[en] Purpose: The binary multileaf collimator (MLC) is one of the most important components in helical tomotherapy (HT), as it modulates the dose delivered to the patient. However, methods to ensure MLC quality in HT treatments are lacking. The authors obtained data on the performance of the MLC in treatments administered in their department in order to assess possible delivery errors due to the MLC. Correction methods based on their data are proposed. Methods: Twenty sinograms from treatments delivered using both of the authors HT systems were measured and analyzed by recording the fluence collected by the imaging detector. Planned and actual sinograms were compared using distributions of leaf open time (LOT) errors, as well as differences in fluence reconstructed at each of the 51 projections into which the treatment planning system divides each rotation for optimization purposes. They proposed and applied a method based on individual leaf error correction and the increase in projection time to prevent latency effects when LOT is close to projection time. In order to analyze the dosimetric impact of the corrections, inphantom measurements were made for four corrected treatments. Results: The LOTs measured were consistent with those planned. Most of the mean errors in LOT distributions were within 1 ms with standard deviations of over 4 ms. Reconstructed fluences showed good results, with over 90% of points passing the 3% criterion, except in treatments with a short mean LOT, where the percentage of passing points was as low as 66%. Individual leaf errors were as long as 4 ms in some cases. Corrected sinograms improved error distribution, with standard deviations of over 3 ms and increased percentages of points passing 3% in the fluence per angle analysis, especially in treatments with a short mean LOT and those that were more subject to latency effects. The minimum percentage of points within 3% increased to 86%. In-phantom measurements of the corrected treatments showed that, while treatments affected by latency effects were improved, those affected by individual leaf errors were not. Conclusions: Measurement of MLC performance in real treatments provides the authors with a valuable tool for ensuring the quality of HT delivery. The LOTs of MLC are very accurate in most cases. Sources of error were found and correction methods proposed and applied. The corrections decreased the amount of LOT errors. The dosimetric impact of these corrections should be evaluated more thoroughly using 3D dose distribution analysis.

[en] Radiation therapy plays a central role in the management of many childhood malignancies and Helical Tomotherapy (HT) provides potential to decrease toxicity by limiting the radiation dose to normal structures. The aim of this article was to report preliminary results of our clinical experience with HT in pediatric malignancies. In this study 66 consecutive patients younger than 14 years old, treated with HT at our center between January 2006 and April 2010, have been included. We performed statistical analyses to assess the relationship between acute toxicity, graded according to the RTOG criteria, and several clinical and treatment characteristics such as a dose and irradiation volume. The median age of patients was 5 years. The most common tumor sites were: central nervous system (57%), abdomen (17%) and thorax (6%). The most prevalent histological types were: medulloblastoma (16 patients), neuroblastoma (9 patients) and rhabdomyosarcoma (7 patients). A total of 52 patients were treated for primary disease and 14 patients were treated for recurrent tumors. The majority of the patients (72%) were previously treated with chemotherapy. The median prescribed dose was 51 Gy (range 10-70 Gy). In 81% of cases grade 1 or 2 acute toxicity was observed. There were 11 cases (16,6%) of grade 3 hematological toxicity, two cases of grade 3 skin toxicity and one case of grade 3 emesis. Nine patients (13,6%) had grade 4 hematological toxicity. There were no cases of grade 4 non-hematological toxicities. On the univariate analysis, total dose and craniospinal irradiation (24 cases) were significantly associated with severe toxicity (grade 3 or more), whereas age and chemotherapy were not. On the multivariate analysis, craniospinal irradiation was the only significant independent risk factor for grade 3-4 toxicity. HT in pediatric population is feasible and safe treatment modality. It is characterized by an acceptable level of acute toxicity that we have seen in this highly selected pediatric patient cohort with clinical features of poor prognosis and/or aggressive therapy needed. Despite of a dosimetrical advantage of HT technique, an exhaustive analysis of long-term follow-up data is needed to assess late toxicity, especially in this potentially sensitive to radiation population

[en] The purpose of this paper is to determine the correlation between dose-volume histogram (DVH) and dose wall-histogram (DWH) in the evaluation of rectal complications for prostate cancer patients treated with three-dimensional conformal radiotherapy (3D-CRT). A retrospective analysis of DVHs and DWHs of a subset of 25 prostate cancer patients treated with 3D-CRT was performed. For every patient the rectum and the rectal wall (inner and outer surface) were contoured. Median ICRU radiation dose of 79.4 Gy was administered. Correlation between DVHs and DWHs parameters was investigated by the nonparametric Spearman test and by linear regression analysis. The results showed a statistically significant linear correlation between pairs of DVH and DWH dosimetric parameters with Spearman correlation values (S) bigger than 0.8, with p values better than 0.0005 (two-sided) when the emptied rectum is considered. The variation of S and linear fit slope values [b(1)] showed a very similar functional shape with a minimum at 91% ICRU dose [S=0.83,b(1)=0.65]. The present study confirms a high correlation (>80%) between DVH and DWH of the rectum following 3D-CRT for prostate cancer. The derived advantage is that the contouring of inner surface of rectum could be obviated in almost 90% of patients when performing predictive models for rectal complications based on dosimetric variables under the standard treatment conditions specified in this study