[en] Purpose: To evaluate the adequacy of tumor volume coverage using a three dimensional (3D) margin growing algorithm compared to a two dimensional (2D) margin growing algorithm in the conformal radiotherapy planning of prostate cancer. Methods and Materials: Two gross tumor volumes (GTV) were segmented in each of ten patients with localized prostate cancer: prostate gland only (PO) and prostate with seminal vesicles (PSV). A margin of 10 mm was applied to these two groups (PO and PSV) using both the 2D and 3D margin growing algorithms. The true planning target volume (PTV) was defined as the region delineated by the 3D algorithm. Adequacy of geometric coverage of the GTV with the two algorithms was examined throughout the target volume. Discrepancies between the two margin methods were measured in the transaxial plane. Results: The 2D algorithm underestimated the PTV by 17% (range 12-20) in the PO group and by 20% (range 13-28) for the PSV group when compared to the 3D algorithm. For both the PO and PSV groups, the inferior coverage of the PTV was consistently underestimated by the 2D margin algorithm when compared to the 3D margins with a mean radial distance of 4.8 mm (range 0-10). In the central region of the prostate gland, the anterior, posterior, and lateral PTV borders were underestimated with the 2D margin in both the PO and PSV groups by a mean of 3.6 mm (range 0-9), 2.1 mm (range 0-8), and 1.8 (range 0-9) respectively. The PTV coverage of the PO group superiorly was radially underestimated by 4.5mm (range 0-14) when comparing the 2D margins to the 3D margins. For the PSV group, the junction region between the prostate and the seminal vesicles was underestimated by the 2D margin by a mean transaxial distance of 18.1 mm in the anterior PTV border (range 4-30), 7.2 mm posteriorly (range 0-20), and 3.7 mm laterally (range 0-14). The superior region of the seminal vesicles in the PSV group was also consistently underestimated with a radial discrepancy of 3.3 mm (range 0-12). The maximum underestimation using the 2D algorithm occurred when the target volume angulated sharply to 90 deg. within successive adjacent slices resulting in transaxial plane differences of up to 20 and 55 mm respectively for the PO and PSV groups when compared to coverage by the 3D margin. This was most evident in the junction region of the PSV group. In this region, the 2D algorithm was inadequate, often not providing any margin (range 0-3 mm) in both the sagittal and coronal planes to the GTV compared to the 10 mm margin delineated with the 3D algorithm. Conclusion: This study illustrates the problem of assuming margins delineated in the transaxial plane are adequate to cover a 3D target volume. An appreciation of spatial margins in 3D is required if 2D margin growing algorithms are used. If 2D margin methods are utilised, beams eye view evaluations are required in sagittal and coronal planes to ensure adequate margin and coverage of the target volume

No abstract available

[en] A treatment-planning case study has been performed on a patient with a medium-sized, convex brain tumour. The study involved the application of advanced treatment-plan optimization techniques to improve on the dose distribution of the 'standard plan' used to treat the patient. The standard plan was created according to conventional protocol at the Royal Marsden NHS Trust, and consisted of a three-field (one open and two wedged) non-coplanar arrangement, with field shaping to the beam's-eye view of the planning target volume (PTV). Three optimized treatment plans were created corresponding to (i) the optimization of the beam weights and wedge angles of the standard plan, (ii) the optimization of the beam orientations, beam weights and wedge angles of the standard plan, and (iii) a full fluence tomotherapy optimization of 1 cm wide (at isocentre), 270 deg. arcs. (i) and (ii) were created on the VOXELPLAN research 3D treatment-planning system, using in-house developed optimization algorithms, and (iii) was created on the PEACOCK tomotherapy planning system. The downhill-simplex optimization algorithm is used, in conjunction with 'threshold-dose' cost-function terms enabling the algorithm to optimize specific regions of the dose-volume histogram (DVH) curve. The 'beam-cost plot' tool is presented as a visual aid to the selection of beneficial beam directions. The methods and pitfalls in the transfer of plans and patient data between the two planning systems are discussed. Each optimization approach was evaluated, relative to the standard plan, on the basis of DVH and dose statistics in the PTV and organs at risk (OARs). All three optimization approaches were able to improve on the dose distribution of the standard plan. The magnitude of the improvement was greater for the optimized beam-orientation and tomotherapy plans (up to 15% and 30% for the maximum and mean OAR doses). A smaller improvement was observed in the beam-weight and wedge-angle optimized plan (up to 5% and 10% in the maximum and mean OAR doses). In the tomotherapy plan, difficulty was encountered achieving an acceptable homogeneity of dose in the PTV. This was improved by treating the gross tumour volume (GTV) and (PTV-GTV) regions as separate targets in the inverse planning, with the latter region prescribed a slightly higher dose to reduce edge under-dosing. In conclusion, for the medium-sized convex tumour studied, the tomotherapy dose distribution showed a significant improvement on the standard plan, but no significant improvement over a conventional three-field plan where the beam orientations, beam weights and wedge angles had been optimized. (author)

No abstract available