The requirement of 5 % overall accuracy for the target-absorbed dose in radiotherapy implies that the accuracy of the relative dose calculation should be only within a few percent. According to the recommendation of the International Commission on Radiation Units and Measurements (ICRU), a computer generated dose distribution is considered accurate enough if it differs from the results of relative dose measurements by less than 2%, or 2 mm in the position of isodose curves involving very steep dose gradients. Considering these requirements, eight treatment planning systems of 27 systems, currently used in Czech hospitals, were compared. The intercomparison was made using a multi-purpose phantom involving dose measurements based on thermoluminescent (TLD) and film dosimetry. Seven typical cases of irradiation were tested: regular fields, irregular fields, wedge fields, oblique incidence and non-homogenities. The absorbed dose measured by TLD was compared with the absorbed dose calculated by the treatment planning systems. The results indicate that the dose distributions generated by different treatment planning systems can differ from each other as well as from the measured dose distributions up to a level not acceptable in terms of the ICRU requirement. The method using the multipurpose phantom seems to be a good tool for regular independent quality control of the treatment planning systems since it shows the real dose outcome resulting from all procedures involved in the radiotherapy chain.

        Key words: radiotherapy - treatment planning system - quality assurance - TLD