(EPSRC) Generation of an AI predictive model capable of identifying prostate and bladder cancer patients that will benefit from radiotherapy.

   Faculty of Biology, Medicine and Health

  , , , ,  Saturday, June 15, 2024  Competition Funded PhD Project (UK Students Only)

About the Project

Prostate and bladder cancer are both malignancies of the urological system affecting 1.8 million people. However, 50% of prostate and 25% of bladder cancer patients are respectively diagnosed as high risk aggressive diseases, both requiring either radical surgery (RS) or radiotherapy (RT)1,2. RS has been traditionally considered the gold-standard treatment for both prostate and bladder cancer, but it causes severe chronic side effects (e.g. erectile dysfunction, urinary incontinence)1,2. RT allows for organ preservation, showing no difference in overall survival to RS in either cancer. However, both prostate and bladder tumours can be RT-resistant, leading to recurrence1,2. Currently, no algorithm can identify which patients benefit from RT over RS. Since 1955, hypoxia (<2% O2) is known to drive radioresistance. It correlates with prostate cancer progression3. In bladder cancer, hypoxia-modifying treatment is standard-of-care, improving patient prognosis through enhancing RT2. Mechanistically, hypoxia drives genomic instability, epigenetics changes, and extracellular matrix remodelling, all processes associated with RT resistance2,4. We previously developed hypoxia signatures for prostate and bladder cancers. However, the lack of specific models to objectively analyse and interpret gene expression data impairs their implementation in the clinic. Recent advances in artificial intelligence (AI; e.g. gradient random forest models, artificial neural networks, gradient neural networks) are changing the paradigm, as they can analyse massive datasets and establish radiotherapy predictive models5. Here, we aim to establish a hypoxia-associated AI model to identify which bladder and prostate cancer patients benefit from radiotherapy over radical surgery. 

From our previous work, we have normalised transcriptomic and clinical data from eight bladder (n=1,259) and four prostate (n=917) cancer cohorts. Patients within each cohort will be stratified into “high” and “low” hypoxia-score groups using available normalised gene expression data and previously generated hypoxia gene signatures. Then, we will calculate the difference in gene expression within both groups for each cohort, establishing a consensus hypoxia expression dataset. Ontology enrichment analysis will identify genes related to radioresistance, using them to constitute a radiosensitivity signature. After retrospective validation of the signature, we will use AI models (e.g. artificial neural networks) to generate an algorithm capable of predicting outcomes from RS and RT, based on the developed signature and clinical parameters). A successful outcome of this project will be to develop a predictive AI model which can be tested prospectively in the clinical, helping patients and clinicians to identify the most suitable treatment for an individual patient and promoting personalised medicine approaches. 



Entry requirements

Candidates are expected to hold (or be about to obtain) a minimum UK Upper First or 2:1 (or equivalent) in a related area / subject. Candidates with an interest in Cancer Studies are encouraged to apply.

How to apply

For information on how to apply for this project, please visit the Faculty of Biology, Medicine and Health Doctoral Academy website (https://www.bmh.manchester.ac.uk/study/research/apply/).

Interested candidates must first make contact with the Primary Supervisor prior to submitting a formal application, to discuss their interest and suitability for the project. On the online application form select PhD Biochemistry (this is for application purposes only).

Equality, Diversity and Inclusion

Equality, diversity and inclusion is fundamental to the success of The University of Manchester, and is at the heart of all of our activities. The full equality, diversity and inclusion statement can be found on the website https://www.bmh.manchester.ac.uk/study/research/apply/equality-diversity-in

Biological Sciences (4) Mathematics (25)

Funding Notes

EPSRC DTP Studentship funding is for a duration of 3.5 years to commence in September 2024 and covers UK tuition fees and an annual stipend at the UKRI rate. Due to funding restrictions the studentship is only open to UK nationals.


1. Arrayeh, E. et al. Int J Radiat Oncol Biol Phys 2012.
2. Lodhi, T. et al. Clin Oncol 2021.
3. Bharti, S. K. et al. Neoplasia 2019.
4. Bigos et al. Frontiers in Oncology 2024.
5. Siddique, S. & Chow, J. C. L. Reports of Practical Oncology and Radiotherapy 2020

Register your interest for this project