Purpose: Since 1991 the use of three-dimensional planned radiation therapy for rectal cancer was evaluated as a part of a clinical project funded by the Deutsche Krebshilfe at the Radiological Department of the University of Heidelberg. First clinical experiences are reported.
Patients and methods: 32 patients who were irradiated loco-regional for rectal cancer were analysed retrospectively. Radiation therapy techniques: patient's position was prone by the use of a belly board. Three-dimensional treatment (3D) planning system was voxelplan. The radiation therapy was started using a 3-field-box technique modified by 3D-technique after 22 to 42 Gy, depending on therapeutic concept.
Results: On average 92% of the planning target volume were encompassed with the 80% isodose. The average maximum dose was 108% (isocenter = 100%), of the small bowel volume received less than 34% of the dose, 70% of the bladder volume less than 40% respectively. Total average dose was 52.4% Gy. Irregular fields were used in all patients, a non-coplanar field was used to optimize dose-distribution in 41% of the patients, 21 patients received chemotherapy simultaneously. Acute side effects according to gastrointestinal tract reached grade 1 (WHO) in 9 patients, grade 2 in 2 patients. Side effects according to bladder reached grade 1 in 5 patients grade 2 in 1 patient respectively. In 4 patients suffered from acute side effects concerning the bladder. In one patient acute side effects grade 1 concerning gastrointestinal tract was observed. Portal films were analysed to evaluate precision of radiation therapy. On average the error was 3.1/4.5/4.0 mm in the x/y/z direction respectively. The standard deviation was 4.4/6.8/6.3 mm for x/y/z respectively. A median time of 2 hours was necessary for all planning procedures.
Conclusions: Three-dimensional treatment planning optimizes dose-distribution in a relevant number of patients. Its clinical use for treatment of large targets or in order to applicate high doses is justified. Planning target volume should cover the clinical target volume with a margin of 1.5 cm.