No abstract

The purpose of our study was to examine the potential benefits of integrating functional MRI (fMRI) information into the 3D-based planning process for central nervous system (CNS) malignancies. Between 01.01.2008 and 01.12.2009, ten patients with astrocytoma (both low and high-grade histological type) were enrolled in this study. Before the planning process, conventional CT planning, postoperative MR, and individual functional MRI examinations were conducted. For the functional MRI examination four types of conventional stimuli were applied: acoustic, visual, somatosensory, and numeric. To examine the potential benefits of using fMRI-based information, three different types of theoretical planning were applied and compared: 3D conformal plan without fMRI information, 3D conformal plan with fMRI information, and IMRT plan with fMRI information. DVH analysis and the NTCP model were used for plan comparison. When comparing planning methods, distance-related subgroups were generated and studied. By using the additional fMRI information, a significantly higher sparing effect can be achieved on these ORs (both with conventional 3D-based planning and IMRT). In cases when the OR-PTV distance is less than 1 cm, IMRT seems to be a significantly better choice than conventional 3D-based techniques. IMRT also has an additional sparing effect on the optic tract and brainstem, especially for locations close to the midline. Our results demonstrated that using fMRI information in conventional 3D-based treatment planning has the potential benefit of significant dose reduction for the critical organs, with no compromise in PTV coverage even when using conventional 3D planning. fMRI can be widely used in low-grade cases (long life expectancy, lower acute and late toxicity) and also in cases with high-grade astrocytomas or distant metastases (higher dose to PTV with better sparing of risk organs). In cases when the OR-PTV distance is less than 1 cm, IMRT should be the choice of treatment for a higher sparing effect on functional active areas. Longer imaging and clinical follow up are needed to confirm the real sparing effect on these functional areas.

Dynamic MR is a sensitive and well tolerated method for tumor motion monitoring in high precision 3D therapy planning of lung cancer patients. Our results demonstrate that tumors located in the upper and mid lobes have moderate breath synchronous movements. The greatest deviation occur in cranio-caudal direction.

In the treatment of early stage breast cancer breast conserving surgery (BCS) followed by whole breast irradiation (WBI) is a standard method. The impact of the tumor bed boost following WBI is well-defined, but there are various delivery methods. In this study the electron and the photon boost techniques were compared. For 78 early stage breast cancer patients both CT based 3D conformal photon boost and electron boost plans were created. For dosimetric comparison coverage index (CI), external volume index (EI) and conformality index (COIN) were studied. Lung volume receiving a dose of 2 Gy was also reviewed. Seventy-eight patients with 156 plans were compared. The mean tumor bed volume was measured as 61.39 cm3 the mean tumor bed-skin distance was 3.13 cm. In the case of CI and COIN significant differences were found in favor of the photon boost. In the comparison of EI no significant difference was detected between the two techniques. The mean lung volume receiving 2 Gy were 42.3 and 168.35 cm3, for photons and electrons respectively. In the adjuvant treatment of early stage breast cancer WBI followed by conformal photon boost showed to be superior to electron boost in focus of the COIN and CI.

Magnetic resonance imaging (MRI) provides superior visualization of the prostate, its substructure, surrounding tissues, and, most important, focal lesions or cancer. The purpose of our canine study was to demonstrate the feasibility of a low-field (0.35 T) transperineal system that enables precise MR image guidance of prostate interventions. The canines were placed in the right lateral decubitus position. Template reconstruction, trajectory planning, contouring were based on T2-weighted FSE images. For image guidance and target confirmation, fast gradient spoiled-echo (FSPGR) sequence was used. MR compatible coaxial needles were manually inserted through the perineum to the base of the prostate. After satisfactory position was confirmed, brachytherapy catheters were placed through the coaxial needles. The mean deviation of the needle displacements was 2.9 mm with a median value of 2.7 mm. 97% of the errors were less than 4.0 mm. The needle placement accuracy was modelled by the Rayleigh distribution with a sigma value of 2.3 mm. Visual confirmation of needle placements was demonstrated on pathology tissue slices. The time needed for each step was: anaesthesia - 15 min, setup and positioning - 15 min, initial imaging - 15 min, template registration, projection - 15 min, contouring, trajectory planning, insertion of 12 needles - 60 min Based on our canine experiences our method seems to be a promising approach for performing feasible, accurate, reliable and high-quality prostate MR guidance within a reasonable time span.