G. Meijer

Three-dimensional analysis of delineation errors, setup errors, and organ motion during radiotherapy of bladder cancer

PURPOSE To quantify in three dimensions the geometric uncertainties of bladder irradiation (i.e., uncertainties in target delineation, organ motion, and patient setup). METHODS AND MATERIALS Pelvic CT images were obtained for 10 male bladder cancer patients. Apart from the initial planning CT scan, three follow-up scans were made for each of the patients. The bladder volumes in the planning CT scan were outlined by seven radiation oncologists. One observer also delineated the bladder volumes in the follow-up scans. Two-dimensional scalar maps of the interobserver variation and organ motion of the bladder surfaces were constructed. The setup errors were derived from the portal imaging results of the pooled group of bladder and prostate patients. RESULTS All bladder volumes were consistently outlined by all observers. Generally small variations occurred (1.5-3 mm, 1 SD), although in 50% of the patients, larger discrepancies were observed in discriminating the bladder from the base of the prostate. Analysis of the portal imaging data showed setup errors of up to 3 mm (1 SD). Organ motion is the predominant geometric uncertainty in the radiotherapy process (5 mm, 1 SD, at the cranial side of the bladder), although 9 of 10 patients were able to preserve a fairly reproducible bladder volume during the complete treatment course. CONCLUSION Anisotropic margins between the clinical target volume and planning target volume are needed in conformal radiotherapy of the bladder. Especially at the cranial side of the bladder, larger margins are needed because of the impact of bladder shape variation.

Dose-wall histograms and normalized dose-surface histograms for the rectum : A new method to analyze the dose distribution over the rectum in conformal radiotherapy

PURPOSE To develop an accurate method to generate a dose-volume histogram (DVH) of the rectum wall, solely based on the outer contours of the rectum wall. METHODS AND MATERIALS A mathematical model for the rectum wall is developed, incorporating the stretching of the rectum wall due to variable rectal filling and neighboring structures. The model is based on the assumption that the amount of intersected rectum wall tissue normal to the central axis of the rectum is constant. The main objective of the model is to determine the thickness of the rectum wall in each wall element. Two approaches are described, each yielding a DVH of the rectum wall, based only on the delineated outer contours of the rectum. In the first approach, the model is used to create a set of inner contours out of the axial outer contours. Both sets of contours are used to derive a dose-wall histogram (DWH) of the rectum. In the second approach, the model is used to generate a normalized 2D sampling space, which is subsequently binned into a normalized dose-surface histogram (NDSH). The model is verified using 20 sets of CT data (5 patients x 4 scans) in which both outer and inner contours of the rectum are carefully delineated. The DWHs and NDSHs are compared with DVHs of the rectum wall, which require contouring of the outer and inner surfaces of the rectum wall, and with DVHs of the total rectum (including rectal filling). The variation between DWHs, NDSHs, and DVHs is investigated using normal tissue complication probability (NTCP) calculations. RESULTS The local wall thickness of the rectum as outlined on CT data was in conformity with the described rectum model. The amount of rectum wall tissue per unit length rectum varied considerably between patients (27%, 1 SD). In all analyzed patients, the DWHs and NDSHs corresponded well to the DVHs of the rectum wall. Much more discrepancies were observed between the DVHs of the total rectum and the DVHs of the rectum wall. CONCLUSION The applied methods yield accurate dose distributions of the rectum wall, without delineating the inner surface of the rectum. This reduces both the workload and variations due to inaccurate delineation of the rectum wall. The DWH and NDSH are effective tools to evaluate 3D dose distributions of the rectum wall and to estimate the complication probability of the rectum in high-dose conformal radiotherapy.

