Abstract
In this thesis nodal characteristics have been assessed with high field Magnetic Resonance Imaging (MRI) using a clinical scanner in order to discriminate non-metastatic from metastatic nodes of breast cancer patients. The final goal is to non-invasively determine nodal and tumor stage of breast cancer patients during one MR exam, thereby identifying which patients have non-metastatic nodes in order to prevent them from undergoing a surgical nodal staging procedure.
First ex vivo axillary lymph nodes of human female cadavers were characterized at 7T MRI. The findings were correlated with pathological analysis. Morphological features were identified and quantitative parametric analyses were performed. MRI detected all 45 nodes in the specimens and 6 additional nodes that were not detected by routine pathological analysis. B-cell follicles, efferent- and afferent lymph vessels and blood vessels were identified on high resolution (180µm isotropic voxel size) images.
Next the association of 7T MRI characteristics with metastatic nodal invasion, as determined by histopathology, in dissected sentinel nodes of breast cancer patients was investigated. Morphological criteria showed no discriminatory effect. Cortical thickness did show a high negative predictive value for the presence of metastases (threshold 3mm; p=0.021). T1, T2, T2* relaxation time constants and the apparent diffusion coefficients were also measured. Metastatic invasion resulted in a pan-nodal lengthening of T2*, even if only micro-metastatic invasion was present. The other parameters did not differ significantly.
Nodal lipid composition was determined by 1H-Magnetic Resonance Spectroscopy.The ratios of signals from unsaturated fatty acids to the total lipid signal differed significantly. Lipid content could serve as a marker for metastatic nodal invasion.
The potential of saturation prepared MR methods for discriminating nodes was also studied. The mean amide proton transfer difference and the mean magnetization transfer ratio at 2800Hz were not statistically significantly different.
As part of tumor staging of breast cancer patients, in addition to nodal staging, the feasibility of 7T breast MRI was evaluated. At 7T a 5.7 fold higher signal-to-noise ratio was measured than at 3T. Subsequently, the clinical application of 7T MRI in breast cancer patients receiving neo-adjuvant chemotherapy was explored. High morphological detail was obtained on fat-suppressed images (isotropic resolution 450µm³). The first results of diffusion weighted imaging and 1H-MRS were shown, proving that these techniques could potentially be performed clinically.
Without the use of contrast agents, discriminating metastatic from benign nodes is not straightforward, even at high resolution ex vivo 7T MRI. Therefore a contrast material with a T2* shortening effect, gadofosveset, was studied at 3T. Gadofosveset-enhanced diffusion weighted imaging was able to discriminate metastatic from benign axillary nodes on the basis of a significantly higher signal intensity on post-contrast b=500 images.
This thesis has explored the potential of high field MRI as a non-invasive method for staging breast cancer patients. Following the optimization of scan sequences, the combination of contrast-enhanced imaging with a negative contrast agent, using a breast coil which also depicts the axilla, 3T or even 7T MRI is a promising candidate for a one-stop staging tool for breast cancer patients.
Original language | English |
---|---|
Qualification | Doctor of Philosophy |
Awarding Institution |
|
Supervisors/Advisors |
|
Award date | 22 Nov 2011 |
Publisher | |
Print ISBNs | 978-90-393-5681-4 |
Publication status | Published - 22 Nov 2011 |
Keywords
- Econometric and Statistical Methods: General
- Geneeskunde(GENK)
- Medical sciences
- Bescherming en bevordering van de menselijke gezondheid