Search

To evaluate the performance of a model-based optimisation process for volumetric modulated arc therapy, VMAT, applied to whole breast irradiation.A set of 150 VMAT dose plans with simultaneous integrated boost were selected to train a model for the prediction of dose-volume constraints. The dosimetric validation was done on different groups of patients from three institutes for single (50 cases) and bilateral breast (20 cases).Quantitative improvements were observed between the model-based and the reference plans, particularly for heart dose. Of 460 analysed dose-volume objectives, 13% of the clinical plans failed to meet the constraints while the respective model-based plans succeeded. Only in 5 cases did the reference plans pass while the respective model-based failed the criteria. For the bilateral breast analysis, the model-based plans resulted in superior or equivalent dose distributions to the reference plans in 96% of the cases.Plans optimised using a knowledge-based model to determine the dose-volume constraints showed dosimetric improvements when compared to earlier approved clinical plans. The model was applicable to patients from different centres for both single and bilateral breast irradiation. The data suggests that the dose-volume constraint optimisation can be effectively automated with the new engine and could encourage its application to clinical practice.

Abstract Background The study summarised in this report aimed to investigate the interplay between fluence complexity, dose calculation algorithms, dose calculation spatial resolution and delivery characteristics (monitor units, effective field width and dose delivery against dose prediction agreement) was investigated. A sample set of complex planning cases was selected and tested using a commercial treatment planning system capable of inverse optimisation and equipped with tools to tune fluence smoothness. Methods A set of increasingly smoothed fluence patterns was correlated to a generalised expression of the Modulation Index (MI) concept, in nature independent from the specific planning system used that could therefore be recommended as a predictor to score fluence "quality" at a very early stage of the IMRT QA process. Fluence complexity was also correlated to delivery accuracy and characteristics in terms of number of MU, dynamic window width and agreement between calculation and measurement (expressed as percentage of field area with a γ > 1 (%FA)) when comparing calculated vs. delivered modulated dose maps. Different resolutions of the calculation grid and different photon dose algorithms (pencil beam and anisotropic analytical algorithm) were used for the investigations. Results and Conclusion i) MI can be used as a reliable parameter to test different approaches/algorithms to smooth fluences implemented in a TPS, and to identify the preferable default values for the smoothing parameters if appropriate tools are implemented; ii) a MI threshold set at MI < 19 could ensure that the planned beams are safely and accurately delivered within stringent quality criteria; iii) a reduction in fluence complexity is strictly correlated to a corresponding reduction in MUs, as well as to a decrease of the average sliding window width (for dynamic IMRT delivery); iv) a smoother fluence results in a reduction of dose in the healthy tissue with a potentially relevant clinical benefit; v) increasing the smoothing parameter s, MI decreases with %FA: fluence complexity has a significant impact on the accuracy of delivery and the agreement between calculation and measurements improves with the advanced algorithms.

Purpose: In breast radiotherapy with intensity modulation, it is a well established practice to extend the dose fluence outside the limit of the body contour to account for small changes in size and position of the target and the rest of the tissues due to respiration or to possible oedema. A simple approach is not applicable with RapidArc volumetric modulated arc therapy not being based on a fixed field fluence delivery. In this study, a viable technical strategy to account for this need is presented.Methods: RapidArc (RA) plans for six breast cancer patients (three right and three left cases), were optimized (PRO version III) on the original CT data set (O) and on an alternative CT (E) generated with an artificial expansion (and assignment of soft-tissue equivalent HU) of 10 mm of the body in the breast region and of the PTV contours toward the external direction. Final dose calculations for the two set of plans were performed on the same original CT data set O, normalizing the dose prescription (50 Gy) to the target mean. In this way, two treatment plans on the same CT set O for each patient were obtained: the no action plan (OO) and the alternative plan based on an expanded optimization (EO). Fixing MU, these two plans were then recomputed on the expanded CT data set and on an intermediate one (with expansion = 5 mm), to mimic, possible changes in size due to edema during treatment or residual displacements due to breathing not properly controlled. Aim of the study was to quantify the robustness of this planning strategy on dose distributions when either the OO or the EO strategies were adopted. For all the combinations, a DVH analysis of all involved structures is reported.Results: I. The two optimization approaches gave comparable dose distributions on the original CT data set. II. When plans were evaluated on the expanded CTs (mimicking the presence of edema), the EO approach showed improved target coverage if compared to OO: on CT_10 mm, DV = 98% [%] = 92.5 ± 0.9 and 68.5 ± 3.1, respectively, for EO and OO. Minor changes were registered in organs at risk sparing for both EO and OO. III. From dose distributions and DVHs, EO approach allowed to irradiate at near to prescription levels also the expanded fraction of the target: this would account also for residual intrafraction movements.Conclusions: The proposed plan strategy could represent a robust approach to account for moderate changes in target or body volume during the course of breast radiotherapy and to account for residual intrafractional respiratory motion in volumetric modulated arc therapy. The strategy, logistically simple to implement requiring only modifications to the standard planning workflow was routinely implemented at author's institute for treatment of breast patients with RapidArc. [ABSTRACT FROM AUTHOR]

