A comparison of dose calculations at points around an intracavitary cervix applicator

An intercomparison was made between dose-rate calculations performed in 10 institute in The Netherlands for intracavitary applications with the Selectron-LDR remote afterloading machine. The results of 11 computer planning systems of five different manufacturers were compared with reference dose-rate calculations. The difference in clinically relevant dose-points like rectum and bladder, were less than 2% if the values of the coordinates of source and dose-points were given, except for one type of planning system, that showed differences up to 10%. The errors observed for the latter system were probably due to a coarse calculation grid, which becomes important in regions with a high dose gradient. If the reconstruction process of source and dose-point positions from radiographs was included as it usually is in practice, then the accuracy became worse and the planning system that showed large errors before, gave unacceptable errors of 30% to 40% in the rectal area. This is due to additional errors introduced in the reconstruction. The intercomparison showed that it is very important that computer treatment planning systems for brachytherapy calculations are tested before clinical use. The reconstruction technique for source and dose-point positions should be part of this test. Radiographs should be made of a phantom that simulates the cervix application, and the radiography technique should be similar to the one that is clinically employed.