PDF, English Download (24MB) | Terms of use

Abstract

Tumor hypoxia has been widely recognized as a significant factor that increases treatment resistance and promotes malignant progression. High linear energy transfer (LET) radiotherapy (RT), e.g. with carbon ions (12C-ions), is expected to overcome this resistance factors as its lethality is less dependent on tumor oxygenation as compared to conventional low LET photon irradiation. However, the exact interplay between irradiation response, vascular changes, perfusion, and hypoxia is still not well understood, especially with respect to high LET RT. In the present thesis, the hypoxic status of syngeneic Dunning R3327 rat prostate tumor model sublines was characterized prior to and after irradiation with either low LET photons or high LET 12C-ions by multimodal imaging and histology. The initial oxygenation status of three subcutaneously transplanted Dunning tumor sublines (H, HI and AT1) was determined by photoacoustic imaging (PAI) which included the development and validation of a new PAI analysis protocol. The new protocol enabled the distinction of the three sublines based on their oxygenation profiles and their response towards external changes in oxygen supply. Subsequently, the effects of curative and sub-curative single dose irradiations with either photons or 12C-ions on the two hypoxic tumor sublines (HI and AT1) were investigated by pharmacokinetic modeling of longitudinal dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) data. For this, a novel method for estimating the contrast agent’s arrival time was developed in cooperation with the group for Image Analysis and Learning of the Interdisciplinary Center for Scientific Computing of Heidelberg University. It enables a delay correction of the contrast agent arrival time which improves fit accuracy and the reliability of the pharmacokinetic modeling results. The moderately differentiated HI-tumor showed increased vascular permeability 7 days after irradiation without any modality or dose dependency, while the anaplastic and chronic hypoxic AT1-tumor revealed an earlier and stronger treatment response after 12C-ion irradiation as compared to the more delayed response after photon irradiation. Again, no dose dependency was detected. Finally, a longitudinal histology study after irradiation of the AT1-tumor with curative doses of either photons or 12C-ions revealed that hypoxia developed slightly faster after 12C-ion than after photon irradiation. Furthermore, this study validated the relative biological effectiveness (RBE) for 12C-ions, which was determined previously for the endpoint local tumor control, on a microscopic level within the first 10 days. Additionally, reasonable time points for a future multimodal imaging study with PAI, sequential positron emission tomography (PET) and DCE-MRI measurements as well as histology were determined. In conclusion, this thesis proved PAI and the novel analysis protocol to be a feasible method for the characterization of the three Dunning tumor sublines with respect to their oxygenation. The different sensitivities of the HI- and AT1-tumors towards the two irradiation modalities indicate that the irradiation-induced vascular response depends on the structural-functional status of the tumor vasculature. The dose-independent response of both tumor sublines towards the two irradiation modalities suggests that the initial vascular response only plays a minor role with respect to local tumor control at high single doses.