Use this URL to cite or link to this record in EThOS:
Title: Multi-parametric MRI of prostate cancer : assessment of spectroscopy, diffusion, dynamic contrast and relaxometry for active surveillance and staging
Author: Papadopoulos, Ioannis
ISNI:       0000 0004 7967 447X
Awarding Body: Swansea University
Current Institution: Swansea University
Date of Award: 2018
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Magnetic Resonance Imaging (MRI) was first applied back in the 1970s and is a medical imaging technique for anatomical and physiological processes of the body. It is characterized by high sensitivity and distinctive diagnostic ability on soft tissues. A variety of pulse sequences with different characteristics are available for the assessment and subsequent therapeutic approaches to prostate cancer. Advanced MR techniques such as Diffusion Weighted Imaging (DWI), Dynamic Contrast Enhanced imaging (DCE), MR Spectroscopy (MRS) and T2 Relaxometry (R2) can be applied along with conventional T2-weighted (T2w) imaging, called multiparametric MRI (mp-MRI), to add useful clinical value. However, MRI's low specificity can make tumour detection difficult or impossible. Moreover, not every technique is appropriate for low-grade low-volume prostate cancer evaluation and thus some have been excluded from several studies and clinical protocols, the optimization of which is critical for best results. In the present study a combination of T2w and DWI was evaluated as a potential diagnostic tool for detection, staging and guided biopsies for patients with possible early-stage prostate cancer. Quantification of the noise floor in the DWI images and careful fitting of the data suggests that the monoexponential model provides a very good fit to the data and there is no evidence of non-Gaussian diffusion for b-values up to 1000 s/mm2. The kurtosis, a measure of non-Gaussianity, can therefore not be used as a biomarker. However, the ADC value was significantly higher for suspicious regions than for normal regions. In addition, DWI showed high correlation with overall scores assigned according to the Prostate Imaging Reporting And Data System (PIRADS) and the results suggest that a simplified protocol combining T2w and DWI can be cost and time efficient for early-stage prostate cancer diagnostic programme, combining robust MR biomarkers that can be reliably quantified and appear well-suited for general clinical practice. DCE MRI was also evaluated in differentiation between radiologically defined normal and suspicious regions in the prostate from patients suspected of early-stage prostate cancer. The MRI signal increase following contrast agent administration was evaluated for regions defined as suspicious for malignancy or healthy, respectively. Background noise levels were calculated and signal intensities below a set threshold were excluded. The data was analysed using a phenomenological model that enables both quantification of the signal increase due to contrast agent uptake and automatic classification of the curve type according to previously established criteria. Curve-type classification did not provide a clear differentiation between normal and suspicious regions either by visual or computer analysis. However, statistically significant differences in the magnitude of the signal increase for normal and suspicious regions were observed. The results strengthen evidence that DCE MRI has limited value in the assessment of early stage, small volume prostate cancer in mp-MRI and that curve-type classification is an unreliable technique for differentiation of normal and suspicious prostatic tissue. MRS scores did not significantly correlate with overall PIRADS scores and added no clinical value compared to an mp-MRI protocol without MRS. In this study several post-processing steps were used to evaluate the acquired clinical spectra without the use of endorectal coil. The post-processing steps, especially baseline correction, affected the signals heavily. Two different approaches were used for peak integrations but the noisy spectra made their identification ambiguous. MRS was removed from our mp-MRI protocol. In view of significant current interest in quantitative MRI and considering the value of T2w imaging as mainstay in the diagnosis of disease, R2 mapping was investigated as a potential addition to the protocol. To this end, different R2 mapping protocols were tested on contrast phantoms and a small number of volunteers and patients. Although promising, R2 mapping was not incorporated in the current clinical mp-MRI protocol due to a lack of a standardized parametrization and differences observed between different protocols, which require further investigation and the development of optimal, standardized protocol.
Supervisor: Shermer, Sophie ; Phillips, Jonathan Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral