Two-dimensional time-resolved scintillating sheet monitoring of proton pencil beam scanning FLASH mouse irradiations.

BACKGROUND: Dosimetry in pre-clinical FLASH studies is essential for understanding the beam delivery conditions that trigger the FLASH effect. Resolving the spatial and temporal characteristics of proton pencil beam scanning (PBS) irradiations with ultra-high dose rates (UHDR) requires a detector with high spatial and temporal resolution. PURPOSE: To implement a novel camera-based system for time-resolved two-dimensional (2D) monitoring and apply it in vivo during pre-clinical proton PBS mouse irradiations. METHODS: Time-resolved 2D beam monitoring was performed with a scintillation imaging system consisting of a 1 mm thick transparent scintillating sheet, imaged by a CMOS camera. The sheet was placed in a water bath perpendicular to a horizontal PBS proton beam axis. The scintillation light was reflected through a system of mirrors and captured by the camera with 500 frames per second (fps) for UHDR and 4 fps for conventional dose rates. The raw images were background subtracted, geometrically transformed, flat field corrected, and spatially filtered. The system was used for 2D spot and field profile measurements and compared to radiochromic films. Furthermore, spot positions were measured for UHDR irradiations. The measured spot positions were compared to the planned positions and the relative instantaneous dose rate to equivalent fiber-coupled point scintillator measurements. For in vivo application, the scintillating sheet was placed 1 cm upstream the right hind leg of non-anaesthetized mice submerged in the water bath. The mouse leg and sheet were both placed in a 5 cm wide spread-out Bragg peak formed from the mono-energetic proton beam by a 2D range modulator. The mouse leg position within the field was identified for both conventional and FLASH irradiations. For the conventional irradiations, the mouse foot position was tracked throughout the beam delivery, which took place through repainting. For FLASH irradiations, the delivered spot positions and relative instantaneous dose rate were measured. RESULTS: The pixel size was 0.1 mm for all measurements. The spot and field profiles measured with the scintillating sheet agreed with radiochromic films within 0.4 mm. The standard deviation between measured and planned spot positions was 0.26 mm and 0.35 mm in the horizontal and vertical direction, respectively. The measured relative instantaneous dose rate showed a linear relation with the fiber-coupled scintillator measurements. For in vivo use, the leg position within the field varied between mice, and leg movement up to 3 mm was detected during the prolonged conventional irradiations. CONCLUSIONS: The scintillation imaging system allowed for monitoring of UHDR proton PBS delivery in vivo with 0.1 mm pixel size and 2 ms temporal resolution. The feasibility of instantaneous dose rate measurements was demonstrated, and the system was used for validation of the mouse leg position within the field.

Oxygen Enhancement Ratio-Weighted Dose Quantitatively Describes Acute Skin Toxicity Variations in Mice After Pencil Beam Scanning Proton FLASH Irradiation With Changing Doses and Time Structures.

PURPOSE: The aim of this work was to investigate the ability of a biological oxygen enhancement ratio-weighted dose, DOER, to describe acute skin toxicity variations observed in mice after proton pencil beam scanning irradiations with changing doses and beam time structures. METHODS AND MATERIALS: In five independent experiments, the right hind leg of a total of 621 CDF1 mice was irradiated previously in the entrance plateau of a pencil beam scanning proton beam. The incidence of acute skin toxicity (of level 1.5-2.0-2.5-3.0-3.5) was scored for 47 different mouse groups that mapped toxicity as function of dose for conventional and FLASH dose rate, toxicity as function of field dose rate with and without repainting, and toxicity when splitting the treatment into 1 to 6 identical deliveries separated by 2 minutes. DOER was calculated for all mouse groups using a simple oxygen kinetics model to describe oxygen depletion. The three independent model parameters (oxygen-depletion rate, oxygen-recovery rate, oxygen level without irradiation) were fitted to the experimental data. The ability of DOER to describe the toxicity variations across all experiments was investigated by comparing DOER-response curves across the five independent experiments. RESULTS: After conversion from the independent variable tested in each experiment to DOER, all five experiments had similar MDDOER50 (DOER giving 50% toxicity incidence) with standard deviations of 0.45 - 1.6 Gy for the five toxicity levels. DOER could thus describe the observed toxicity variations across all experiments. CONCLUSIONS: DOER described the varying FLASH-sparing effect observed for a wide range of conditions. Calculation of DOER for other irradiation conditions can quantitatively estimate the FLASH-sparing effect for arbitrary irradiations for the investigated murine model. With appropriate fitting parameters DOER also may be able to describe FLASH effect variations with dose and dose rate for other assays and endpoints.

Evaluation of in vitro irradiation setup: Designed for the horizontal beamline at the Danish Centre for Particle Therapy.

BACKGROUND: Radiobiological experimental setups are challenged by precise sample positioning along depth dose profile, scattering conditions, and practical difficulties that must be addressed in individual designs. The aim of this study was to produce cell survival curves with several irradiation modalities, by using a setup designed at the Danish Centre for Particle Therapy (DCPT) for in vitro proton irradiations using a horizontal beam line and thereby evaluating the setups use for in vitro irradiations experiments. MATERIALS AND METHODS: The setup is a water phantom suitable for in vitro research with multiple irradiation modalities, in particular the pencil scanning proton beam available from a horizontal experimental beamline. The phantom included a water tank of 39.0 × 17.0 × 20.5 cm. Cell survival-curves were produced using the cell line V79 Chinese hamster lung fibroblast cells (V79s) in biological triplicates of clonogenic assays. Cell survival curves were produced with both a 18 MeV electron beam, 6 MV photon beam, and a Spread-Out Bragg Peak (SOBP) proton beam formed by pristine energies of 85-111 MeV where three positions were examined. RESULTS: Survival curves with uncertainty areas were made for all modalities. Dosimetric uncertainty amounted to, respectively, 4%, 3% and 3% for proton, electron, and high energy photon irradiations. Cell survival fraction uncertainty was depicted as the standard deviation between replications of the experiment. CONCLUSION: Cell survival curves could be produced with acceptable uncertainties using this novel water phantom and cellular laboratory workflow. The setup is useful for future in vitro irradiation experiments.

