Pencil Beam Scanning Research Articles

Extended in-field and out-of-field validation of a compact Monte Carlo model of an IBA PROTEUS®ONE proton beam in TOPAS/GEANT4

Objective: This study evaluates a compact Monte Carlo (MC) model of a pencil beam scanning clinical proton beam using TOPAS to estimate the dose out-of-field (OOF). Compact modelling means that the model starts from a pristine proton beam at the nozzle exit, customised based on acceptance and commissioning data, instead of modelling the full treatment head and room. Approach: First, in-field validation tests were performed. Then, the OOF dose was validated in an RW3 phantom with bubble detectors for personal neutron dosimetry (measuring the neutron dose equivalent) and thermoluminiescent detectors (measuring the absorbed dose by protons and gammas). Measurements were performed at 15 and 35 cm from the distal edge of the field for five different irradiation plans, covering different beam orientations, proton energies and a 40 mm range shifter. TOPAS simulations were performed with QGSP Binary Cascade HP (BIC) and QGSP Bertini HP (Bertini) hadron physics lists. Main results: In-field validation shows that MC simulations agree with point dose measurements within −2.5 % and +1.5 % at locations on- and off-axis and before, in and after the Bragg peak or plateau. The gamma passing rate 2%/3mm of four simulated treatment plans compared to the dose distribution calculated by the TPS exceeds 97 % agreement score. OOF dose simulations showed an average overestimation of 27 % of the neutron dose equivalent for the BIC hadron physics list and an average underestimation of 20 % for the Bertini hadron physics list. The simulated absorbed dose of protons and gammas showed a systematic underestimation which was on average 21 % and 51 % for BIC and Bertini respectively. Significance: Our study demonstrates that a compact MC model can reliably produce in-field data, while out-of-field dose data are within the uncertainties of the detector systems and MC simulations nuclear models, and do so with shorter modelling and faster calculation time.

Integration and dosimetric validation of a dynamic collimation system for pencil beam scanning proton therapy

Objective. To integrate a Dynamic Collimation System (DCS) into a pencil beam scanning (PBS) proton therapy system and validate its dosimetric impact. Approach. Uncollimated and collimated treatment fields were developed for clinically relevant targets using an in-house treatment plan optimizer and an experimentally validated Monte Carlo model of the DCS and IBA dedicated nozzle (DN) system. The dose reduction induced by the DCS was quantified by calculating the mean dose in 10- and 30-mm two-dimensional rinds surrounding the target. A select number of plans were then used to experimentally validate the mechanical integration of the DCS and beam scanning controller system through measurements with the MatriXX-PT ionization chamber array and EBT3 film. Absolute doses were verified at the central axis at various depths using the IBA MatriXX-PT and PPC05 ionization chamber. Main results. Simulations demonstrated a maximum mean dose reduction of 12% for the 10 mm rind region and 45% for the 30 mm rind region when utilizing the DCS. Excellent agreement was observed between Monte Carlo simulations, EBT3 film, and MatriXX-PT measurements, with gamma pass rates exceeding 94.9% for all tested plans at the 3%/2 mm criterion. Absolute central axis doses showed an average verification difference of 1.4% between Monte Carlo and MatriXX-PT/PPC05 measurements. Significance. We have successfully dosimetrically validated the delivery of dynamically collimated proton therapy for clinically relevant delivery patterns and dose distributions with the DCS. Monte Carlo simulations were employed to assess dose reductions and treatment planning considerations associated with the DCS.

The Applications and Pitfalls of Cone-Beam Computed Tomography-Based Synthetic Computed Tomography for Adaptive Evaluation in Pencil-Beam Scanning Proton Therapy.

The study evaluates the efficacy of cone-beam computed tomography (CBCT)-based synthetic CTs (sCT) as a potential alternative to verification CT (vCT) for enhanced treatment monitoring and early adaptation in proton therapy. Seven common treatment sites were studied. Two sets of sCT per case were generated: direct-deformed (DD) sCT and image-correction (IC) sCT. The image qualities and dosimetric impact of the sCT were compared to the same-day vCT. The sCT agreed with vCT in regions of homogeneous tissues such as the brain and breast; however, notable discrepancies were observed in the thorax and abdomen. The sCT outliers existed for DD sCT when there was an anatomy change and for IC sCT in low-density regions. The target coverage exhibited less than a 5% variance in most DD and IC sCT cases when compared to vCT. The Dmax of serial organ-at-risk (OAR) in sCT plans shows greater deviation from vCT than small-volume dose metrics (D0.1cc). The parallel OAR volumetric and mean doses remained consistent, with average deviations below 1.5%. The use of sCT enables precise treatment and prompt early adaptation for proton therapy. The quality assurance of sCT is mandatory in the early stage of clinical implementation.

The implementation of an image-guided system at a proton therapy center facility.

