A generalized model for monitor units determination in ocular proton therapy using machine learning: A proof-of-concept study.

Objective.Determining and verifying the number of monitor units is crucial to achieving the desired dose distribution in radiotherapy and maintaining treatment efficacy. However, current commercial treatment planning system(s) dedicated to ocular passive eyelines in proton therapy do not provide the number of monitor units for patient-specific plan delivery. Performing specific pre-treatment field measurements, which is time and resource consuming, is usually gold-standard practice. This proof-of-concept study reports on the development of a multi-institutional-based generalized model for monitor units determination in proton therapy for eye melanoma treatments.Approach.To cope with the small number of patients being treated in proton centers, three European institutes participated in this study. Measurements data were collected to address output factor differences across the institutes, especially as function of field size, spread-out Bragg peak modulation width, residual range, and air gap. A generic model for monitor units prediction using a large number of 3748 patients and broad diversity in tumor patterns, was evaluated using six popular machine learning algorithms: (i) decision tree; (ii) random forest, (iii) extra trees, (iv) K-nearest neighbors, (v) gradient boosting, and (vi) the support vector regression. Features used as inputs into each machine learning pipeline were: Spread-out Bragg peak width, range, air gap, fraction and calibration doses. Performance measure was scored using the mean absolute error, which was the difference between predicted and real monitor units, as collected from institutional gold-standard methods.Main results.Predictions across algorithms were accurate within 3% uncertainty for up to 85.2% of the plans and within 10% uncertainty for up to 98.6% of the plans with the extra trees algorithm.Significance.A proof-of-concept of using machine learning-based generic monitor units determination in ocular proton therapy has been demonstrated. This could trigger the development of an independent monitor units calculation tool for clinical use.

Multi-institutional consensus on machine QA for isochronous cyclotron-based systems delivering ultra-high dose rate (FLASH) pencil beam scanning proton therapy in transmission mode.

BACKGROUND: The first clinical trials to assess the feasibility of FLASH radiotherapy in humans have started (FAST-01, FAST-02) and more trials are foreseen. To increase comparability between trials it is important to assure treatment quality and therefore establish a standard for machine quality assurance (QA). Currently, the AAPM TG-224 report is considered as the standard on machine QA for proton therapy, however, it was not intended to be used for ultra-high dose rate (UHDR) proton beams, which have gained interest due to the observation of the FLASH effect. PURPOSE: The aim of this study is to find consensus on practical guidelines on machine QA for UHDR proton beams in transmission mode in terms of which QA is required, how they should be done, which detectors are suitable for UHDR machine QA, and what tolerance limits should be applied. METHODS: A risk assessment to determine the gaps in the current standard for machine QA was performed by an international group of medical physicists. Based on that, practical guidelines on how to perform machine QA for UHDR proton beams were proposed. RESULTS: The risk assessment clearly identified the need for additional guidance on temporal dosimetry, addressing dose rate (constancy), dose spillage, and scanning speed. In addition, several minor changes from AAPM TG-224 were identified; define required dose rate levels, the use of clinically relevant dose levels, and the use of adapted beam settings to minimize activation of detector and phantom materials or to avoid saturation effects of specific detectors. The final report was created based on discussions and consensus. CONCLUSIONS: Consensus was reached on what QA is required for UHDR scanning proton beams in transmission mode for isochronous cyclotron-based systems and how they should be performed. However, the group discussions also showed that there is a lack of high temporal resolution detectors and sufficient QA data to set appropriate limits for some of the proposed QA procedures.

Geometrical accuracy of magnetic resonance imaging for ocular proton therapy planning.