A simple method to measure the gating latencies in photon and proton based radiotherapy using a scintillating crystal.

BACKGROUND: In respiratory gated radiotherapy, low latency between target motion into and out of the gating window and actual beam-on and beam-off is crucial for the treatment accuracy. However, there is presently a lack of guidelines and accurate methods for gating latency measurements. PURPOSE: To develop a simple and reliable method for gating latency measurements that work across different radiotherapy platforms. METHODS: Gating latencies were measured at a Varian ProBeam (protons, RPM gating system) and TrueBeam (photons, TrueBeam gating system) accelerator. A motion-stage performed 1 cm vertical sinusoidal motion of a marker block that was optically tracked by the gating system. An amplitude gating window was set to cover the posterior half of the motion (0-0.5 cm). Gated beams were delivered to a 5 mm cubic scintillating ZnSe:O crystal that emitted visible light when irradiated, thereby directly showing when the beam was on. During gated beam delivery, a video camera acquired images at 120 Hz of the moving marker block and light-emitting crystal. After treatment, the block position and crystal light intensity were determined in all video frames. Two methods were used to determine the gate-on (τon ) and gate-off (τoff ) latencies. By method 1, the video was synchronized with gating log files by temporal alignment of the same block motion recorded in both the video and the log files. τon was defined as the time from the block entered the gating window (from gating log files) to the actual beam-on as detected by the crystal light. Similarly, τoff was the time from the block exited the gating window to beam-off. By method 2, τon and τoff were found from the videos alone using motion of different sine periods (1-10 s). In each video, a sinusoidal fit of the block motion provided the times Tmin of the lowest block position. The mid-time, Tmid-light , of each beam-on period was determined as the time halfway between crystal light signal start and end. It can be shown that the directly measurable quantity Tmid-light - Tmin  = (τoff +τon )/2, which provided the sum (τoff +τon ) of the two latencies. It can also be shown that the beam-on (i.e., crystal light) duration ΔTlight increases linearly with the sine period and depends on τoff - τon : ΔTlight  = constant•period+(τoff - τon ). Hence, a linear fit of ΔTlight as a function of the period provided the difference of the two latencies. From the sum (τoff +τon ) and difference (τoff - τon ), the individual latencies were determined. RESULTS: Method 1 resulted in mean (±SD) latencies of τon  = 255 ± 33 ms, τoff  = 82 ± 15 ms for the ProBeam and τon  = 84 ± 13 ms, τoff  = 44 ± 11 ms for the TrueBeam. Method 2 resulted in latencies of τon  = 255 ± 23 ms, τoff  = 95 ± 23 ms for the ProBeam and τon  = 83 ± 8 ms, τoff  = 46 ± 8 ms for the TrueBeam. Hence, the mean latencies determined by the two methods agreed within 13 ms for the ProBeam and within 2 ms for the TrueBeam. CONCLUSIONS: A novel, simple and low-cost method for gating latency measurements that work across different radiotherapy platforms was demonstrated. Only the TrueBeam fully fulfilled the AAPM TG-142 recommendation of maximum 100 ms latencies.

Intrafraction tumor motion monitoring and dose reconstruction for liver pencil beam scanning proton therapy.

Background: Pencil beam scanning (PBS) proton therapy can provide highly conformal target dose distributions and healthy tissue sparing. However, proton therapy of hepatocellular carcinoma (HCC) is prone to dosimetrical uncertainties induced by respiratory motion. This study aims to develop intra-treatment tumor motion monitoring during respiratory gated proton therapy and combine it with motion-including dose reconstruction to estimate the delivered tumor doses for individual HCC treatment fractions. Methods: Three HCC-patients were planned to receive 58 GyRBE (n=2) or 67.5 GyRBE (n=1) of exhale respiratory gated PBS proton therapy in 15 fractions. The treatment planning was based on the exhale phase of a 4-dimensional CT scan. Daily setup was based on cone-beam CT (CBCT) imaging of three implanted fiducial markers. An external marker block (RPM) on the patient's abdomen was used for exhale gating in free breathing. This study was based on 5 fractions (patient 1), 1 fraction (patient 2) and 6 fractions (patient 3) where a post-treatment control CBCT was available. After treatment, segmented 2D marker positions in the post-treatment CBCT projections provided the estimated 3D motion trajectory during the CBCT by a probability-based method. An external-internal correlation model (ECM) that estimated the tumor motion from the RPM motion was built from the synchronized RPM signal and marker motion in the CBCT. The ECM was then used to estimate intra-treatment tumor motion. Finally, the motion-including CTV dose was estimated using a dose reconstruction method that emulates tumor motion in beam's eye view as lateral spot shifts and in-depth motion as changes in the proton beam energy. The CTV homogeneity index (HI) The CTV homogeneity index (HI) was calculated as D 2 %  -  D 98 % D 50 %   × 100 % . Results: The tumor position during spot delivery had a root-mean-square error of 1.3 mm in left-right, 2.8 mm in cranio-caudal and 1.7 mm in anterior-posterior directions compared to the planned position. On average, the CTV HI was larger than planned by 3.7%-points (range: 1.0-6.6%-points) for individual fractions and by 0.7%-points (range: 0.3-1.1%-points) for the average dose of 5 or 6 fractions. Conclusions: A method to estimate internal tumor motion and reconstruct the motion-including fraction dose for PBS proton therapy of HCC was developed and demonstrated successfully clinically.

Time-resolved dose rate measurements in pencil beam scanning proton FLASH therapy with a fiber-coupled scintillator detector system.

BACKGROUND: The spatial and temporal dose rate distribution of pencil beam scanning (PBS) proton therapy is important in ultra-high dose rate (UHDR) or FLASH irradiations. Validation of the temporal structure of the dose rate is crucial for quality assurance and may be performed using detectors with high temporal resolution and large dynamic range. PURPOSE: To provide time-resolved in vivo dose rate measurements using a scintillator-based detector during proton PBS pre-clinical mouse experiments with dose rates ranging from conventional to UHDR. METHODS: All irradiations were performed at the entrance plateau of a 250 MeV PBS proton beam. A detector system with four fiber-coupled ZnSe:O inorganic scintillators and 20 µs temporal resolution was used for dose rate measurements. The system was first characterized in terms of precision and stem signal. The detector precision was determined through repeated irradiations with the same field. The stem signal contribution was quantified by irradiating two of the detector probes alongside a bare fiber (fiber without a coupled scintillator). Next, the detector system was calibrated against an ionization chamber (IC) with all four detector probes and the IC placed in a water bath at 2 cm depth. A scan pattern covering 9.6 × 9.6 cm was used. Multiple irradiations with different requested nozzle currents provided instantaneous dose rates at the detector positions in the range of 7-1270 Gy/s. The correspondence of the detector signal (in Volts) to the instantaneous dose rate (in Gy/s) was found. The instantaneous dose rate was calculated from the beam current and the spot-to-detector distance assuming a Gaussian beam profile at distances up to 8 mm from the spot. Afterwards, the calibrated system was used in vivo, in mouse experiments, where mouse legs were irradiated with a constant dose and varying field dose rates of 0.7-87.5 Gy/s. The instantaneous dose rate was measured for each delivered spot and the delivered dose was determined as the integrated instantaneous dose rate. The spot dose profile and PBS dose rate map were calculated. The dose contamination to neighbouring mice were measured together with the upper limit of the dose to the mouse body. RESULTS: The detectors showed high precision with ≤0.4% fluctuations in the measured dose. The stem signal exceeded 10% for spots <5 mm from the optical fiber and >18 mm from the scintillator. It contributed up to 0.2% to the total dose, which was considered negligible. All four detectors showed a non-linear relation between signal and instantaneous dose rate, which was modelled with a polynomial response function. In the mouse experiments, the measured scintillator dose showed 1.8% fluctuations, independent of the field dose rate. The in vivo measured spot dose profile had tails that deviated from a Gaussian profile with measurable dose contributions from spots up to 85 mm from the detector. Neighbour mouse irradiation contributed ∼1% of the total mouse dose. The upper limit of the mouse body dose was 6% of the mouse leg dose. CONCLUSIONS: A fiber-coupled inorganic scintillator-based detector system can provide high precision in vivo measurements of the instantaneous dose rate if correction for the non-linear dose rate dependency is applied.

The dosimetric error due to uncorrected tumor rotation during real-time adaptive prostate stereotactic body radiation therapy.

BACKGROUND: During prostate stereotactic body radiation therapy (SBRT), prostate tumor translational motion may deteriorate the planned dose distribution. Most of the major advances in motion management to date have focused on correcting this one aspect of the tumor motion, translation. However, large prostate rotation up to 30° has been measured. As the technological innovation evolves toward delivering increasingly precise radiotherapy, it is important to quantify the clinical benefit of translational and rotational motion correction over translational motion correction alone. PURPOSE: The purpose of this work was to quantify the dosimetric impact of intrafractional dynamic rotation of the prostate measured with a six degrees-of-freedom tumor motion monitoring technology. METHODS: The delivered dose was reconstructed including (a) translational and rotational motion and (b) only translational motion of the tumor for 32 prostate cancer patients recruited on a 5-fraction prostate SBRT clinical trial. Patients on the trial received 7.25 Gy in a treatment fraction. A 5 mm clinical target volume (CTV) to planning target volume (PTV) margin was applied in all directions except the posterior direction where a 3 mm expansion was used. Prostate intrafractional translational motion was managed using a gating strategy, and any translation above the gating threshold was corrected by applying an equivalent couch shift. The residual translational motion is denoted as T r e s $T_{res}$ . Prostate intrafractional rotational motion R u n c o r r $R_{uncorr}$ was recorded but not corrected. The dose differences from the planned dose due to T r e s $T_{res}$ + R u n c o r r $R_{uncorr}$ , ΔD( T r e s $T_{res}$ + R u n c o r r $R_{uncorr}$ ) and due to T r e s $T_{res}$ alone, ΔD( T r e s $T_{res}$ ), were then determined for CTV D98, PTV D95, bladder V6Gy, and rectum V6Gy. The residual dose error due to uncorrected rotation, R u n c o r r $R_{uncorr}$ was then quantified: Δ D R e s i d u a l $\Delta D_{Residual}$ = ΔD( T r e s $T_{res}$ + R u n c o r r $R_{uncorr}$ ) - ΔD( T res ${T}_{\textit{res}}$ ). RESULTS: Fractional data analysis shows that the dose differences from the plan (both ΔD( T r e s $T_{res}$ + R u n c o r r $R_{uncorr}$ ) and ΔD( T r e s $T_{res}$ )) for CTV D98 was less than 5% in all treatment fractions. ΔD( T r e s $T_{res}$ + R u n c o r r $R_{uncorr}$ ) was larger than 5% in one fraction for PTV D95, in one fraction for bladder V6Gy, and in five fractions for rectum V6Gy. Uncorrected rotation, R u n c o r r $R_{uncorr}$ induced residual dose error, Δ D R e s i d u a l $\Delta D_{Residual}$ , resulted in less dose to CTV and PTV in 43% and 59% treatment fractions, respectively, and more dose to bladder and rectum in 51% and 53% treatment fractions, respectively. The cumulative dose over five fractions, ∑D( T r e s $T_{res}$ + R u n c o r r $R_{uncorr}$ ) and ∑D( T r e s $T_{res}$ ), was always within 5% of the planned dose for all four structures for every patient. CONCLUSIONS: The dosimetric impact of tumor rotation on a large prostate cancer patient cohort was quantified in this study. These results suggest that the standard 3-5 mm CTV-PTV margin was sufficient to account for the intrafraction prostate rotation observed for this cohort of patients, provided an appropriate gating threshold was applied to correct for translational motion. Residual dose errors due to uncorrected prostate rotation were small in magnitude, which may be corrected using different treatment adaptation strategies to further improve the dosimetric accuracy.

Accuracy and potential improvements of surface-guided breast cancer radiotherapy in deep inspiration breath-hold with daily image-guidance.

Objective.Radiotherapy of left-sided breast cancer in deep inspiration breath-hold (DIBH) reduces the heart dose. Surface guided radiotherapy (SGRT) can guide the DIBH, but the accuracy is subject to variations in the chest wall position relative to the patient surface.Approach.In this study, ten left-sided breast cancer patients received DIBH radiotherapy with tangential fields in 15-18 fractions. After initial SGRT setup in free breathing an orthogonal MV/kV image pair was acquired during SGRT-guided breath-hold. The couch was corrected to align the chest wall during another breath-hold, and a new SGRT reference surface was acquired for the gating. The chest wall position error during treatment was determined from continuous cine MV images in the imager direction perpendicular to the cranio-caudal direction. A treatment error budget was made with individual contributions from the online registration of the setup MV image, the difference in breath-hold level between setup imaging and SGRT reference surface acquisition, the SGRT level during treatment, and intra-fraction shifts of the chest wall relative to the SGRT reference surface. In addition to the original setup protocol (Scenario A), SGRT was also simulated with better integration of image-guidance by capturing either the new reference surface (Scenario B) or the SGRT positional signal (Scenario C) simultaneously with the setup MV image, and accounting for the image-guided couch correction by shifting the SGRT reference surface digitally.Main results.In general, the external SGRT signal correlated well with the internal chest wall position error (correlation coefficient >0.7 for 75% of field deliveries), but external-to-internal target position offsets above 2 mm occasionally occurred (13% of fractions). The PTV margin required to account for the treatment error was 3.5 mm (Scenario A), 3.4 mm (B), and 3.1 mm (C).Significance. Further integration of SGRT with image-guidance may improve treatment accuracy and workflow although the current study did not show large accuracy improvements of scenario B and C compared to scenario A.

Time structure of pencil beam scanning proton FLASH beams measured with scintillator detectors and compared with log files.

PURPOSE: Key factors in FLASH treatments are the ultra-high dose rate (UHDR) and the time structure of the beam delivery. Measurement of the time structure in pencil beam scanning (PBS) proton FLASH treatments is challenging for many types of detectors since high temporal resolution is needed. In this study, a fast scintillator detector system was developed and used to measure the individual spot durations as well as the time when the beam moves between two positions (transition duration) during PBS proton FLASH and UHDR treatments. The spot durations were compared with machine log-file recordings. METHODS: A detector system based on inorganic scintillating crystals was developed. The system consisted of four detector probes made of a sub-millimeter ZnSe:O crystal that was coupled via an optical fiber to an optical reader with 50 kHz sampling rate. The detector system was used in two experiments, both performed with a PBS proton beam with 250 MeV beam energy and 215 nA requested nozzle beam current. The sampling rate enabled multiple measurements during each spot delivery and during the beam transition between spots. First, the detector was tested in a phantom experiment, where a total of 305 scan sequences were delivered to the four detectors. The number of spots delivered without beam interruption in a single scan sequence ranged from one to 35. The spot duration and transition duration were measured for each individual spot. Secondly, the detector system was used in vivo in preclinical experiments with FLASH irradiation of mouse legs placed in the entrance plateau of the beam. A single detector was placed 1 cm downstream of the irradiated mouse leg. The mouse dose ranged from 30.5 to 44.2 Gy and the field consisted of 35 spots. The spot durations as well as the mean dose rate (field dose divided by the measured field duration) for each mouse were determined using the detector and then compared with the corresponding log files. RESULTS: The phantom experiment showed that the logged total duration of an uninterrupted spot sequence was consistently shorter than the measured duration with a difference of -0.252 ms (95% confidence interval: [-0.255, -0.249 ms]). This corresponded to 0.05%-0.07% of the spot sequence duration in the mice experiments. For individual spots, the mean ± 1SD difference between logged and measured spot duration was -0.39 ± 0.05 ms for the first spot in a sequence, 0.13 ± 0.04 ms for the last spot in a sequence, and -0.0017 ± 0.09 ms for the intermediate spots in a sequence. The measured spot transition durations were 0.20 ± 0.04 ms (5.1 mm horizontal steps) and 0.50 ± 0.04 ms (5.0 mm vertical steps). For the mouse experiments, the mean dose rate calculated from the measured field duration was 84.1-92.5 Gy/s. It agreed with log files with a root mean square difference of 0.02 Gy/s. CONCLUSIONS: Fiber-coupled scintillator detectors were designed with sufficient temporal resolution to measure the spot and transition duration during PBS proton UHDR deliveries. Their small volume makes them feasible for in vivo use in preclinical FLASH studies. The logged spot durations were in excellent agreement with measurements but showed small systematic errors in the logged duration for the first and last spot in a sequence.

