4-year PhD Studentship: An investigation into the performance of screening and diagnostic variables to predict 15-year prostate cancer mortality informing diagnostic and treatment strategies.

Prostate cancer is the UK’s commonest male cancer and represents a significant public health challenge. The UK National Screening Committee currently does not recommend prostate cancer screening1. The CAP randomised controlled trial (RCT) was influential in this recommendation, showing no significant prostate cancer-specific mortality benefit of one-time PSA screening at 10-years follow-up [rate ratio 0.96 (95%CI:0.85,1.08)].2 There are also concerns about screening’s adverse-effects on men’s quality-of-life through biopsy-related complications, overdiagnosis and overtreatment.

Utilising the long-term follow-up data from CAP provides the opportunity to consider whether any diagnostic factors affect the long-term outcomes for an individual man. This studentship could provide information that would help doctors decide diagnostic tests and treatments for prostate cancer, that would maximise the benefits and potentially reduce the harms from screening. Questions such as, what PSA cut point should be selected to start the screening programme? Answers to these questions could be used to build prognostic tools.3

Aims and objectives

Determine if measurements at diagnosis (age, PSA level and tumour characteristics) help doctors predict long-term outcomes for an individual man, to better decide his diagnostic tests and treatments for prostate cancer. Enabling the provision of high-quality information to help men and their doctors understand the possible benefits and harms of prostate cancer screening and will potentially be used by the National Screening Committee to update their prostate cancer-screening recommendations for the NHS.

1. Determine the performance of screening (age, PSA) and diagnostic (Gleason, stage) variables to predict 15-year PCa-mortality to inform diagnostic and treatment strategies.

Methodology

CAP randomised 415,357 men aged 50-69 from 573 primary-care practices across 8 centres in England and Wales between 2001-2009. This work exploits the continued linkage of >96% of the originally randomised CAP trial men to routine electronic NHS databases, making data collection and analysis extremely efficient.

This PhD studentship offers the opportunity to develop working relationships with the experienced CAP RCT research team of clinical and non-clinical researchers. Whilst working with linked data, and potentially conducting supplementary data collection of granular prostate cancer Gleason grade, provides the opportunity for students to develop data science and statistical skills. Beginning with a review of existing prostate cancer screening literature, and through discussion with clinical advisors the student will develop clinically meaningful comparisons for inclusion in the model to determine the performance of screening and diagnostic variables.

The student will fit a Cox regression model to evaluate the individual and joint performance of routine screening (age, PSA) and diagnosis (Gleason, T stage) variables in predicting 15-year prostate cancer-specific mortality. Testing assumptions (e.g., proportional hazards) and examine alternative methods (e.g., time-specific mortality rate-ratios4). Predictive performance will be evaluated using discrimination, calibration, decision-curve analysis and comparison with existing models.

How to apply for this project

This project will be based in Bristol Medical School - Population Health Sciences in the Faculty of Health Sciences at the University of Bristol.

Please visit the Faculty of Health Sciences website for details of how to apply