A study was carried out to evaluate the possibility of delivering volumetric modulated arc therapy with the RapidArc technology under respiratory-gated conditions. The experiments were performed in the framework of a non-clinically released environment. Plans of six patients, all realized for a single arc, were used for the experiments. The Real-time Position Management (RPM) respiratory gating system from Varian was used to generate gate-open signals of different durations. Arcs were delivered applying the different gates creating sequences of beam-hold/beam-on during the dose delivery: the average number of interruptions for a single arc ranged from 0 to 45. Dose prescription was set to 2 Gy and different gate-open periods of 30, 15 and 5 s to keep gantry speed constant at maximum. 5 Gy and 15 Gy doses were then applied to gate-open signals of 5 and 8 s, respectively, to mimic the most challenging conditions of slow gantry rotation and high-frequency interruptions. The 5 and 15 Gy experiments represent dose conditions of clinical interest for stereotactic treatments. For each patient and gating condition, pre-treatment 2D verification measurements were performed using the PTW-729 array in conjunction with the Octavius phantom (PTW, Freiburg); measurements were performed on different days (one per patient, with the complete setup of phantom and detectors every time), while each gating experiment was repeated seven consecutive times for reproducibility (without a new setup of the measurement equipment). Measurements were compared with dose calculations in the treatment planning system (performed without any gating) to appraise the dosimetric impact of the presence of gating and the eventual dependence from the number of interruptions during a single arc. Analysis of machine-registered log files in terms of average deviations between actual and expected positions (from automatic measurements every 50 ms) resulted in mean DMU (monitor units) <0.02% for all gating conditions. D(Gantry angle) = 0.38 +- 0.01deg for 2 Gy (all gate periods), 0.24 +- 0.01deg for 5 Gy, and 0.10 +- 0.01deg for 15 Gy deliveries. Average deviations for multileaf collimator (MLC) positions (root mean square over all 120 leaves) were 0.45 +- 0.01 mm for 2 Gy (all gate periods), 0.32 +- 0.01 mm for 5 Gy and 0.14 +- 0.01 mm for 15 Gy. Results in terms of dose measurements confirmed that the application of gating to RapidArc delivery does not affect the quality of the dose delivery. With criteria of DD = 3%, DTA = 3 mm, the gamma test was passing in a range of 99 to 100% of the measured points for most of the cases (with maximum number of interruptions of about 20 per arc) and from 97 to 98% for the extreme case of 15 Gy and 8 s gate-open signal (corresponding to almost 50 interruptions per arc). In conclusion, RapidArc delivery proved, in a pre-clinical phase and non-clinically released framework, to be reliable and dosimetrically accurate also when applied in conjunction with gating procedures. [ABSTRACT FROM AUTHOR]

A planning and dosimetric study was carried out on a cohort of six CT datasets from patients treated for prostate cancer to assess the impact of couch modelling on the accuracy of dose calculation for the volumetric modulated arc technique RapidArc. For each patient, RapidArc plans were optimized using the couch while final dose calculation was performed with different conditions (thin, medium, thick and no couch). Analysis was performed in terms of dose volume histograms, dose difference histograms and 3D-g tests. Pre-treatment verification measurements were performed using the PTW-729 array in conjunction with the Octavius phantom (PTW, Freiburg); similarly, HU characterization of couch was performed with the same phantom and ion chamber measurements comparing calculations and experimental data. A set of Hounsfield Units (HU) valid for low and high energy and the entire couch length was found as internal structure HU = [?]960, surface shell HU = [?]700. Analysis of dose plans showed that differences larger than 1.5 Gy for a 70 Gy prescription might be observed on significant fractions of PTVs. Smaller differences are visible in the medium low-dose regions. Pre-treatment verification on composite delivery confirmed these observations and, at the same time, showed good accuracy of dose calculations in the presence of couch modelling compared to delivery in the same conditions (GAI ranging from 95% to 100%). Results confirmed the reliability of the geometrical model build in the planning system Eclipse, and (i) there is no measurable effect if the wrong segment of the couch is used in the calculations; (ii) there are significant discrepancies of potential clinical impact at the level of the target volumes if calculations are performed without couch and delivery is performed with couch, and (iii) the effect is particularly relevant at low energy (6 MV in this case) that is the configuration clinically used by most of the centres adopting technologies based on intensity-modulated arcs. [ABSTRACT FROM AUTHOR]

Background: The study summarised in this report aimed to investigate the interplay between fluence complexity, dose calculation algorithms, dose calculation spatial resolution and delivery characteristics (monitor units, effective field width and dose delivery against dose prediction agreement) was investigated. A sample set of complex planning cases was selected and tested using a commercial treatment planning system capable of inverse optimisation and equipped with tools to tune fluence smoothness. Methods: A set of increasingly smoothed fluence patterns was correlated to a generalised expression of the Modulation Index (MI) concept, in nature independent from the specific planning system used that could therefore be recommended as a predictor to score fluence "quality" at a very early stage of the IMRT QA process. Fluence complexity was also correlated to delivery accuracy and characteristics in terms of number of MU, dynamic window width and agreement between calculation and measurement (expressed as percentage of field area with a γ > 1 (%FA)) when comparing calculated vs. delivered modulated dose maps. Different resolutions of the calculation grid and different photon dose algorithms (pencil beam and anisotropic analytical algorithm) were used for the investigations. Results and Conclusion: i) MI can be used as a reliable parameter to test different approaches/algorithms to smooth fluences implemented in a TPS, and to identify the preferable default values for the smoothing parameters if appropriate tools are implemented; ii) a MI threshold set at MI < 19 could ensure that the planned beams are safely and accurately delivered within stringent quality criteria; iii) a reduction in fluence complexity is strictly correlated to a corresponding reduction in MUs, as well as to a decrease of the average sliding window width (for dynamic IMRT delivery); iv) a smoother fluence results in a reduction of dose in the healthy tissue with a potentially relevant clinical benefit; v) increasing the smoothing parameter s, MI decreases with %FA: fluence complexity has a significant impact on the accuracy of delivery and the agreement between calculation and measurements improves with the advanced algorithms. [ABSTRACT FROM AUTHOR]