Uniform versus non-uniform dose prescription for proton stereotactic body radiotherapy of liver tumors investigated by extensive motion-including treatment simulations.

Compared to x-ray-based stereotactic body radiotherapy (SBRT) of liver cancer, proton SBRT may reduce the normal liver tissue dose. For an optimal trade-off between target and liver dose, a non-uniform dose prescription is often applied in x-ray SBRT, but lacks investigation for proton SBRT. Also, proton SBRT is prone to breathing-induced motion-uncertainties causing target mishit or dose alterations by interplay with the proton delivery. This study investigated non-uniform and uniform dose prescription in proton-based liver SBRT, including effects of rigid target motion observed during planning-4DCT and treatment. The study was based on 42 x-ray SBRT fractions delivered to 14 patients under electromagnetic motion-monitoring. For each patient, a non-uniform and uniform proton plan were made. The uniform plan was renormalized to be iso-toxic with the non-uniform plan using a NTCP model for radiation-induced liver disease. The motion data were used in treatment simulations to estimate the delivered target dose with rigid motion. Treatment simulations were performed with and without a repainting scheme designed to mitigate interplay effects. Including rigid motion, the achieved CTV mean dose after three fractions delivered without repainting was on average (±SD) 24.8 ± 8.4% higher and the D98%was 16.2 ± 11.3% higher for non-uniform plans than for uniform plans. The interplay-induced increase in D2%relative to the static plans was reduced from 3.2 ± 4.1% without repainting to -0.5 ± 1.7% with repainting for non-uniform plans and from 1.5 ± 2.0% to 0.1 ± 1.3% for uniform plans. Considerable differences were observed between estimated CTV doses based on 4DCT motion and intra-treatment motion. In conclusion, non-uniform dose prescription in proton SBRT may provide considerably higher tumor doses than uniform prescription for the same complication risk. Due to motion variability, target doses estimated from 4DCT motion may not accurately reflect the delivered dose. Future studies including modelling of deformations and associated range uncertainties are warranted to confirm the findings.

Strategies for Motion Robust Proton Therapy With Pencil Beam Scanning for Esophageal Cancer.

PURPOSE: Proton therapy of esophageal cancer is superior to photon radiation therapy in terms of normal tissue sparing. However, respiratory motion and anatomical changes may compromise target dose coverage owing to density changes, geometric misses, and interplay effects. Here we investigate the combined effect on clinical target volume (CTV) coverage and compare proton therapy with intensity modulated radiation therapy (IMRT). METHODS AND MATERIALS: This study includes 26 patients with esophageal cancer previously treated with IMRT planned on 4-dimensional computed tomography (4D-CT). For each patient, 7 proton pencil beam scanning (PBS) plans were created with different field configurations and optimization strategies. The effect of respiration was investigated by calculating the phase doses, 4D dose, and 4D dynamic dose (including interplay effects). The effect of anatomical changes was investigated by recalculating all plans on all phases of a 4D-CT surveillance scan. RESULTS: The most robust PBS plans were achieved using 2 posterior beams requiring coverage of planning target volume (PTV) and simultaneously using robust optimization (RO) of CTV (2PAPTVRO), resulting in only 1 patient showing V95%CTV <97% in 1 or more phases of the planning CT. For the least robust PBS plans obtained using lateral + posterior beams and CTV-RO, but not requiring PTV coverage (2LPRO), 10 patients showed underdosage. For IMRT, 2 patients showed underdosage. Interplay effects reduced V95%CTV significantly when delivering only 1 fraction, but the effects generally averaged out after 10 fractions. The effect of interplay was significantly larger for RO-only plans compared with plans optimized with RO combined with PTV coverage. Combining the effect of anatomical changes and respiration on the 4D-CT surveillance scan resulted in V95%CTV <97% for 3 2PAPTVRO, 16 2LPRO, and 8 IMRT patients. CONCLUSIONS: PBS using posterior beam angles was more robust to anatomical changes and respiration than IMRT. The effect of respiration was enhanced when anatomical changes were present. Single fraction interplay effects deteriorated the dose distribution but were averaged out after 10 fractions.

Six degrees of freedom dynamic motion-including dose reconstruction in a commercial treatment planning system.

PURPOSE: Intrafractional motion during radiotherapy delivery can deteriorate the delivered dose. Dynamic rotational motion of up to 38 degrees has been reported during prostate cancer radiotherapy, but methods to determine the dosimetric consequences of such rotations are lacking. Here, we create and experimentally validate a dose reconstruction method that accounts for dynamic rotations and translations in a commercial treatment planning system (TPS). Interplay effects are quantified by comparing dose reconstructions with dynamic and constant rotations. METHODS: The dose reconstruction accumulates the dose in points of interest while the points are moved in six degrees of freedom (6DoF) in a precalculated time-resolved four-dimensional (4D) dose matrix to emulate dynamic motion in a patient. The required 4D dose matrix was generated by splitting the original treatment plan into multiple sub-beams, each representing 0.4 s dose delivery, and recalculating the dose of the split plan in the TPS (Eclipse). The dose accumulation was performed via TPS scripting by querying the dose of each sub-beam in dynamically moving points, allowing dose reconstruction with any dynamic motion. The dose reconstruction was validated with film dosimetry for two prostate dual arc VMAT plans with intra-prostatic lesion boosts. The plans were delivered to a pelvis phantom with internal dynamic rotational motion of a film stack (21 films with 2.5 mm separation). Each plan was delivered without motion and with three prostate motion traces. Motion-including dose reconstruction was performed for each motion experiment using the actual dynamic rotation as well as a constant rotation equal to the mean rotation during the experiment. For each experiment, the 3%/2 mm γ failure rate of the TPS dose reconstruction was calculated with the film measurement being the reference. For each motion experiment, the motion-induced 3%/2 mm γ failure rate was calculated using the static delivery as the reference and compared between film measurements and TPS dose reconstruction. DVH metrics for RT structures fully contained in the film volume were also compared between film and TPS. RESULTS: The mean γ failure rate of the TPS dose reconstructions when compared to film doses was 0.8% (two static experiments) and 1.7% (six dynamic experiments). The mean (range) of the motion-induced γ failure rate in film measurements was 35.4% (21.3-59.2%). The TPS dose reconstruction agreed with these experimental γ failure rates with root-mean-square errors of 2.1% (dynamic rotation dose reconstruction) and 17.1% (dose reconstruction assuming constant rotation). By DVH metrics, the mean (range) difference between dose reconstructions with dynamic and constant rotation was 4.3% (-0.3-10.6%) (urethra D 2 % ), -0.6% (-5.6%-2.5%) (urethra D 99 % ), 1.1% (-7.1-7.7%) (GTV D 2 % ), -1.4% (-17.4-7.1%) (GTV D 95 % ), -1.2% (-17.1-5.7%) (GTV D 99 % ), and -0.1% (-3.2-7.6%) (GTV mean dose). Dose reconstructions with dynamic motion revealed large interplay effects (cold and hot spots). CONCLUSIONS: A method to perform dose reconstructions for dynamic 6DoF motion in a TPS was developed and experimentally validated. It revealed large differences in dose distribution between dynamic and constant rotations not identifiable through dose reconstructions with constant rotation.

