Identifying biomarkers with predictive utility in the clinical management of prostate cancer

Green, W., 2018. Identifying biomarkers with predictive utility in the clinical management of prostate cancer. MPhil, Nottingham Trent University.

William Green 2018 for IREP.pdf - Published version

Download (2MB) | Preview


Background: There is significant variation in clinical outcome between patients diagnosed with prostate cancer (CaP). Although useful, statistical nomograms and risk stratification tools alone do not always accurately predict an individual's need for and response to treatment. As a result there remains a need to identify and validate biomarkers for predicting prostate cancer outcomes using robust and routinely available pathology techniques to recognize men at most risk of premature death due to prostate cancer.

The day-to-day treatment options available to clinicians are continually evolving, with newer technologies and a greater understanding of the tumour biology prompting innovative approaches. However, despite this, all techniques have considerable associated side effects and there is a great deal of disagreement regarding which patients need radical treatment and which can be safely monitored, therefore avoiding unnecessary morbidity.

If more accurate risk stratification can be achieved using newly developed biomarkers (probably in addition to conventional staging techniques) then the aim is to reduce unnecessary treatment and assist timely and appropriate surgical and oncological intervention.

Aims and Objectives: We aimed to develop biomarkers predictive of outcome in prostate cancer, in particular ones that could be used in a mainstream NHS laboratory to help clinicians and patients make informed decisions regarding the management of their disease. We also intended to investigate whether bioinfomatic techniques such as artificial neural network analysis (ANN) could play a role in prostate cancer biomarker identification. We also felt it was important to validate our clinical data set and associated tissue micro array (TMA) by comparing its performance against previously identified biomarkers shown to have predictive utility in prostate cancer.

Methods: A tissue microarray (TMA) was constructed from transurethral resection of prostate (TURP) and transrectal ultrasound-guided (TRUS) prostate biopsy samples that werehistologically proven to demonstrate prostate cancer. Patients had undergone these procedures either to deal with troublesome urinary symptoms or had presented with an elevated prostate specific antigen (PSA) blood test. A comprehensive clinical data set of parameters conventionally used to decide upon treatment and monitor clinical response was collected. ANN analysis was used to identify candidate markers conferring increased risk of death and metastasis interrogating a public cDNA array, alongside a conventional literature review identifying previously published biomarkers that could be used to validate the TMA and clinical dataset. Immunohistochemical analysis of the TMA was carried out and univariate and multivariate tests performed to explore the association of tumour protein levels of identified biomarkers with various clinical endpoints, particularly time to death and metastasis.

Results: We successfully demonstrated associations between various biomarkers, and in particular validated our TMA and clinical dataset against the previously published marker Ki67, showing that Ki67 is predictive of CaP-specific survival and development of future metastases. In addition we were able to identify an entirely novel prostate cancer marker, DLX2, using artificial neural network analysis and demonstrate it has a statistically significant association with the development of prostate cancer metastases.

Conclusion: The Nottingham TMA has been shown to have utility in the investigation of candidate biomarkers in prostate cancer. We have also demonstrated that bioinfomatic techniques such as artificial neural network analysis can be employed to isolate candidate markers. During this work we have identified two cancer cell proliferation markers, Ki67 and DLX2, that may be able to inform clinical decision making when identifying patients for suitable for prompt active treatment versus active surveillance.

Item Type: Thesis
Creators: Green, W.
Date: 2018
Divisions: Schools > School of Science and Technology
Record created by: Linda Sullivan
Date Added: 27 Mar 2018 13:47
Last Modified: 27 Mar 2018 13:47

Actions (login required)

Edit View Edit View


Views per month over past year


Downloads per month over past year