Novel Approaches to Prostate Cancer Detection, Diagnosis and Stratification

Abstract

Prostate cancer is the most prevalent cancer type in men in the UK. Prostate Specific Antigen (PSA) is currently used as a marker for this solid cancer, being secreted by the prostate and measured in peripheral blood samples. PSA is used for diagnosis, prognosis, and clinical management of patients undergoing treatment. However, due to PSA’s non-specific nature, testing has led to over-diagnosis in patients with indolent disease, and an inability to solely define stratification of disease, there is an unmet need to find other non-invasive biomarkers of prostate cancer in early detection, increased sensitivity in determining benign prostate hyperplasia (BPH) from cancer, and indicating the severity of cancer in those who may or may not need treatment. Data independent acquisition (DIA) tandem mass spectrometry (MS/MS) is defined as one of the best methods for analysing and quantifying large amounts of biomolecules in a sample, suitable for biomarker discovery. Sequential Window Acquisition all Theoretical fragment ion spectra Mass Spectrometry (SWATH-MS) and High Definition Mass Spectrometry E (HDMSe) are well-established DIA-MS/MS methods that have been proven to have high, un-bias identify rates of various analytes. Combining omics data collected for proteomics via SWATH-MS, and lipidomics and small-metabolites from HDMSe provides an in-depth biological understanding of prostate cancer when comparing controls, BPH and prostate cancer patient samples from the Guernsey Prostate Cancer cohort. Differential expression using LogFC and feature selection, machine learning using Boruta Random Forest was used to select critical analytes involved in disease progression and revealed the Liver-, Retinoid-, Farnesoid-X receptors and complement pathway as significant. From the same cohort, a group of samples was collected from men before the diagnosis of prostate cancer. These samples are vital in identifying an early detection biomarker and revealing key analytes involved in progression of symptomatic disease. SWATH-MS was used again to show a comprehensive list of proteins combined from 3 reference libraries. Differential expression revealed a group of proteins which reached a cumulative AUC of 1, potentially being a panel for early detection of prostate cancer. EGFR’s involvement in angiogenesis, growth of tumours and development of vasculature systems was found to be activated and significant. There is currently little literature on the analysis of early detection in patient samples making these results extremely novel. With PSA’s downfall in specificity, it is still the only approved biomarker for prostate cancer in the UK. Once a biomarker has been revealed, it is predominately measured in the clinic via Enzyme-Linked Immuno- Sorbent Assay (ELISA), specifically sandwich assay. Three polyclonal antibodies were raised against PSA p eptides and optimised for use in a sandwich assay. These antibodies were proven to bind to whole protein PSA, but high background and low specificity was seen in trying to quantify PSA in patient samples. These antibodies may be more appropriately used in the targeted, antibody-enriched quantify MS technique Stable Isotope Standards and Capture by Anti-Peptide Antibodies (SISCAPA-MS), but more testing would be needed. Overall, this research attempts to identify new stratification and early detection biomarkers for prostate cancer and optimise new PSA antibodies to use via ELISA. Ultimately the results may inform essential pathways in prostate cancer progression in early and symptomatic disease. They also reveal issues with in the biomarker pathway, such as the importance of having a hypothesis when collecting samples, availability of patient clinical information, the influence of the method used for acquisition and data analysis, and challenges ensued when developing an ELISA.

Date of Award	1 Aug 2023
Original language	English
Awarding Institution	The University of Manchester
Supervisor	Anthony Whetton (Supervisor), Richard Unwin (Supervisor) & Paul Townsend (Supervisor)