AAPM Task Group 264: The safe clinical implementation of MLC tracking in radiotherapy.

The era of real-time radiotherapy is upon us. Robotic and gimbaled linac tracking are clinically established technologies with the clinical realization of couch tracking in development. Multileaf collimators (MLCs) are a standard equipment for most cancer radiotherapy systems, and therefore MLC tracking is a potentially widely available technology. MLC tracking has been the subject of theoretical and experimental research for decades and was first implemented for patient treatments in 2013. The AAPM Task Group 264 Safe Clinical Implementation of MLC Tracking in Radiotherapy Report was charged to proactively provide the broader radiation oncology community with (a) clinical implementation guidelines including hardware, software, and clinical indications for use, (b) commissioning and quality assurance recommendations based on early user experience, as well as guidelines on Failure Mode and Effects Analysis, and (c) a discussion of potential future developments. The deliverables from this report include: an explanation of MLC tracking and its historical development; terms and definitions relevant to MLC tracking; the clinical benefit of, clinical experience with and clinical implementation guidelines for MLC tracking; quality assurance guidelines, including example quality assurance worksheets; a clinical decision pathway, future outlook and overall recommendations.

Dosimetric impact of intrafraction prostate rotation and accuracy of gating, multi-leaf collimator tracking and couch tracking to manage rotation: An end-to-end validation using volumetric film measurements.

BACKGROUND AND PURPOSE: Both gating and tracking can mitigate the deteriorating dosimetric impact of intrafraction translation during prostate stereotactic body radiotherapy (SBRT). However, their ability to manage intrafraction rotation has not yet been thoroughly investigated. The dosimetric accuracy of gating, MLC tracking and couch tracking to manage intrafraction prostate rotation was investigated. MATERIALS AND METHODS: Treatment plans for end-to-end tests of prostate SBRT with focal boosting were generated for a dynamic anthropomorphic pelvis phantom. The phantom applied internal lateral rotation (up to 25°) and coupled vertical and longitudinal translation of a radiochromic film stack that was used for dose measurements. Dose was delivered for each plan while the phantom applied motion according to three typical prostate motion traces without compensation (i), with gating (ii), with MLC tracking (iii) or with couch tracking (iv). Measured doses for the four motion compensation strategies were compared with the planned dose in terms of γ-index analysis, target coverage and organs at risk (OAR) sparing. RESULTS: Intrafraction rotation reduced the 3%(global)/2mm γ-index passing rate (γPR) for the prostate target volume by median (range) -33.2% (-68.6%, -4.1%) when no motion compensation was applied. The use of motion compensation improved the γPR by 13.2% (-0.4%, 32.9%) for gating, by 6.0% (-0.8%, 27.7%) for MLC tracking and by 11.1% (1.2%, 22.9%) for couch tracking. The three compensation techniques improved the target coverage in most cases. Gating showed better OAR sparing than MLC tracking or couch tracking. CONCLUSIONS: Compensation of intrafraction prostate rotation with gating, MLC tracking and couch tracking was investigated experimentally for the first time. All three techniques improved the dosimetric accuracy, but residual motion-related dose errors remained due to the lack of rotation correction.

Fully automated detection of heart irradiation in cine MV images acquired during breast cancer radiotherapy.

PURPOSE: To develop robust automated detection of heart irradiation in continuous portal images (cine MV images) of tangential breast cancer treatments. METHODS: Cine MV images of 302 tangential field deliveries were recorded for ten left-sided breast cancer patients receiving deep-inspiration breath-hold radiotherapy. An algorithm for fully automated heart edge detection in cine MV images was developed and tested for all images. The algorithm first enhances the heart edge contrast greatly by exploiting that pixels on the heart edge change their intensity cyclically, and highly correlated, at 1-3 Hz due to heartbeat. The algorithm then detects the heart edge in the enhanced image and calculates the exposed heart area within the field aperture. RESULTS: The algorithm correctly identified the heart edge in all cine MV series with heart exposure (169 of 302 field deliveries). With conservative selection criteria the algorithm on average identified 70 heart edge pixels in the heart-including field deliveries (range: 10-230) without false positives. With less strict criteria 106 heart edge pixels were identified on average (range: 13-262) with 0.6% being false positives. The heart edge bordering the lung was segmented highly reliably even a few millimeters outside the field edge. For six patients with frequent heart irradiation, the exposed heart area showed large interfraction variations and smaller intrafraction variations. CONCLUSIONS: Automated heart edge detection in cine MV images was proposed, developed and shown to be highly efficient for heart exposure detection in tangential breast fields. It may allow unsupervised surveillance of heart exposure at all tangential breast cancer treatments in a clinic.

