Y. Tsunashima

Efficiency of respiratory-gated delivery of synchrotron-based pulsed proton irradiation

Significant differences exist in respiratory-gated proton beam delivery with a synchrotron-based accelerator system when compared to photon therapy with a conventional linear accelerator. Delivery of protons with a synchrotron accelerator is governed by a magnet excitation cycle pattern. Optimal synchronization of the magnet excitation cycle pattern with the respiratory motion pattern is critical to the efficiency of respiratory-gated proton delivery. There has been little systematic analysis to optimize the accelerators operational parameters to improve gated treatment efficiency. The goal of this study was to estimate the overall efficiency of respiratory-gated synchrotron-based proton irradiation through realistic simulation. Using 62 respiratory motion traces from 38 patients, we simulated respiratory gating for duty cycles of 30%, 20% and 10% around peak exhalation for various fixed and variable magnet excitation patterns. In each case, the time required to deliver 100 monitor units in both non-gated and gated irradiation scenarios was determined. Based on results from this study, the minimum time required to deliver 100 MU was 1.1 min for non-gated irradiation. For respiratory-gated delivery at a 30% duty cycle around peak exhalation, corresponding average delivery times were typically three times longer with a fixed magnet excitation cycle pattern. However, when a variable excitation cycle was allowed in synchrony with the patients respiratory cycle, the treatment time only doubled. Thus, respiratory-gated delivery of synchrotron-based pulsed proton irradiation is feasible and more efficient when a variable magnet excitation cycle pattern is used.

The precision of respiratory-gated delivery of synchrotron-based pulsed beam proton therapy

A synchrotron-based proton therapy system operates in a low repetition rate pulsed beam delivery mode. Unlike cyclotron-based beam delivery, there is no guarantee that a synchrotron beam can be delivered effectively or precisely under the respiratory-gated mode. To evaluate the performance of gated synchrotron treatment, we simulated proton beam delivery in the synchrotron-based respiratory-gated mode using realistic patient breathing signals. Parameters used in the simulation were respiratory motion traces (70 traces from 24 patients), respiratory gate levels (10%, 20% and 30% duty cycles at the exhalation phase) and synchrotron magnet excitation cycles (T(cyc)) (fixed T(cyc) mode: 2.7, 3.0-6.0 s and each patient breathing cycle, and variable T(cyc) mode). The simulations were computed according to the breathing trace in which the proton beams were delivered. In the shorter fixed T(cyc) (<4 s), most of the proton beams were delivered uniformly to the target during the entire expiration phase of the respiratory cycle. In the longer fixed T(cyc) (>4 s) and the variable T(cyc) mode, the proton beams were not consistently delivered during the end-expiration phase of the respiratory cycle. However we found that the longer and variable T(cyc) operation modes delivered proton beams more precisely during irregular breathing.

MO‐FF‐A3‐02: What Is the Maximum Number of Beam Spots Deliverable within One Gating Window for Synchrotron Based Scanning Proton Beam Therapy of Lung Cancer?

Purpose: To analyze the maximum number of spots achievable for each energy layer in respiratory‐gated scanning beam delivery of protonradiotherapy when optimizing efficiency and minimizing the impact of cycle‐to‐cycle breathing irregularity. The ideal delivery would be to deliver all spots in each energy level within a single respiratory gate. Method and Materials: In‐house treatmentdelivery simulation software was developed based on manufacturers specification to study treatment statistics using timing information from patients respiration patterns, and scanning beam protontreatment plans. Data from 10 patients with lungtumor were analyzed.Proton scanning beam plans were developed using 3 beams/plan. The patients breathing traces were used to simulate respiratory gated delivery with 30%, 20% and 10 % duty cycles (DC). In the simulation, we measured 1) typical respiratory gate duration, and 2) the number of spots within this duration (spill time) in each expiration phase. We also compared the above data with number of available spots without respiratory gating. Results: The average duration of each respiratory gate signal was 1161±475, 841±323 and 517±208 msec for 30%, 20% and 10 % DCs respectively. This translated to 166, 120 and 74 available spots for 30%, 20% and 10 % DC respectively. Simulation of respiratory‐gated scanning beam delivery over 10 respiratory patterns and 10 tumors indicated that, maximum number of spots required per energy level were 85, 83 and 77 for 30%, 20% and 10% DC respectively. Conclusion:Treatment plans for respiratory‐gated scanning beam protondelivery on a synchrotron accelerator should be designed such that the number of available spots energy level are less than can be delivered in the respiratory gate. Our simulation study indicated that this condition could be easily met for various DCs of respiratory gating.

