Monitoring breast tumour response to radiotherapy using super-resolution ultrasound and quantitative magnetic resonance imaging
File(s)
Author(s)
Morris, Megan
Type
Thesis
Abstract
Super-resolution ultrasound (SRUS), diffusion-weighted (DW), and dynamic contrast-enhanced (DCE) MRI can detect microenvironmental changes in tumours following treatment, with advantages over size-based methods for evaluating response. Assessing repeatability, prognostic value, and sensitivity of imaging biomarkers is critical for clinical translation.
Clinical data were collected from locally advanced breast cancer patients undergoing radiotherapy. Test-retest and longitudinal imaging assessments were conducted pre-radiotherapy, and at 2-weeks, 6-, 12-, and 24-months post-radiotherapy. Ultrafast contrast-enhanced ultrasound was used to generate 2D SRUS maps of microvascular structure and dynamics. DW-MRI analyses used mono-exponential and stretched exponential models, and DCE-MRI utilised Tofts modelling and non-parametric analysis. Quantitative parameters were extracted and evaluated. Radiological response was determined by expert breast radiologists. Biopsies from a subset of participants at baseline and 2-weeks post-radiotherapy were stained for CD31 and scored by a consultant breast histopathologist.
Repeatable and sensitive MRI response biomarkers included ADC from DW-MRI, reflecting changes in cellular density, and DCE-MRI parameters including Ktrans, initial wash-in rate (iWIR), final wash-out rate (fWOR), and final wash-out AUC (AUCfWO), which captured microvascular properties. ADC demonstrated the greatest sensitivity to early changes. DCE-MRI parameters were more informative for long-term monitoring, successfully detecting relapse nodules not evident on ADC maps. Trends showed non-responders at 12-months had higher pre-treatment ADC values and smaller post-treatment changes, and higher pre-treatment iWIR; though these trends did not reach statistical significance. SRUS-derived vessel density, vessel diameter, and blood velocity were repeatable and sensitive to early treatment changes. SRUS vessel density did not correlate significantly with CD31 count.
This is the first study to utilise SRUS, DW-MRI, and DCE-MRI to monitor long-term response (> 6-months) of breast cancer post-radiotherapy. Imaging biomarkers demonstrated repeatability and sensitivity to early and late treatment changes. Variability in treatment regimens and a small cohort classified for radiological response introduced challenges in evaluating prognostic potential.
Clinical data were collected from locally advanced breast cancer patients undergoing radiotherapy. Test-retest and longitudinal imaging assessments were conducted pre-radiotherapy, and at 2-weeks, 6-, 12-, and 24-months post-radiotherapy. Ultrafast contrast-enhanced ultrasound was used to generate 2D SRUS maps of microvascular structure and dynamics. DW-MRI analyses used mono-exponential and stretched exponential models, and DCE-MRI utilised Tofts modelling and non-parametric analysis. Quantitative parameters were extracted and evaluated. Radiological response was determined by expert breast radiologists. Biopsies from a subset of participants at baseline and 2-weeks post-radiotherapy were stained for CD31 and scored by a consultant breast histopathologist.
Repeatable and sensitive MRI response biomarkers included ADC from DW-MRI, reflecting changes in cellular density, and DCE-MRI parameters including Ktrans, initial wash-in rate (iWIR), final wash-out rate (fWOR), and final wash-out AUC (AUCfWO), which captured microvascular properties. ADC demonstrated the greatest sensitivity to early changes. DCE-MRI parameters were more informative for long-term monitoring, successfully detecting relapse nodules not evident on ADC maps. Trends showed non-responders at 12-months had higher pre-treatment ADC values and smaller post-treatment changes, and higher pre-treatment iWIR; though these trends did not reach statistical significance. SRUS-derived vessel density, vessel diameter, and blood velocity were repeatable and sensitive to early treatment changes. SRUS vessel density did not correlate significantly with CD31 count.
This is the first study to utilise SRUS, DW-MRI, and DCE-MRI to monitor long-term response (> 6-months) of breast cancer post-radiotherapy. Imaging biomarkers demonstrated repeatability and sensitivity to early and late treatment changes. Variability in treatment regimens and a small cohort classified for radiological response introduced challenges in evaluating prognostic potential.
Version
Open Access
Date Issued
2025-01-05
Date Awarded
2025-07-01
Copyright Statement
Attribution-NonCommercial 4.0 International Licence (CC BY-NC)
License URL
Advisor
Tang, Mengxing
Somaiah, Navita
Blackledge, Matthew
Sponsor
Cancer Research UK
Grant Number
A26234
Publisher Department
Department of Bioengineering
Publisher Institution
Imperial College London
Qualification Level
Doctoral
Qualification Name
Doctor of Philosophy (PhD)