Single-sequence multi-dimensional MRI for customised intervention planning of brachytherapy in cervical cancer

Project reference: SIE_23 
First supervisor: Isabel Dregely 
Second supervisor: Sebastien Roujol 

Start date: October 2021  

Project summary:  Brachytherapy is an important part of treatment for patients with cervical cancer. Advanced MRI-based image guidance has potential to transform clinical practice in the direction of customised brachytherapy enabling better accuracy of target and organs-at-risk delineation as well as individualised dose administration for improved clinical outcome regarding local control and overall survival. Primary Aim: To develop a novel multi-dimensional MRI sequence (3D spatial + quantitative T2 and diffusion information) for precision brachytherapy treatment guidance Secondary Aims: 1) To assess the geometric accuracy of the sequence including depiction of the applicator in phantoms and patients; 2) To assess the accuracy and reproducibility of quantitative biomarkers (T2/ADC) in phantom, healthy volunteers and patients. 
Project description: Around 3,200 women are diagnosed with cervical cancer in the UK each year. High Radiation Dose Brachytherapy (HDR-BT) involves the use of a locally placed radioactive source and is an important component in the treatment. MR imaging is used for detection and staging of cervical cancer and now also increasingly used for planning of brachytherapy treatment. Compared to standard CT-based planning, MR has clear advantages due to superior soft tissue contrast providing opportunities for:

  1. Superior identification of target volume while simultaneously minimising toxicity to adjacent organs;

  2. Customisation of dose delivery by providing multidimensional information (biomarker changes vs time) during the entire course of HDR-BT; and

  3. Checking the accurate placement of the applicator before radiation dosing.

However, clinical challenges are that current MR sequences are limited to 2D thick slices, are prone to geometric distortions and image artifacts can induce errors when estimating the exact applicator position. Importantly, ‘functional’ MR sequences, despite their great promise for ‘biology-based’ radiation dose customisation, are currently not feasible as image quality is not sufficient for intervention.

Hypothesis: A novel multi-dimensional MRI sequence to acquire both anatomical and functional data with geometric precision in a single scan will reduce treatment margins, improve applicator placement and enable customisation of dose delivery to improve image-guided brachytherapy compared to our current standard imaging practice. Project plan:  WP1: Develop a novel multi-dimensional MRI sequence optimised for precision radiotherapy A novel single-scan 3D MRI sequence, using interleaved magnetization preparation to encode both T2 and diffusion-weighted image contrast will be developed to specifically address the challenges for brachytherapy planning: To achieve 1) 3D isotropic high-resolution, 2) excellent geometric fidelity and 3) accurate depiction of applicator position. 1)-3) will be assessed using a 3D MRI geometric phantom, healthy volunteer and pilot patient data. WP2: Quantification of anatomical and ‘functional’ biomarkers for customised intervention planning; Assessment of accuracy & reproducibility Extension of the above MR sequence to obtain quantitative co-registered anatomical and functional biomarkers to facilitate customization of dose delivery. Accuracy and reproducibility of quantitative T2 relaxation time and diffusion apparent diffusion co-efficient measurements will be assessed in a MRI phantom study, healthy volunteers (n=10) and patients with cervical cancer (n=10).  WP3: Towards a clinical advanced MR-guided brachytherapy treatment pathway Pilot data will be used to develop a pilot study with the aim of implementing an MRI based pipeline in the brachytherapy pathway.


Project timelines:

0-24 months (WP1&2): MRI sequence developed; phantom testing and healthy volunteer study commenced; data analysis commenced.


  • Novel single-scan MRI sequence incorporating 3D spatial, T2 & diffusion information as a WIP (Work-In-Progress package towards product in collaboration with industry partner Siemens) including image analysis package and manual;

  • Publications submitted summarising technical evaluation.

24-36 months (WP3): Patient study commenced; Data analysis; Write up of publications.  


  • Implementation of MR workflows in clinical pilot study in collaboration with radiology and radiotherapy clinical consultants; 

  • Publication submitted summarising pilot patient study.  

36-42 months: Pilot studies completed; Finalise data analysis; Thesis write up.