This feasibility study is investigating the application of magnetic resonance imaging (MRI) and transperineal ultrasound (TPUS) to visualise potential sites of relapse and surrounding normal tissue in patient having post-operative prostate cancer radiotherapy. Structure visualisation on MRI and TPUS will be compared to current standard computer tomography (CT) and cone-beam computer tomography (CBCT) imaging.

Approximately 47,000 men are diagnosed with prostate cancer yearly in the UK, 15% of which receive surgery (radical prostatectomy) as primary treatment. Of these men one third develop recurrence after prostatectomy and are offered post-operative radiotherapy.

Radiotherapy is targeted at the area where the prostate used to be, this includes potential relapse sites (PRS); vesicourethral anastomosis, bladder neck and retrovesicle space. When planning post-operative radiotherapy, the lack of a visible target means surrounding normal tissue is used to guide clinical target volume (CTV) delineation. The CTV is the area of tissue to be treated to the full radiotherapy prescription and includes PRS. Typically, the CTV is defined on a radiotherapy planning computer tomography (CT) scan.

The CTV lies adjacent to the bladder and rectum and is highly susceptible to positional changes influenced by organ filling. To identify and correct for CTV displacement during daily radiotherapy a cone-beam CT (CBCT) is acquired, this is an image similar to a CT scan but taken immediately prior to treatment delivery.

CT visualisation of PRS has been reported as inadequate, increasing the risk of missing or miss-identifying CTV structures during radiotherapy planning and delivery. To compensate for restricted visualisation on CT and CBCT an additional 1cm circumference around the CTV is treated to the prescription dose, this volume is referred to as the planning target volume (PTV). This margin reduces the risk of missing the target however its inclusion of normal tissue increases the risk of patient toxicity.

We propose that visualisation of PRS and surrounding normal tissue structures is enhanced on magnetic resonance imaging (MRI) compared to CT and on transperineal ultrasound (TPUS) compared to CBCT. This project investigates the reality of this statement by quantifying the confidence and reliability with which anatomical structures can be identified and delineated on all four image types.

Inclusion Criteria:

• Patient has undergone radical prostatectomy
• Histologically confirmed prostate cancer
• Patient due to receive post-operative prostate bed +/- pelvic lymph node radiotherapy
• WHO performance status 0-1
• Age ≥ 18 years
• Written informed consent

Exclusion Criteria:

• Recruitment into POPS trial
• Contraindications to MRI such as having a pacemaker or non-MRI compatible implanted device