Pencil Beam Scanning (PBS) proton therapy has similar requirements on patient alignment to within 1mm and 1-degree accuracy as photon radiosurgery. This study describes general workflow, acceptance, and commissioning test procedures and their respective results for an independent robotic arm used for Image Guided Radiotherapy (IGRT) for a Proton Therapy System. The system is equipped with kV-imaging techniques capable of orthogonal and Cone-Beam Computed Tomography (CBCT) imaging modalities mounted on an independent robotic arm gantry attached to the ceiling. The imaging system is capable of 360-degree rotation around patients to produce CBCT and kilovoltage orthogonal images. The imaging hardware is controlled by Ehmet Health XIS software, and MIM Software handles the image fusion and registration to an acceptable accuracy of ≤1-mm shifts for patients' alignment. The system was tested according to the requirements outlined in the American Association of Physicists in Medicine (AAPM) Task Group (TG) 142 and TG 179. The system tests included (1) safety, functionality, and connectivity, (2) mechanical testing, (3) image quality, (4) image registration, and (5) imaging dose. Additional tests included imaging gantry isocentricity with a laser tracker and collision-avoiding system checks. The orthogonal and volumetric imaging are comparable in quality to other commercially available On-Board Imagers (OBI) systems. The resulting spatial resolution values were 1.8-, 0.8-, and 0.5-Line Pairs per Millimeter (lp/mm) for orthogonal, full-fan CBCT, and half-fan CBCT, respectively. The image registration is accurate to within 1mm and 1 degree. The data shows consistent imaging-guided system performance with standard deviations in x, y, and z of 0.7, 0.8, and 0.7mm, respectively. The system provides excellent image quality and performance, which can be used for IGRT. The proven accuracy of the x-ray imaging and positioning system at McLaren Proton Therapy Center (MPTC) is 1mm, making it suitable for proton therapy.

Breast Reconstruction Complications After Postmastectomy Proton Radiation Therapy for Breast Cancer

Our purpose was to report complications requiring surgical intervention among patients treated with postmastectomy proton radiation therapy (PMPRT) for breast cancer in the setting of breast reconstruction (BR). Patients enrolled on a prospective proton registry who underwent BR with immediate autologous flap, tissue expander (TE), or implant in place during PMPRT (50/50.4 Gy +/- chest wall boost) were eligible. Major reconstruction complication (MRC) was defined as a complication requiring surgical intervention. Absolute reconstruction failure was an MRC requiring surgical removal of BR. A routine revision (RR) was a plastic surgery refining cosmesis of the BR. Kaplan-Meier method was used to assess disease outcomes and MRC. Cox regression was used to assess predictors of MRC. Seventy-three courses of PMPRT were delivered to 68 women with BR between 2013 and 2021. Median follow-up was 42.1 months. Median age was 47 years. Fifty-six (76.7%) courses used pencil beam scanning PMPRT. Of 73 BR, 29 were flaps (39.7%), 30 implants (41.1%), and 14 TE (19.2%) at time of irradiation. There were 20 (27.4%) RR. There were 9 (12.3%) MRC among 5 implants, 2 flaps, and 2 TE, occurring a median of 29 months from PMPRT start. Three-year freedom from MRC was 86.9%. Three (4.1%) of the MRC were absolute reconstruction failure. Complications leading to MRC included capsular contracture in 5, fat necrosis in 2, and infection in 2. On univariable analysis, BR type, boost, proton technique, age, and smoking status were not associated with MRC, whereas higher body mass index trended toward significance (hazard ratio, 1.07; 95% CI, 0.99-1.16; P = .10). Patients undergoing PMPRT to BR had a 12.3% incidence of major complications leading to surgical intervention, and total loss of BR was rare. MRC rates were similar among reconstruction types. Minor surgery for RR is common in our practice.

Proton Radiation Therapy for Stage IIA/IIB Testicular Seminoma.