What CTV-to-PTV Margins Should Be Applied for Prostate Irradiation? Four-Dimensional Quantitative Assessment Using Model-Based Deformable Image Registration Techniques

PURPOSE To quantify adequate anisotropic clinical target volume (CTV)-to-planning target volume (PTV) margins for three different setup strategies used during prostate irradiation: (1) no setup corrections, (2) on-line corrections determined from bony anatomy, and (3) on-line corrections determined from gold markers. METHOD AND MATERIALS Three radiation oncologists independently delineated the CTV on computed tomography images of 30 prostate cancer patients. Eight repeat scans were acquired to allow simulation of the delivered dose distributions in changing geometry. Different registration approaches were taken to mimic the different setup strategies. A surface model-based deformable image registration system was used to warp the delivered dose distributions back to the dose in the planning computed tomography scan. On the basis of the geometric extent of the underdosed areas, a set of anisotropic margins was derived to ensure a minimal dose to the CTV of 95% for 90% of the patients. RESULTS Without setup correction, margins of approximately 11 mm for the corpus of the prostate and 15 mm for the seminal vesicles were required. These margins could be reduced to 8 and 13 mm when aligning the patient to the bony anatomy and to 3 and 8 mm aligning the patient to implanted gold markers. A larger margin at the apex was required to account for the significant observer variability and steep dose gradients at this location (11 mm using skin marker registration, 9 mm using bony anatomy registration, and 6 mm using gold marker registration). CONCLUSIONS Novel voxel tracking techniques have enabled us to calculate accumulated dose distributions and design accurate three-dimensional CTV-to-PTV margins for prostate irradiation.

High precision bladder cancer irradiation by integrating a library planning procedure of 6 prospectively generated SIB IMRT plans with image guidance using lipiodol markers

PURPOSE To increase local control and decrease side effects for urinary bladder cancer patients by integrating a library planning procedure with image guidance using lipiodol markers. METHODS AND MATERIALS Twenty patients with T2-T4N0M0 grade 2-3 invasive bladder carcinoma were treated according to an online adaptive protocol. Initially, the gross tumour volume (GTV) was demarcated during cystoscopy by injecting several drops of lipiodol in the submucosa around the tumour. Subsequently two CT scans were acquired with a full bladder and a voided bladder. On both scans, the boost volume (GTV) and the low-risk bladder volume were delineated. Using an interpolation tool, six concomitant boost IMRT plans with increasing bladder volumes were generated. For each fraction the procedure at the treatment unit was as follows: Firstly, a ConeBeam-CT was acquired and based on the amount of bladder filling the best fitting bladder contours and corresponding GTV and IMRT plans were selected. Secondly, the lipiodol markers were registered using the corresponding GTV contours and it was verified that the corresponding 95%-isodose surface covered the entire bladder. Finally, an online setup correction was applied based on this registration and the corresponding treatment plan was irradiated. RESULTS The lipiodol markers were very useful in outlining the GTV at the planning CT and for daily setup correction. While the patients strived for a full bladder filling at time of the treatment, this was seldom accomplished. Due to our protocol an appropriate plan with adequate coverage of the PTV and without excessive dose to healthy tissue was delivered every day. The treatment was very well tolerated by all patients. At the end of the treatment no grade 3 urinary or gastro-intestinal toxicity was observed. After a median follow-up of 28 months two local relapses occurred. CONCLUSION Using the library planning approach combined with online image guidance using lipiodol markers, we were able to deliver a highly conformal dose distribution to all bladder cancer patients achieving promising clinical results.

Comparison of prostate cancer treatment in two institutions: a quality control study

PURPOSE To minimize differences in the treatment planning procedure between two institutions within the context of a radiotherapy prostate cancer trial. PATIENTS AND METHODS Twenty-two patients with N0 M0 prostate cancer underwent a computed tomography (CT) scan for radiotherapy treatment planning. For all patients, the tumor and organs at risk were delineated, and a treatment plan was generated for a three-field technique giving a dose of 78 Gy to the target volume. Ten of the 22 cases were delineated and planned in the other institution as well. The delineated volumes and dose distributions were compared. RESULTS All treatments fulfilled the trial criteria. The mean volume ratio of the gross tumor volumes (GTVs) in both institutions was 1.01, while the mean volume ratio of the planning target volumes (PTVs) was 0.88. The three-dimensional (3D) PTV difference was 3 mm at the prostate apex and 6-8 mm at the seminal vesicles. This PTV difference was mainly caused by a difference in the method of 3D expansion, and disappeared when applying an improved algorithm in one institution. The treated volume (dose > or =95% of isocenter dose) reflects the size of the PTV and the conformity of the treatment technique. This volume was on average 66 cm3 smaller in institution A than in institution B; the effect of the PTV difference was 31 cm3 and the difference in technique accounted for 36 cm3. The mean delineated rectal volume including filling was 112 cm3 and 125 cm3 for institution A and B, respectively. This difference had a significant impact on the relative dose volume histogram (DVH) of the rectum. CONCLUSION Differences in GTV delineation were small and comparable to earlier quantified differences between observers in one institution. Different expansion methods for generation of the PTV significantly influenced the amount of irradiated tissue. Strict definitions of target and normal structures are mandatory for reliable trial results.