Simulated real-time dose reconstruction for moving tumors in stereotactic liver radiotherapy.

PURPOSE: In radiotherapy, tumor motion may deteriorate the planned dose distribution. However, the dosimetric consequences of the motion are normally unknown for individual treatments. We here present a method for real-time motion-including tumor dose reconstruction and demonstrate its use for simulated stereotactic body radiotherapy (SBRT) of patients with liver cancer previously treated with Calypso-guided gating. METHODS: Real-time motion-including dose reconstruction was performed using in-house developed software, DoseTracker, on offline replays of previous clinical treatments. The patient cohort consisted of fifteen patients previously treated in our clinic with three-fraction SBRT to the liver using conformal or IMRT plans. The tumor motion at treatment was monitored with implanted electromagnetic transponders. The dose reconstruction was performed for both the actual gated treatments and simulated nongated treatments using a 21 Hz data stream containing accelerator parameters and the recorded motion. The dose was reconstructed in the same calculation points within the planning target volume (PTV) as used by the treatment planning system (TPS). The reconstructed doses were compared with calculations performed in the TPS, in which the motion was modeled as a series of isocenter shifts. The comparison included point doses as a function of treatment time and the dose volume histogram (DVH) for the clinical target volume (CTV). The motion-induced reduction in the dose to 95% of the CTV, Δ D 95 % , and in the mean CTV dose, ΔDMean , was compared between DoseTracker and the TPS for each simulated fraction. DoseTracker currently assumes water density within the patient contour, so for comparison, the TPS calculations were performed with both CT density and water density. The calculation times were additionally analyzed. RESULTS: Dose reconstruction was carried out for ninety SBRT sessions with calculation volumes ranging from 9.9 to 366.4 cm3 and median calculation times of 55-155 ms (equivalent to 18.2-6.5 Hz). Time-resolved trends of doses to a single calculation point in the patient were well replicated and dose differences between actual and planned calculations matched well. ΔDMean had a range of -0.1%-30.7%-points and was estimated by DoseTracker with a root-mean-square deviation (RMSD) to the TPS calculations of 0.43%-points (water density) and 0.79%-points (CT density). Similarly, Δ D 95 % had a range of 0.0%-35.2%-points and was estimated by DoseTracker with an RMSD of 0.80%-points (water density) and 1.33%-points (CT density). DoseTracker predicted losses in tumor dose coverage above 5%-points with high sensitivity (91.7%) and specificity (97.6%). CONCLUSIONS: Real-time dose reconstruction to moving tumors was demonstrated on offline replays of previous clinical treatments. DVHs of actually delivered dose are made available immediately after the end of treatment fractions. It shows promising results for liver SBRT with accurate estimation of CTV dose deteriorations caused by motion during treatment.

First clinical real-time motion-including tumor dose reconstruction during radiotherapy delivery.

PURPOSE: To clinically implement and characterize real-time motion-including tumor dose reconstruction during radiotherapy delivery. METHODS: Seven patients with 2-3 fiducial markers implanted near liver tumors received stereotactic body radiotherapy on a conventional linear accelerator. The 3D marker motion during a setup CBCT scan was determined online from the CBCT projections and used to generate a correlation model between tumor and external marker block motion. During treatment, the correlation model was updated by kV imaging every three seconds and used for real-time tumor localization. Using streamed accelerator parameters and tumor positions, in-house developed software, DoseTracker, calculated the dose to the moving tumor in real-time assuming water density in the patient. Post-treatment, the real-time tumor localization was validated by comparison with independent marker segmentations and 3D motion estimations. Dose reconstruction was validated by comparison with treatment planning system (TPS) calculations that modeled motion as isocenter shifts and used both actual CT densities and water densities. RESULTS: The real-time estimated tumor position had a mean 3D root-mean-square error of 1.7 mm (range: 0.9-2.6 mm). The motion-induced reduction in the minimum dose to 95% of the clinical target volume (CTV D95) per fraction was up to 12.3%-points. It was estimated in real-time by DoseTracker during patient treatment with a root-mean-square difference relative to the TPS of 1.3%-points (TPS CT) and 1.0%-points (TPS water). CONCLUSIONS: The world's first clinical real-time motion-including tumor dose reconstruction during radiotherapy was demonstrated. This marks an important milestone for real-time in-treatment quality assurance and paves the way for real-time dose-guided treatment adaptation.

The accuracy and precision of the KIM motion monitoring system used in the multi-institutional TROG 15.01 Stereotactic Prostate Ablative Radiotherapy with KIM (SPARK) trial.

