- Title
- Multiparametric magnetic resonance imaging for liver function quantification
- Creator
- Phonlakrai, Monchai
- Relation
- University of Newcastle Research Higher Degree Thesis
- Resource Type
- thesis
- Date
- 2023
- Description
- Research Doctorate - Doctor of Philosophy (PhD)
- Description
- The ultimate goal of radiation therapy is to achieve precise treatment while minimising damage from the radiation beams to functioning tissue to improve patient outcomes and quality of life post-treatment. However, radiation-induced liver toxicity remains a risk in cirrhotic patients undergoing stereotactic body radiotherapy for hepatocellular carcinoma (HCC). This side-effect can limit the standard approach of liver radiotherapy in terms of dose escalation and re-irradiation. Currently, only global dose constraints are used in radiation treatment planning, without accounting for the spatial heterogeneity of liver function. However, if spatial liver function (function at the image voxel level) is known prior to treatment, it could allow for the mitigation of treatment side effects through functional avoidance radiation treatment planning. Magnetic resonance imaging (MRI) could play a vital role in radiation oncology by providing spatial heterogeneity of liver dysfunction in cirrhotic patients. MRI-derived functional mapping would allow for dose sparing in known healthy regions, reducing adverse consequences and maximising well-functioning liver tissue to improve patient outcomes. This dissertation investigates multiparametric (mp) MRI for spatial liver function quantification, specifically for radiotherapy use. The mp-MRI consists of low-temporal resolution dynamic contrast-enhanced MRI (LTR-DCE MRI), intravoxel incoherent motion diffusion-weighted imaging (IVIM-DWI), and B1 corrected dual-flip angle T1 mapping (DFA T1 map). The LTR-DCE MRI and DFA T1 mapping are based on Gadoxetic acid disodium, a hepatocyte-specific contrast agent, which allows investigation of the hepatocyte extraction efficacy. Additionally, changes in microtissue architecture and tissue perfusion of the liver were investigated using IVIM-DWI, which allows for the measurement of water diffusion and perfusion separately. Gadoxetate LTR-DCE MRI was extensively used throughout this dissertation to assess the feasibility of liver function mapping and the clinical application of functional avoidance radiotherapy in liver cancer. The studies were performed with a retrospective patient dataset of 64 patients and a prospectively collected dataset of 13 patients. The overall underlying hypothesis of this study is that MRI can be used to investigate liver tissue function, estimate pre-treatment spatial liver function, assess radiation-induced liver damage, and develop functional sparing radiotherapy plans. Firstly, the feasibility of LTR-DCE MRI data in quantifying liver function with hepatic extraction fraction (HEF), input-relative blood flow (irBF) and mean transit time (MTT) using a deconvolution analysis method was investigated. The method's variability was deemed acceptable, with coefficient of variation (CV) of 19.7% for HEF and 27% for irBF. A significant correlation was observed between MRI-derived HEF, irBF, and MTT parameters and ALBI score. Additionally, a significant difference in quantitative values of MRI-derived parameters was noted across different clinically defined liver function patient groups, with a p-value < 0.001. To assess the usefulness of IVIM-based measurements and T1 mapping in quantifying liver function, a comparison was made with HEF measurements. The results indicated that IVIM-derived Fp and Dfast values hold promise in distinguishing healthy liver from patients with mild liver function impairment for global liver function (p < 0.05). Moreover, these parameters exhibit high diagnostic accuracy in identifying liver dysfunction, with an area under the curve (AUC) of 0.89 (95% CI: 0.76, 1.00) for Fp and 0.950 (95% CI: 0.87, 1.00) for Dfast. At the voxel level, HEF was found to have a moderate positive correlation with IVIM-based perfusion parameter (Fp) and percentage T1 decrease (∆T1%) with r = 0.45, p < 0.001, and r= 0.43, p< 0.001, respectively. To investigate the dose-response of the liver to radiation, pre-treatment and post-treatment (~6 months) MRI-derived HEF maps were compared. The average dose-response of liver function reduction and the half dose of liver function reduction were 0.18% per Gy (EQD2) and 46.3 Gy (EQD2), respectively. Additionally, the HEF reproducibility was evaluated. This dissertation also explored the feasibility of integrating MRI-based liver function into radiation treatment plans. The results from nineteen patients revealed that a function-sparing plan could reduce the mean dose of functional sparing regions up to 37.5% compared to clinically approved plans. These findings lay the foundations for personalised liver radiotherapy for cirrhotic patients with HCC and improving routine clinical practice by incorporating functional maps for radiotherapy dose sparing. The study was approved by the Hunter New England Local Health District Human Research Ethics Committee (reference number 2019/ETH13629).
- Subject
- multiparametric MRI; DCE-MRI; IVIM-DWI; T1 map; radiation-induced liver disease
- Identifier
- http://hdl.handle.net/1959.13/1491350
- Identifier
- uon:53161
- Rights
- Copyright 2023 Monchai Phonlakrai
- Language
- eng
- Full Text
- Hits: 606
- Visitors: 1142
- Downloads: 576
| Thumbnail | File | Description | Size | Format | |||
|---|---|---|---|---|---|---|---|
| View Details Download | ATTACHMENT01 | Thesis | 5 MB | Adobe Acrobat PDF | View Details Download | ||
| View Details Download | ATTACHMENT02 | Abstract | 354 KB | Adobe Acrobat PDF | View Details Download |