Background & purpose: Magnetic resonance imaging (MRI) is increasingly used in treatment preparation of ocular proton therapy, but its spatial accuracy might be limited by geometric distortions due to susceptibility artefacts. A correct geometry of the MR images is paramount since it defines where the dose will be delivered. In this study, we assessed the geometrical accuracy of ocular MRI. Materials & methods: A dedicated ocular 3 T MRI protocol, with localized shimming and increased gradients, was compared to computed tomography (CT) and X-ray images in a phantom and in 15 uveal melanoma patients. The MRI protocol contained three-dimensional T2-weighted and T1-weighted sequences with an isotropic reconstruction resolution of 0.3-0.4 mm. Tantalum clips were identified by three observers and clip-clip distances were compared between T2-weighted and T1-weighted MRI, CT and X-ray images for the phantom and between MRI and X-ray images for the patients. Results: Interobserver variability was below 0.35 mm for the phantom and 0.30(T1)/0.61(T2) mm in patients. Mean absolute differences between MRI and reference were below 0.27 ± 0.16 mm and 0.32 ± 0.23 mm for the phantom and in patients, respectively. In patients, clip-clip distances were slightly larger on MRI than on X-ray images (mean difference T1: 0.11 ± 0.38 mm, T2: 0.10 ± 0.44 mm). Differences did not increase at larger distances and did not correlate to interobserver variability. Conclusions: A dedicated ocular MRI protocol can produce images of the eye with a geometrical accuracy below half the MRI acquisition voxel (<0.4 mm). Therefore, these images can be used for ocular proton therapy planning, both in the current model-based workflow and in proposed three-dimensional MR-based workflows.

Improving organs-at-risk sparing for choroidal melanoma patients: A CT-based two-beam strategy in ocular proton therapy with a dedicated eyeline.

PURPOSE: To investigate the potential clinical benefit of a two-beam arrangement technique using three-dimensional (3D) imaging of uveal melanoma (UM) patients treated with proton therapy and a dedicated eyeline. MATERIAL/METHODS: Retrospective CT-based treatment plans of 39 UM patients performed using a single beam (SB) were compared to plans with two beams (TB) optimized for better trade-offs in organs-at-risk sparing. The RBE-weighted prescribed dose was 60 Gy (DRBE, GTV = 60 Gy) in four fractions, assuming an RBE of 1.1. Dosimetric findings were analyzed for three patient groups based on tumor-optic nerve distance and UM staging (group GrA: ≤3 mm, T1 T2 UM; GrB: ≤3 mm, T3 UM; GrC: >3 mm, T1 T2 T3 UM). Finally, two schedules were compared on biologically effective dose (BED): both beams being delivered either the same day (TB), or on alternate days (TBalter). RESULTS: All strategies resulted in dosimetrically acceptable plans. A dose reduction to the anterior structures was achieved in 23/39 cases with the two-beam plans. D25% was significantly lowered compared to SB plans by 12.4 and 15.4 Gy RBE-weighted median dose in GrA and GrB, respectively. D2% was reduced by 18.6 and 6.0 Gy RBE-weighted median dose in GrA and GrB, respectively. A cost to the optic nerve was observed with a median difference up to 3.8 Gy RBE-weighted dose in GrB. BED differences were statistically significant for all considered parameters in favor of two beams delivered the same day. CONCLUSION: A two-beam strategy appears beneficial for posterior tumors abutting the optic nerve. This strategy might have a positive impact on the risk of ocular complications.

Characterization of the HollandPTC proton therapy beamline dedicated to uveal melanoma treatment and an interinstitutional comparison.

