MR Thermometry Accuracy and Prospective Imaging-Based Patient Selection in MR-Guided Hyperthermia Treatment for Locally Advanced Cervical Cancer
by
, ,
Iva VilasBoas-Ribeiro
1,*, 
Sergio Curto
1
,
Gerard C. van Rhoon
1,2
,
Martine Franckena
1 and 
Margarethus M. Paulides
1,3
1
Department of Radiotherapy, Erasmus MC Cancer Institute, University Medical Center Rotterdam, 3015 GD Rotterdam, The Netherlands
2
Department of Radiation Science and Technology, Faculty of Applied Sciences, Delft University of Technology, 2629 JB Delft, The Netherlands
3
Center for Care and Cure Technologies Eindhoven (C3Te), Department of Electrical Engineering, Eindhoven University of Technology, 5600 MB Eindhoven, The Netherlands
*
Author to whom correspondence should be addressed.
Cancers 2021, 13(14), 3503; https://doi.org/10.3390/cancers13143503
Submission received: 11 June 2021
/
Revised: 2 July 2021
/
Accepted: 5 July 2021
/
Published: 13 July 2021
(This article belongs to the Special Issue MR Guidance for Real-Time Interventions in Oncology)
Simple Summary
Monitoring and controlling the temperature distribution combined with precise energy delivery are key components for hyperthermia treatment success. Magnetic resonance (MR) imaging is used clinically to monitor the temperature of the treated volume non-invasively. However, there are no comprehensive systematic studies on MR thermometry accuracy during deep pelvic hyperthermia, and the few investigational studies suffer from a high probability of bias due to lacking objective criteria for data inclusion. This study presents the first systematic analysis and defines an imaging-based criterion for prospective patient selection to standardize clinical MR thermometry accuracy assessments.
Abstract
The efficacy of a hyperthermia treatment depends on the delivery of well-controlled heating; hence, accurate temperature monitoring is essential for ensuring effective treatment. For deep pelvic hyperthermia, there are no comprehensive and systematic reports on MR thermometry. Moreover, data inclusion generally lacks objective selection criteria leading to a high probability of bias when comparing results. Herein, we studied whether imaging-based data inclusion predicts accuracy and could serve as a tool for prospective patient selection. The accuracy of the MR thermometry in patients with locally advanced cervical cancer was benchmarked against intraluminal temperature. We found that gastrointestinal air motion at the start of the treatment, quantified by the Jaccard similarity coefficient, was a good predictor for MR thermometry accuracy. The results for the group that was selected for low gastrointestinal air motion improved compared to the results for all patients by 50% (accuracy), 26% (precision), and 80% (bias). We found an average MR thermometry accuracy of 2.0 °C when all patients were considered and 1.0 °C for the selected group. These results serve as the basis for comprehensive benchmarking of novel technologies. The Jaccard similarity coefficient also has good potential to prospectively determine in which patients the MR thermometry will be valuable.
