engleski

Comparison of three calculation algorithms for post- operative radiotherapy of breast cancer

Introduction: In Croatia, breast cancer makes 11.1% of all cancers and it is the most common diagnosed cancer in females (25%). It is reported that 63% of breast cancer patients require radiotherapy as part of their primary treatment. The quality of radiation therapy depends on the ability to optimize the ratio of tumour control probability and normal tissue complication probability. This ratio is directly related to the absorbed dose distribution and the accuracy of its calculation and delivery. Dose calculation accuracy varies significantly in the presence of tissue inhomogeneities. Motivation for this work was to investigate the influence of three different calculation algorithms on dose distribution in breast cancer irradiation. Methods and Materials: Retrospective analysis of dose distributions calculated using three different calculation algorithms for fifteen patients requiring post-operative left breast radiotherapy was carried out. Treatment method was whole breast irradiation with total prescribed dose of 5000cGy in 25 fractions. Planning technique was forward IMRT using field-in-field technique. Dose distributions were created in Elekta Monaco treatment planning system (TPS) using Collapsed Cone Convolution (CCC) calculation algorithm following international guidelines. Absorbed dose distributions were calculated so that 100% of prescribed dose covers ≥80% of PTV. Pre-treatment dosimetric verification was performed for each patient. Dose distributions for fifteen patients were retrospectively recalculated using Monte Carlo (MC) (Elekta, Monaco) and Anisotropic Analytical Algorithm (AAA) (Varian, Eclipse) calculation algorithm and compared to CCC. For all three algorithms, 6MV beam of the same linear accelerator was used. Gantry angles, field number and field shapes were kept constant. Dose distributions were recalculated to achieve the same planning target volume (PTV) coverage with 100% isovolumes in original CCC plan. Parameters related to PTV and OARs were analyzed. Results and Discussion: The average number of therapy fields was 7.73 per patient. The average coverage of PTV with D100% was 85.23%. To achieve the same PTV coverage, it was observed that Dmax increases when using MC (109, 64%, p=0.000000) and AAA (107.69%, p=0.000876) as compared to CCC (106.84%). Looking at D95%, lower coverage of 97.49% was obtained using AAA which was shown to be statistically significant different with p=0.035551. Considering absorbed dose to ipsilateral lung, dose increases when AAA is applied, however it is not statistically significant. This difference could be attributed to the fact that MC and CCC algorithms are based on dose-to-medium (Dm) calculation formalism as opposed to dose-to-water (Dw) used by AAA algorithm. No statistically significant difference was found when comparing doses to the heart. Difference between average numbers of monitor units (MUs) was also found to be statistically insignificant. Conclusion: Retrospective analysis of absorbed dose distributions for 15 patients using three calculation algorithms showed statistically significant difference only in Dmax and D95%. Differences for other parameters were small and therefore any of these algorithms can be used as an algorithm of choice for post-operative radiotherapy of breast cancer. However, for AAA, lower PTV dose coverage and higher Dmax is observed which should be taken into consideration.

calculation algorithms, absorbed dose distributions, breast cancer radiotherapy

nije evidentirano