SU‐FF‐T‐132: Efficiency of Respiratory‐Gated Synchrotron Based Scanning Beam Delivery of Proton Therapy for Lung

Purpose: To quantify the efficiency of respiratory‐gated synchrotron based scanning beam delivery of proton irradiation for lungtumors.Method and Materials: An in‐house software program was developed to simulate the interaction of the respiratory‐gate signal based on respiratory traces of patients from an external respiratory monitor and the synchrotron magnet operation pattern in scanning beam protondelivery. A simulation study was performed using 10 proton scanning beam plans (total 33 beam angles) for lungtumors. For each plan, 10 patient breathing traces were used to simulate respiratory gated delivery. The total time to deliver each beam was determined for non‐gated and gated modes (30%, 20% and 10 % duty cycles respectively). Efficiency was determined using the effective dose rate (MU/min). The relationship between effective dose rate with and without gating was examined for different irradiated volumes along each beam direction and the number of control points in each plan. Results: The average effective dose rate was 34.2±18.8 (MU/min) for non‐gated treatment. This represented an approximately three fold reduction as compared to the passively‐scattered protondelivery (90.1 MU/min). With gating, effective dose ratios were further reduced to 13.8±4.7, 9.3±3.6 and 3.7±1.7 MU/min for 30%, 20% and 10 % duty cycles respectively. Using an average MU per beam at 70.9, the estimated treatment delivery time for a single beam was 2.1 min for non‐gated treatment, and 5.1, 7.6, and 19.1 min for gated treatment with 30%, 20% and 10 % duty cycles, respectively. No strong relationship was found between the effective dose rate and the irradiated volume or the number of control points. Conclusion: Respiratory gated delivery of pulsed scanning beam proton therapy for lungcancer is less efficient than passively‐scattered proton irradiation. Further studies are needed to improve delivery efficiency in the treatment plan optimization using patient specific breathing patterns.

WE‐E‐BRB‐03: Precision of Respiratory‐Gated Delivery of Synchrotron‐Based Proton Therapy

Purpose: To investigate the precision of synchrotron based passively scattered respiratory gated delivery of proton irradiation. A simulation study was undertaken for the analysis of the residual target motion uncertainty during synchrotron based respiratory gated proton treatment. Method and Materials: In‐house software was developed to simulate a synchrotron based respiratory gated proton treatment. The interactions of respiratory motion traces (70 traces, 22 patients), respiratory gate threshold levels (10%, 20% and 30% duty cycle around peak exhalation) and synchrotron Tcyc patterns (fixed Tcyc =2.7, 3.0 ∼ 6.0 second, average patient breathing‐cycle and variable Tcyc) were plotted along the same time scale, similar to an oscilloscope display. Proton beam delivery within a gate threshold only occurred during a portion of each Tcyc pattern. A specific pattern of Tcyc acts as a “gate‐within‐a‐gate”, which produces a smaller effective gating window. Precision of respiratory gated protondelivery was analyzed by examining distribution of distance from the respiratory gate threshold where 95% of gated beam delivery (DGT95) occurred. Results: With shorter fixed Tcyc ( 4 sec) average DGT95 values were 0.25 cm, 0.19 cm and 0.14 cm respectively for 30%, 20% and 10% respiratory gate duty cycles. With Tcyc ≈ average patient breathing cycle, average DGT95 values were 0.27 cm, 0.20 cm and 0.15 cm for 30%, 20% and 10% respiratory gate duty cycles. With variable Tcyc, DGT95 values were 0.21 cm, 0.17 cm and 0.14 cm for 30%, 20% and 10% respiratory gate duty cycles. Conclusion: Variable Tcyc mode offered the greatest precision of respiratory gated delivery for passively scattered synchrotron proton irradiation.