In vivo dosimetry during conformal radiotherapy: Requirements for and findings of a routine procedure

PURPOSE Conformal radiotherapy requires accurate knowledge of the actual dose delivered to a patient. The impact of routine in vivo dosimetry, including its special requirements, clinical findings and resources, has been analysed for three conformal treatment techniques to evaluate its usefulness in daily clinical practice. MATERIALS AND METHODS Based on pilot studies, routine in vivo dosimetry quality control (QC) protocols were implemented in the clinic. Entrance and exit diode dose measurements have been performed during two treatment sessions for 378 patients having prostate, bladder and parotid gland tumours. Dose calculations were performed with a CT-based three-dimensional treatment planning system. In our QC-protocol we applied action levels of 2.5% for the prostate and bladder tumour group and 4.0% for the parotid gland patients. When the difference between the measured dose at the dose specification point and the prescribed dose exceeded the action level the deviation was investigated and the number of monitor units (MUs) adjusted. Since an accurate dose measurement was necessary, some properties of the on-line high-precision diode measurement system and the long-term change in sensitivity of the diodes were investigated in detail. RESULTS The sensitivity of all diodes decreased by approximately 7% after receiving an integrated dose of 10 kGy, for 4 and 8 MV beams. For 34 (9%) patients the difference between the measured and calculated dose was larger than the action level. Systematic errors in the use of a new software release of the monitor unit calculation program, limitations of the dose calculation algorithms, errors in the planning procedure and instability in the performance of the accelerator have been detected. CONCLUSIONS Accurate in vivo dosimetry, using a diode measurement system, is a powerful tool to trace dosimetric errors during conformal radiotherapy in the range of 2.5-10%, provided that the system is carefully calibrated. The implementation of an intensive in vivo dosimetry programme requires additional staff for measurements and evaluation. The patient measurements add only a few minutes to the total treatment time per patient and guarantee an accurate dose delivery, which is a prerequisite for conformal radiotherapy.

Accurate in vivo dosimetry of a randomized trial of prostate cancer irradiation

PURPOSE To guarantee an accurate dose delivery, within +/- 2.5%, in a Phase III randomized trial of prostate cancer irradiation (68 vs. 78 Gy) by means of a comprehensive in vivo dosimetry program. METHODS AND MATERIALS Prostate patients are generally treated in our clinic with a 3-field isocentric technique: an 8-MV anteroposterior beam and 2 18-MV wedged laterals. All fields are shaped conformally to the PTV. Patients were randomized between two dose levels of 68 Gy and 78 Gy. During treatment, the entrance and exit dose were measured for each patient with diodes. Special 2.5-mm thick steel build-up caps were applied to make the diodes appropriate for measurements in 18-MV photon beams as well. Portal images were used to verify the correct position of the diodes and to detect and correct for gas filling in the rectum that may influence the exit dose reading. Entrance and exit dose measurements were converted to midplane dose, which was used in combination with a depth dose correction to obtain the dose at the specification point. An action level of 2.5% was applied. RESULTS The added build-up for the diodes in the 18-MV beams resulted in correction factors that were only slightly sensitive to changes in beam setup and comparable to the corrections used in the 8-MV beams for diodes without extra build-up. The calibration factor increased almost linearly with cumulative dose: 0.7%/kGy for the 8-MV and 1.2%/kGy for the 18-MV photon beams. The introduction of average correction factors made the analysis easier, while keeping the accuracy within acceptable limits. In a period of 3 years, 225 patients were analyzed, from which 8 patients needed to be corrected. The average ratio of measured and prescribed dose was 1.009 (standard deviation [SD] 0.012) for the total group treated on two linear accelerators. When the results were analyzed per accelerator, the ratios were 1.002 (SD, 0.001) for Accelerator A and 1.015 (SD, 0.001) for Accelerator B. This difference could be attributed to the cumulative effect of three small imperfections in the performance of Accelerator B that were well within the limits of our quality assurance program. CONCLUSION Diodes can be used for accurate in vivo dosimetry during prostate irradiation in high-energy photon beams. The dose delivery in this randomized trial is guaranteed within the 2.5% limits on an individual patient basis. This could not be achieved without the in vivo dosimetry program, despite our high-standard quality assurance program of treatment delivery.

