Brachytherapy of Squamous Cell Carcinoma in the Lip MDR to PDR Treatment Conversion

Abstract: Introduction: From 1988 – 2004, patients with squamous cell carcinoma in the lower lip were treated with medium dose rate (MDR) 192Ir wires using a manual afterloading tech-nique. An average of 3 wires were used, each with a mean length of 36 mm (range 27 – 65 mm) and a mean dose rate of 4.5 Gy/h (range 1.8 – 8.7 Gy/h). The average prescribed dose was 21.5 Gy (range 21.0 – 32.6 Gy) at the 85% isodose of the basal dose point according to the Paris system. The main purpose of this study is to investigate the equivalence of dose distribution when replacing a manually afterloaded wire with a remotely controlled stepping source. This treatment conversion is considered primarily because of improve-ments in staff radiation protection and hospital logistics. Materials and methods: The 192Ir wire dose distribution plans were previously calculated in Nucletron Planning system, module MPS v11.33® (NPS). These plans were compared to plans with a reconstructed wire (RW) that is planned and calculated in PLATO, module BPS v14.2.3® (PLATO). The RW is a catheter of 36 mm that contains a 192Ir stepping source with equally weighted dwell times separated by either 2.5, 5, 7.5 or 10 mm step length. The comparison of dose distribution difference between the RW plans and the actual wire measurements, made with GAFCHROMIC® EBT2 film, was performed using an in house program developed in MATLAB®. Results: For the midplane parallel to the wire, the prescribed isodose differs just above 2% at equal distances and the distance less than 0.14 mm at equal isodoses for angles smaller than 50º. For the midplane perpendicular to the wire, the difference between the two TPS is just above 1% and less than 0.07 mm for all angles. These results are for RW-4, which is the RW that shows worst agreement in the dose calculation difference. The RW and wire dose distribution is in better agreement at planes further away from the sources. The difference at 1 mm from the midplane is approximately ±20%, about the same difference at 6 mm and less than ±15% at 11 mm evaluation distance. The same trend can be seen no matter what RW is being used. There is an obvious difference the closer the point of interest is to the source. Further away from the source the dose distribution gets smeared out making it difficult to get any conclusive results. Results from the ratio between dose values show a better compliance close to the source, which indicates a large angular depen-dence. A correction of the dwell weights results in better agreement close to the source. Conclusion: The difference between NPS and PLATO are mainly because of the differ-ence between AL and EL but also due to the approximations made by NPS, which are insufficient in capturing the change in source specific attenuation properties. There is a general under dosage outside the catheter and cold spots between source positions for longer step lengths but also hot spots at the actual dwell positions. The differences in dose distribution are mainly due to that the AL is not the same as the EL. Another reason is that equal dwell weights were used initially. The different distance of the point of interest will also have impact on the results; a better agreement closer to the source. A better agreement in dose distribution can be obtained by manipulating the dwell weights.