TH‐C‐M100E‐06: Determining Optimal Respiratory Gating Parameters for Passively Scattered Synchrotron Based Proton Irradiation

Purpose: Respiratory gated irradiation offers potential for margin reduction and dose escalation for treating moving tumors in the thorax or abdomen. Unfortunately, for synchrotron‐based proton irradiation, it may not be efficient. We have determined the optimal respiratory gating parameters for passively scattered proton irradiation on a synchrotron through a simulation study. Method and Materials: An in‐house software program was developed to investigate the interaction of the respiratory gating intervals with different synchrotron magnet excitation cycle patterns. Test data was obtained by using the recorded respiratory trace of 94 patients who underwent 4DCT. A typical magnet excitation cycle, Tcyc consists of proton acceleration, flat top and deceleration periods. Proton beam delivery occurs only during the flat top portion of each such excitation cycle. Respiratory gating was simulated at expiration for a 30% duty cycle around peak exhalation. The time required to deliver 100 MUs was estimated for the following scenarios: (a) Ungated irradiation with Tcyc set to the minimum value (2.7sec) and (b) Gated irradiation with Tcyc set to (i) the minimum value, (ii) approximately equal each patients average respiratory cycle, and (iii) a variable value according to each individual respiratory cycle. Overall treatment time and efficiency of treatment delivery were studied in each case. Results: Average times required to deliver 100 MUs were 1.1 minutes for ungated irradiation; and 3.7 (1.7 – 6.0), 3.2 (1.6 – 7.1), 2.3 (1.4 – 3.1) minutes respectively for gated irradiations at various scenarios mentioned above. For gated irradiation, variable Tcyc mode of operation yielded least overall treatment time and greatest efficiency of proton beam delivery. Conclusion: Respiratory gated passively scattered proton delivery using a synchrotron‐based system is feasible without significantly increasing treatment time. Based on above results, variable Tcyc mode of operation offered least overall treatment time and greatest efficiency for respiratory gated irradiation.

SU‐FF‐J‐43: Correlation Between External Abdominal and Internal Liver Fiducial Motion in 4D‐CT

Purpose: to determine how well the anterior‐posterior location of a marker, placed on a supine patients abdominal surface, correlates with the location of a radio‐opaque fiducial implanted within the liver. We investigate which respiratory phases exhibit the best correlation between the marker and the fiducial. Methods and Materials: data was obtained from five patients, each having a fiducial (either an implanted gold pellet or the tip of a stent) in the liver. Each patient received a cine‐mode 4D‐CT scan of the liver, with the position of an external marker used to associate each cine CTimage with a specific respiratory phase. The trace of the anterior‐posterior motion of the marker was normalized under the assumption that, among ten phases of a full respiratory cycle (0%, 10%, 20% … 90%), the marker displacements correlated linearly with the fiducials superior‐inferior displacements. Each cycle of the normalized marker trace was then superimposed upon a plot of the fiducials displacement; the latter was measured within the 4D‐CT data set for each of the ten phases. Results: comparisons of the marker trace and the fiducial coordinate from 4D‐CT indicated that, in general, the marker motion correlated reasonably well with internal liver motion during the process of exhalation (from end‐inspiration to end‐expiration), with optimum correlation during end‐expiration (40% to 60% phase). The correlation tends to be poorer during inhalation, extending from 70% to 90% phase. Conclusions: a tight correlation near end‐expiration suggests that using an external‐motion‐based respiratory trace, as a guide for gated treatments for livercancer, should enable reliable and reproducible coverage of the target volume. Poorer correlation during inhalation suggests that the markers anterior‐posterior motion may not adequately characterize the internal motion during those phases, and that gating should be avoided during this stage of the respiratory cycle.