Evaluation of Ultrasound Probe for Use in Bladder Radiotherapy

All experiments and data analysis will be performed in the Magnetic Resonance Imaging (MRI) and Radiotherapy departments respectively, at the Queen Elizabeth Hospital.

Five volunteers will be recruited to this study. They will each be required to follow a documented drinking protocol prior to each of three scan sessions. The drinking protocol is specific to this study and will not be intended for use with patients. To address the primary endpoint of the study (i.e. the precision of the ultrasound probe for measurement of bladder volume), measurement of a wide range of bladder volumes is required. In order to address this aim while making efficient use of the limited available MRI scanner time, the drinking protocol has been designed such that the volunteer will begin each scan session with an empty bladder and finish it with a full bladder. To this end, each volunteer will be required to void their bladder upon arrival and then consume 500ml of water immediately prior to the start of the scan session.

Each volunteer will then be positioned on the scanner couch according to the same protocol used for patients and alternate MRI and ultrasound scans will then be acquired. Given the time required for each scan it is anticipated that five MRI scans and fifteen ultrasound scans (three scans at a time to reduce the noise on the results, i.e. ultimately giving five ultrasound bladder volume measurements) will be acquired in each session. Each volunteer will be asked to attend for three sessions over the course of a number of weeks so that the repeatability of bladder filling patterns can be determined.

Data processing and analysis

The volume measurements from the ultrasound probe will be recorded and each of the MRI images will be exported to the Radiotherapy department where the bladder volume will be determined. The precise timings of all scan acquisitions will be recorded during the imaging session. Bladder volumes measured by the ultrasound probe, which can be read instantly, will be recorded during the session. To evaluate the primary endpoint, the precision and accuracy of the ultrasound probe will be quantified by comparing the measured values to those expected from the MRI volumes. Each volume measurement from the MRI scans will be checked by a second observer to minimise any associated researcher error.

To evaluate the secondary endpoint, the imaging protocol described above will provide up to 15 MRI images for each volunteer that can be used to inform a comparison of the different adaptive strategies. However, since patients are always asked to empty their bladder prior to radiotherapy treatment (in order to minimise the volume that receives radiation), the deliberately wide-ranging bladder volumes that will have been measured in this study will not be representative of those that would be observed in clinical practice. To address this problem, a suitable dataset that quantifies the distribution of bladder volumes encountered in a typical patient cohort will be used (Christie dataset, permission obtained.Software will be written to randomly assign a bladder volume to each day of each volunteer's 'treatment'. The MRI scan from that volunteer that most closely corresponds to the selected volume will then be assigned for that day's treatment. This will be repeated for the full 20 days of treatment and the importance of each MRI scan can then be factored into the results analysis accordingly.

The treatment planning system within the Radiotherapy department will then be used with the MRI scans to simulate a range of different 'treatments' according to each of the different adaptive bladder radiotherapy strategies to be compared. The different strategies are summarised here:

• Conventional strategy: This is the current standard treatment, in which the patient plan is prepared based on an initial CT scan of the patient with an empty bladder and no account is made of subsequent daily variations in bladder shape and volume.
• Plan-of-the-day strategy: Based on an initial CT scan of the patient with an empty bladder, three different treatment plans are created based on an assumption of how the bladder might expand. These are intended to simulate empty, partially full and full bladders. The patient is scanned before treatment each day and the most appropriate plan is selected for treatment.
• Dynamic strategy: Three different treatment plans are created based on a series of initial CT scans of the patient with empty, partially full and full bladder. These would be expected to directly represent the empty, partially full and full bladders for the individual patient. The patient is scanned before treatment each day and the most appropriate plan is selected for treatment.
• Composite strategy: The patient is CT scanned with an empty bladder and treated according to the Conventional strategy for the first 7 days of treatment with daily imaging. The first 5 daily scans of the patient are used to generate plans based on small, partially full and full bladder. From the 8th day of treatment the daily image is used to select which of these plans is appropriate for treatment.

Data processing of the MRI scans will be carried out retrospectively. The MRI scans will be anonymised and exported to the Radiotherapy department. The data will then be used to (i) acquire bladder volume information to compare with that acquired from the ultrasound probe and (ii) prepare and evaluate radiotherapy treatment simulations to investigate the benefits of different adaptive bladder radiotherapy strategies

Comparison of different adaptive techniques will focus on how much radiation is received by the bowel and the extent to which the radiation misses the disease on any day of treatment.