Consistency in quality control programmes for electron accelerators in radiotherapy centres

BACKGROUND AND PURPOSE To gain insight into the current practice of quality control (QC) of medical electron accelerators and to reduce possible variations in test frequencies and test procedures. MATERIALS AND METHODS An extensive questionnaire on QC procedures of medical electron accelerators was distributed and completed by all (21) radiotherapy institutions in The Netherlands. The questions were related to safety systems, mechanical parameters, beam profiles, beam energy, absolute dosimetry, wedge filters, the dose monitor system and radiation leakage. The data of the questionnaire were compared with recommendations given in national and international reports on QC of electron accelerators. RESULTS Large variations in time spent on QC exist, especially for accelerators having dual energy photon beams and several electron beam energies. This diversity is mainly due to differences in philosophy with regard to QC and the differences in resources and machine time available. Furthermore, large variations in test frequencies and test methodologies were observed. The staffing level involved in the QC measurements was evaluated and compared with recent recommendations provided by EFOMP-ESTRO. CONCLUSIONS From these recommendations and the results of the questionnaire, a set of minimum guidelines for a QC programme could be formulated and implemented in all radiotherapy institutions in The Netherlands.

A treatment planning method to correct dose distributions distorted by setup verification fields.

PURPOSE Portal images of conformal treatment fields are often not suitable for setup verification purposes because they contain insufficient bony structures. Therefore, additional rectangular fields are frequently applied for setup verification purposes. It is the aim of this study to reduce the dose distortions induced by these extra fields by appropriately adjusting the beam weights and wedge angles of the treatment fields. METHODS AND MATERIALS A second treatment plan for the setup verification session is generated, with an identical beam setup as the original plan, but which also includes two orthogonal setup verification fields. An algorithm has been developed, based on vector analysis methods, that adjusts the beam weights and wedge angles of the treatment fields in such a way that both the dose at the isocenter and the dose homogeneity over the planning target volume (PTV) are conserved. RESULTS The algorithm has been applied to three clinical cases. The number of MUs for the setup verification fields, using a liquid-filled electronic portal imaging device, varied between 16 MU in the head and neck region up to 34 MU for lateral images in the pelvic region. In all cases, the method yielded a treatment plan including two orthogonal setup verification fields with a similar dose distribution over the PTV as the original treatment plan without the setup verification fields. CONCLUSION The dose distortions resulting from the acquisition of orthogonal verification imaging can be neutralized by modifying the original beam weights and wedge angles of the treatment fields.

PD-0283: 4D dose accumulation for dose painting by numbers for lung cancer

In conventional radiotherapy of locally advanced lung cancer (LALC) doses levels are homogeneously delivered to the entire PTV, whereat dose escalation is restricted by normal tissue toxicity. Several studies have shown the geometrical correlation between high FDG uptake in a PET scan and tumour recurrence. This is the rationale for FDG-based local dose escalation, e.g. by dose prescription on the voxel values of a PET scan – dose painting by numbers (DPBN). The aim of this study is to investigate the robustness of the DPBN plans against tumour motion