely due to existing difficulties in identifying therapeutics effective in managing metastasised disease, a task compounded by the heterogeneity associated with pancreatic cancers. Concentrated efforts have been made in recent times to mitigate this issue, with the emergence of a range of subtyping strategies allowing for the stratification of patients. This categorisation of patients into workable groups thus serves to limit the degree of heterogeneity found within subgroups, with hypothetical, and otherwise unobservable, vulnerabilities shared between subtypes. This thesis aims to explore these potential therapeutically exploitable vulnerabilities by describing the extensive characterisation of pancreatic cancer subtypes via a diverse collection of patient derived cell-lines. This characterisation was achieved by profiling of the transcriptome via RNA-seq analyses, the proteome via mass-spectrometric approaches, and activation status of metabolic processes associated with oncogenesis in pancreatic cancer via functional assays. This work therefore facilitates the identification of vulnerabilities by utilising the profiles of subtypes generated in this manner and devising therapeutic strategies effective in treating the disease by interrogating dysregulated pathways. Within PDCLs, two pancreatic cancer subtypes were first identified which aligned with those described in patients: the squamous and classical subtypes. Preliminary profiling efforts highlighted a dysregulation of genes involved in metabolism across these subtypes in vitro, with active glycolysis associated with the aggressive squamous subtype and fatty acid biosynthesis and metabolism upregulated in the classical subtype. Further proteomic characterisation then validated this observation, providing further evidence for the existence of distinct metabolic profiles associated with these subgroups. Follow-up experimentation which focused on metabolic outputs then generated metabolic profiles for each subtype, with in vitro phenotypes reflecting findings at the transcriptome and proteome level and demonstrating enhanced glycolysis and fatty acid oxidation in the squamous and classical subtypes respectively. Subsequent attempts to target arms within those subtype-associated metabolic pathways yielded mixed results. Inhibiting glycolysis via targeting of ALDOA successfully mediated a selective response in cell-lines associated with the squamous subtype, while classical cell-lines required a combination therapy to suppress metabolic flexibility to induce sensitivity to inhibition of fatty acid synthesis via targeting of FASN. An adjacent and complementary arm of research involved collaborative highthroughput drug repurposing screens to identify additional targets for follow-up. This involved an initial screen of ~600 compounds in 8 PDCLs. Results generated as part of this screening approach highlighted the potency of statins in effecting a significant response selectively in squamous cell-lines. Research probing the mechanism by which statins induce this selective inhibition suggested that differences in degradation of the statin target HMGCR and cholesterol homeostasis may confer resistance to cell-lines classified as squamous, with findings demonstrating the potential of dietary components found in commonly ingested foodstuffs to mitigate the effects of statins in the otherwise sensitive, squamous subtype. This thesis therefore identified a range of therapeutic strategies effective in mediating sensitivity in in vitro pancreatic subtypes, with mechanisms of actions determined for each strategy. Results have demonstrated that pancreatic cancer cells exhibit differential sensitivities to metabolic inhibition, with subtype classification found to act as a predictor of sensitivity. As these in vitro subtypes recapitulate stratifications described in patients, these therapeutic strategies are of clinical relevance in the treatment of pancreatic cancer.

• Item Type: Thesis (PhD)
• Qualification Level: Doctoral
• Keywords: Pancreatic cancer, subtyping, patient stratification, cancer metabolism, translational research, drug repurposing.
• Subjects: R Medicine > RC Internal medicine > RC0254 Neoplasms. Tumors. Oncology (including Cancer)
• Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Cancer Sciences > Translational Research Centre
• Supervisor's Name: Biankin, Professor Andrew
• Date of Award: 2020
• Depositing User: Richard Cunningham
• Unique ID: glathesis:2020-80267
• Copyright: Copyright of this thesis is held by the author.
• Date Deposited: 19 Mar 2020 08:58
• Last Modified: 27 Apr 2022 09:03
• Thesis DOI: 10.5525/gla.thesis.80267
• URI: https://theses.gla.ac.uk/id/eprint/80267