The standard treatment for locally advanced cervical cancer is concurrent chemo-radiotherapy. This treatment is associated with long term side effects in around half of patients with up to 10% suffering from grade 3-4 toxicity.

The development of intensity modulated radiotherapy (IMRT) allows for shaping of radiotherapy fields to reduce the doses delivered to organs at risk (OARs). This does appear to reduce the risk of long and short term toxicity (although there is little randomized evidence). However pelvic organ position varies both between and even during radiotherapy fractions; this means that radiotherapy margins must be generous to allow adequate coverage of the clinical target volume (CTV) but this also increases dose to OARs.

There have been a number of studies evaluating pelvic organ motion in cervical cancer as well as assessing different adaptive radiotherapy strategies. These have included individualized margins, plan of the day and adaptive techniques. Most of these studies have been carried out using cone beam computed tomography (CBCT) imaging which is often poor quality with limited soft tissue contrast. MR offers better visualization of the tumour and OARs and is used for imaged guided brachytherapy treatment.

This study will explore the role of MR imaging in adaptive radiotherapy for cervical cancer with development of a number of theoretical treatment strategies.

Patients will undergo standard treatment during the study with five weeks of external beam radiotherapy (25 fractions) followed by a MR guided brachytherapy boost or external beam boost (further 10 fractions.) They will be treated with weekly cisplatin 40mg/m2 if clinically appropriate.

As part of standard treatment patients have a staging MR scan at diagnosis, a radiotherapy planning CT and an MR scan in the 4th week of treatment. Cone beam imaging will also be performed. Response will be assessed as standard with a MR scan at 3 months.

As part of the study patients will undergo 3 additional MR scans. These will take place in the 1st, 2nd and 5th weeks of treatment. Extra sequences will be added to the clinical mid-point scan in the 4th week of treatment. These MR scans will include anatomical images - with full and empty bladder as well as cine data (with a scan every minute for 10 minutes) to assess intra fraction motion. They will also include a DWI sequence, which will be used to assess if early prediction of response is possible. The frequency of cone beam imaging will be increased from approximately 10 scans as standard of care to 25 scans to allow for daily imaging.

The 1st MR scan will be contoured to outline clinical target volumes (CTVs) and OARs. A variety of planning strategies will be developed including standard planning target volume (PTV) margins, a plan of the day (POTD) approach, a POTD+ as well as an online adaptation model. These models will be used to assess coverage of CTV and PTV as well as dose to OARs using the scans obtained during the radiotherapy treatment. The practicality of each approach will also be assessed. Inter and intra fraction organ motion will also be analysed in order to develop patient specific models.

Inclusion Criteria:

• Histologically confirmed diagnosis of cervical cancer
• Treatment with primary curative intent
• Undergoing external beam radiotherapy (+/-chemotherapy) followed by brachytherapy or an external beam boost
• Age over 18

Exclusion Criteria:

• Any contraindications to MR identified after MR safety screening including completion of an MR Safety Screening Form (see appendix 1)
• Previous hysterectomy
• Unable to tolerate MR scans
• Metastatic disease
• Pregnancy