PURPOSE: Eye-dedicated proton therapy (PT) facilities are used to treat malignant intraocular lesions, especially uveal melanoma (UM). The first commercial ocular PT beamline from Varian was installed in the Netherlands. In this work, the conceptual design of the new eyeline is presented. In addition, a comprehensive comparison against five PT centers with dedicated ocular beamlines is performed, and the clinical impact of the identified differences is analyzed. MATERIAL/METHODS: The HollandPTC eyeline was characterized. Four centers in Europe and one in the United States joined the study. All centers use a cyclotron for proton beam generation and an eye-dedicated nozzle. Differences among the chosen ocular beamlines were in the design of the nozzle, nominal energy, and energy spectrum. The following parameters were collected for all centers: technical characteristics and a set of distal, proximal, and lateral region measurements. The measurements were performed with detectors available in-house at each institution. The institutions followed the International Atomic Energy Agency (IAEA) Technical Report Series (TRS)-398 Code of Practice for absolute dose measurement, and the IAEA TRS-398 Code of Practice, its modified version or International Commission on Radiation Units and Measurements Report No. 78 for spread-out Bragg peak normalization. Energy spreads of the pristine Bragg peaks were obtained with Monte Carlo simulations using Geant4. Seven tumor-specific case scenarios were simulated to evaluate the clinical impact among centers: small, medium, and large UM, located either anteriorly, at the equator, or posteriorly within the eye. Differences in the depth dose distributions were calculated. RESULTS: A pristine Bragg peak of HollandPTC eyeline corresponded to the constant energy of 75 MeV (maximal range 3.97 g/cm2 in water) with an energy spread of 1.10 MeV. The pristine Bragg peaks for the five participating centers varied from 62.50 to 104.50 MeV with an energy spread variation between 0.10 and 0.70 MeV. Differences in the average distal fall-offs and lateral penumbrae (LPs) (over the complete set of clinically available beam modulations) among all centers were up to 0.25 g/cm2 , and 0.80 mm, respectively. Average distal fall-offs of the HollandPTC eyeline were 0.20 g/cm2 , and LPs were between 1.50 and 2.15 mm from proximal to distal regions, respectively. Treatment time, around 60 s, was comparable among all centers. The virtual source-to-axis distance of 120 cm at HollandPTC was shorter than for the five participating centers (range: 165-350 cm). Simulated depth dose distributions demonstrated the impact of the different beamline characteristics among institutions. The largest difference was observed for a small UM located at the posterior pole, where a proximal dose between two extreme centers was up to 20%. CONCLUSIONS: HollandPTC eyeline specifications are in accordance with five other ocular PT beamlines. Similar clinical concepts can be applied to expect the same high local tumor control. Dosimetrical properties among the six institutions induce most likely differences in ocular radiation-related toxicities. This interinstitutional comparison could support further research on ocular post-PT complications. Finally, the findings reported in this study could be used to define dosimetrical guidelines for ocular PT to unify the concepts among institutions.

Flattening filter free vs flattened beams for breast irradiation.

PURPOSE: Flattening filter free (FFF) beams offer the potential for a higher dose rate, shorter treatment time, and lower peripheral dose. To investigate their role in large-field treatments, this study compared flattened and FFF beams for breast irradiation. METHODS AND MATERIALS: Ten left breast clinical plans comprising 2 tangential beams and a medially located 3-field simultaneous integrated boost (SIB) were replanned. Full intensity modulated radiotherapy (IMRT), hybrid IMRT, electronic tissue compensator (ETC), and multiple static field treatment plans were created for the elective breast volume using flattened and FFF beams, in combination with a 3-field IMRT SIB. Plan quality was assessed and delivery times were measured for all plans for 1 patient. Out-of-field doses were measured using an ionization chamber for an IMRT plan optimized on a corner of simple cubic phantom for both flattened and FFF beams. RESULTS: For each technique, mean target volume metrics (planning target volume coverage, homogeneity, conformity) were typically within 3% for flattened and FFF beams. Larger mean differences in boost conformity favoring flattened hybrid (7.2%) and full IMRT (5.5%) plans may have reflected limitations in plan normalization. Calculated heart and ipsilateral lung doses were comparable; however, both flattened and FFF low-dose phantom measurements were substantially higher than calculated values, rendering the comparison of low dose in the contralateral breast uncertain. Beam delivery times were on average 31% less for FFF. CONCLUSIONS: In general, target volume metrics for flattened and FFF plans were comparable. The planning system did not seem to allow for accurate peripheral dose evaluation. FFF was associated with a potentially shorter treatment time. All 4 IMRT techniques allowed FFF beams to generate acceptable plans for breast IMRT.