Höfl, R. (2018). Visualization and quantification of tumour vasculature in 3D [Diploma Thesis, Technische Universität Wien]. reposiTUm. http://hdl.handle.net/20.500.12708/158660
In tumour biology, anti-angiogenic therapy is a method to treat cancer by targeting speciőc pro-angiogenic factors such as vascular endothelial growth factors (VEGF). Blocking their signalling pathways can inhibit sprouting angiogenesis, which is the foundation of new vessels in tumours. Due to resistance and toxicity of the treatment, these therapies showed very limited success on reducing tumour progression, especially in metastatic tumours. Vascular Imaging is a developing imaging method that provides information about the area and quantity of vessels, the perfusion and permeability of the vasculature and appearing malfunctions of vessels. Current Magnetic Resonance Imaging (MRI) protocols that are routinely used for vascular imaging, such as angiography and time of ŕight (TOF) based approaches, are not suitable for imaging tumour vasculature, due to reduced velocity of the blood in the tumour vessels. In addition, use of gadolinium-based contrast agents is problematic due to leaky tumour vasculature, and inability of this contrast agent to be contained within the vessels. As part of this thesis, a different approach for visualization and quantifcation of tumour vasculature was explored, which took advantage of the inherent susceptibility contrast of blood and the ultra-high field of a 15.2 Tesla MRI scanner. A murine study with nine mice was set up, for which a targeted therapy had an indirect effect on the tumour vasculature. The goal was to visualize the microvasculature in the tumours and to establish a scan protocol for tumour blood vessel imaging. Therefore a high resolution 3D imaging in vivo before and after the drug Treatment was performed, with a specifc focus on vessel quantifcation and contrast agent performance. Different sizes of iron oxide particles were used as agent for the studies and different scan parameters were tested, in order to determine an optimal set up for the imaging. The results were compared to administration of gadolinium-based agents, to underline the advantages of susceptibility measurements. For Parameter optimization of the scan protocols, a blood phantom prototype was built to look at the contrast-to-noise (CNR) ratio for different used imaging sequences. The results were analysed with the NIH software ImageJ and Matlab. Minimum and Maximum Intensity Projections of the used scan sequences were compared for the different contrast agents and scan sequences used. The results of the blood phantom measurements suggest the 30 nm iron-oxide particles as the most promising contrast agent. The murine studies showed that the blood pool agent leaked into the surrounding tissue and didn’t apply an usable T1 relaxation time effect on blood. Two gradient-echo scanning sequences, FLASH and FISP, showed the most promising results. Comparison with later performed in-vivo MRI measurements and micro Computer Tomography (CT) were done, to show that microvsculature couldn’t be readily visualized by MRI except for the vessels located along the tumour margins.