PURPOSE: Kilovoltage intrafraction monitoring (KIM) allows for real-time image guidance for tracking tumor motion in six-degrees-of-freedom (6DoF) on a standard linear accelerator. This study assessed the geometric accuracy and precision of KIM used to guide patient treatments in the TROG 15.01 multi-institutional Stereotactic Prostate Ablative Radiotherapy with KIM trial and investigated factors affecting accuracy and precision. METHODS: Fractions from 44 patients with prostate cancer treated using KIM-guided SBRT were analyzed across four institutions, on two different linear accelerator models and two different beam models (6 MV and 10 MV FFF). The geometric accuracy and precision of KIM was assessed from over 33 000 images (translation) and over 9000 images (rotation) by comparing the real-time measured motion to retrospective kV/MV triangulation. Factors potentially affecting accuracy, including contrast-to-noise ratio (CNR) of kV images and incorrect marker segmentation, were also investigated. RESULTS: The geometric accuracy and precision did not depend on treatment institution, beam model or motion magnitude, but was correlated with gantry angle. The centroid geometric accuracy and precision of the KIM system for SABR prostate treatments was 0.0 ± 0.5, 0.0 ± 0.4 and 0.1 ± 0.3 mm for translation, and -0.1 ± 0.6°, -0.1 ± 1.4° and -0.1 ± 1.0° for rotation in the AP, LR and SI directions respectively. Centroid geometric error exceeded 2 mm for 0.05% of this dataset. No significant relationship was found between large geometric error and CNR or marker segmentation correlation. CONCLUSIONS: This study demonstrated the ability of KIM to locate the prostate with accuracy below other uncertainties in radiotherapy treatments, and the feasibility for KIM to be implemented across multiple institutions.

A deep learning framework for automatic detection of arbitrarily shaped fiducial markers in intrafraction fluoroscopic images.

PURPOSE: Real-time image-guided adaptive radiation therapy (IGART) requires accurate marker segmentation to resolve three-dimensional (3D) motion based on two-dimensional (2D) fluoroscopic images. Most common marker segmentation methods require prior knowledge of marker properties to construct a template. If marker properties are not known, an additional learning period is required to build the template which exposes the patient to an additional imaging dose. This work investigates a deep learning-based fiducial marker classifier for use in real-time IGART that requires no prior patient-specific data or additional learning periods. The proposed tracking system uses convolutional neural network (CNN) models to segment cylindrical and arbitrarily shaped fiducial markers. METHODS: The tracking system uses a tracking window approach to perform sliding window classification of each implanted marker. Three cylindrical marker training datasets were generated from phantom kilovoltage (kV) and patient intrafraction images with increasing levels of megavoltage (MV) scatter. The cylindrical shaped marker CNNs were validated on unseen kV fluoroscopic images from 12 fractions of 10 prostate cancer patients with implanted gold fiducials. For the training and validation of the arbitrarily shaped marker CNNs, cone beam computed tomography (CBCT) projection images from ten fractions of seven lung cancer patients with implanted coiled markers were used. The arbitrarily shaped marker CNNs were trained using three patients and the other four unseen patients were used for validation. The effects of full training using a compact CNN (four layers with learnable weights) and transfer learning using a pretrained CNN (AlexNet, eight layers with learnable weights) were analyzed. Each CNN was evaluated using a Precision-Recall curve (PRC), the area under the PRC plot (AUC), and by the calculation of sensitivity and specificity. The tracking system was assessed using the validation data and the accuracy was quantified by calculating the mean error, root-mean-square error (RMSE) and the 1st and 99th percentiles of the error. RESULTS: The fully trained CNN on the dataset with moderate noise levels had a sensitivity of 99.00% and specificity of 98.92%. Transfer learning of AlexNet resulted in a sensitivity and specificity of 99.42% and 98.13%, respectively, for the same datasets. For the arbitrarily shaped marker CNNs, the sensitivity was 98.58% and specificity was 98.97% for the fully trained CNN. The transfer learning CNN had a sensitivity and specificity of 98.49% and 99.56%, respectively. The CNNs were successfully incorporated into a multiple object tracking system for both cylindrical and arbitrarily shaped markers. The cylindrical shaped marker tracking had a mean RMSE of 1.6 ± 0.2 pixels and 1.3 ± 0.4 pixels in the x- and y-directions, respectively. The arbitrarily shaped marker tracking had a mean RMSE of 3.0 ± 0.5 pixels and 2.2 ± 0.4 pixels in the x- and y-directions, respectively. CONCLUSION: With deep learning CNNs, high classification performances on unseen patient images were achieved for both cylindrical and arbitrarily shaped markers. Furthermore, the application of CNN models to intrafraction monitoring was demonstrated using a simple tracking system. The results demonstrate that CNN models can be used to track markers without prior knowledge of the marker properties or an additional learning period.

Investigating multi-leaf collimator tracking in stereotactic arrhythmic radioablation (STAR) treatments for atrial fibrillation.

Stereotactic arrhythmia radioablation (STAR) is an emerging treatment option for atrial fibrillation (AF). However, it faces possibly the most challenging motion compensation scenario: both respiratory and cardiac motion. Multi-leaf collimator (MLC) tracking is clinically used for lung cancer treatments but its capabilities with intracardiac targets is unknown. We report the first experimental results of MLC tracking for intracardiac targets. Five AF STAR plans of varying complexity were created. All delivered 5 × 10 Gy to both pulmonary vein antra. Three healthy human target motion trajectories were acquired with ultrasound and programmed into a motion platform. Plans were delivered with a linac to a dosimeter placed on the motion platform. For each motion trace, each plan was delivered with no MLC tracking and with MLC tracking with and without motion prediction. Dosimetric accuracy was assessed with γ-tests and dose metrics. MLC tracking improved the dosimetric accuracy in all measurements compared to non-tracking experiments. The average 2%/2 mm γ-failure rate was improved from 13.1% with no MLC tracking to 5.9% with MLC tracking (p < 0.001) and 7.2% with MLC tracking and no motion prediction (p < 0.001). MLC tracking significantly improved the consistency between planned and delivered target dose coverage. The 95% target coverage with the prescription dose (V100) was improved from 60% of deliveries with no MLC tracking to 80% of deliveries with MLC tracking (p = 0.03). MLC tracking was successfully implemented for the first time for intracardiac motion compensation. MLC tracking provided significant dosimetric accuracy improvements in AF STAR experiments, even with challenging cardiac and respiratory-induced target motion and complex treatment plans. These results warrant further investigation and optimisation of MLC tracking for intracardiac target motion compensation.