Testicular seminoma affects young men and is associated with very favorable prognosis. The evolution in treatment paradigm has focused on minimizing acute and especially late toxicities. Following orchiectomy, while surveillance is favored in Stage I patients, radiotherapy (RT) is a standard treatment option for de novo or relapsed stage IIA or select non-bulky stage IIB disease. Despite low doses, standard RT fields to paraaortic and pelvic lymphatics using x-rays exposes a large volume of uninvolved normal tissue/viscera to excess dose. This young patient population is especially vulnerable to risks of late RT toxicities including secondary malignancy. Proton beam therapy (PBT) has dosimetric advantage over x-ray-based RT due to lack of exit dose, and comparative dosimetric/modeling studies show significant sparing of uninvolved abdominal/pelvic organs. However, there is scant reported clinical data at this time for PBT. We review our early institutional outcomes with PBT for testicular seminoma. Single institution retrospective review from a tertiary care center of patients treated with PBT from 2013-2022 for testicular seminoma. Recurrence free (RFS) and overall survival (OS) were calculated from the completion of PBT. Toxicities were graded (Gr) using CTCAE v5.0. Four patients underwent PBT, median age 39 (range 36-47). All were Stage I at diagnosis (pT1b n = 3; pT2 n = 1) and were treated for recurrent stage II disease (IIA n = 3; IIB n = 1) at a median of 34 months from orchiectomy (range 3 - 74 months). Nodal extent included 2 with multiple paraaortic nodes, 1 with solitary paraaortic node and 1 with solitary pelvic node. PBT was delivered with pencil-beam scanning, treating paraaortic + ipsilateral pelvic fields (20 Gy in 10 fractions), then sequential boost to involved nodes (10 -16 Gy in 5-8 fractions). Typically, PA or posterior oblique fields were used to minimize dose to out-of-field abdominal/pelvic viscera. Treatment was well tolerated with minimal acute toxicities: fatigue Gr 1 (n = 3), nausea Gr 1 (n = 3). No Gr 2 or higher acute toxicities or significant late toxicities were observed. At median follow up of 30 months (range 3 - 54), no recurrences were observed, and RFS and OS were 100%. Two patients are without evidence of disease > 4 years post-treatment. In this case series, PBT for retroperitoneal and pelvic metastases in Stage IIA/IIB testicular seminoma was associated with oncologic efficacy with minimal toxicity. PBT reduces unnecessary dose to abdominal/pelvic organs compared to x-ray techniques, which is advantageous in young patients who have anticipated long-term survival. This is one of the few series reporting clinical outcomes of PBT in the management of seminoma. Randomized comparisons with x-ray approaches are impractical given the relatively low volume of patients receiving RT in modern seminoma management, so it is essential to report and track longitudinal outcomes across institutions to validate this approach.

Proton Therapy for Ocular Melanoma: A Dosimetric Comparison of Three Different Delivery Systems

The purpose of this study is to compare the treatment plans for intraocular tumors with two types of Pencil Beam Scanning (PBS) machine configurations and compare them with the widely used uniform scanning (US)/scattering technique. Two virtual PBS machines were constructed for ocular planning. The first machine (P1) consisted of a standard spot size and a 7.5cm WET range shifter (RS) positioned 50cm upstream from the aperture. The second machine (P2) had a smaller spot size and used a 4cm WET RS but had no aperture support. These PBS delivery systems resemble currently available models. Ten patients with different locations and sizes of intraocular tumors were planned with US using 2-3 conformal fields following our institution's intraocular criteria. PBS spot patterns for each machine were optimized using the same beam geometry as the US plan to achieve identical tumor coverage. PBS distribution was calculated using Monte Carlo. All air gaps were minimized as small as clinically achievable. PBS plans passed robustness criteria with all scenarios meeting CTV D95>95% with 2mm translational offsets and 3.5% range uncertainty. The average conformity index (CI) of 95% reference isodose was better for both PBS techniques (P1: 1.43±0.22, P2:1.5±0.32) versus US (1.72±0.56) for all plans (p = 0.025, p = 0.034). The average CI of 50% reference isodose was worse for the P2 plans (10.4±3.3) compared to US (9.34±2.93). However, CI 50% for P1 plans was better than US (5.21±1.24, p<0.003). There were no significant improvements to OARs using P2. However, the mean dose to anterior structures ciliary body, lacrimal gland, and D0.03cc lens were significantly lower in the P1 plans than in the US plans (p<0.016). Small spot size PBS systems without apertures can achieve similar coverage to US plans but have higher OAR dose. PBS systems with a standard spot size and reconfigured RS and apertures are better for anterior OAR sparing and is non-inferior to US planning of intraocular targets.

Outcomes Following Proton Therapy for Non-Metastatic Central Nervous System Germinoma in Children and Adolescents

Radiation is a key component in the treatment of central nervous system pure germinoma (PG) in children and adolescents. Compared to photons, proton therapy (PT) improves normal tissue sparing and potentially reduces adverse effects, but there are sparse data on long-term patient outcomes. Herein, we present the largest reported single institution experience utilizing PT for the management of PG. We performed an IRB-approved retrospective review of a prospective database containing 35 patients with non-metastatic PG treated with PT between July 2007 and September 2021. The median age at treatment was 13 years. All patients had > 6 months of follow up. Two patients were diagnosed based on cerebral spinal fluid B-HCG levels and the remaining 33 patients through histopathology. Eleven patients had bifocal or multifocal intracranial disease. Most patients (n = 31, 88.6%) received induction chemotherapy with carboplatin + etoposide (n = 25, 80.6%) with all demonstrating a radiographic response to neoadjuvant therapy. Twenty-nine were treated with whole ventricular irradiation (WVI) with an involved field (IF) boost, 2 with craniospinal irradiation + WVI + IF, 2 with IF, 1 with CSI + IF, and 1 with whole brain + IF. Among this subgroup, the most common total dose was 30 GyRBE (n = 18, 51.4%). Of the 4 patients that did not receive chemotherapy, 3 received WVI + IF and 1 was treated with CSI + WVI + IF. The total dose in the patients not receiving chemotherapy was 45 GyRBE. Twenty-nine patients were treated with double scattered PT (DS) and 6 patients with pencil beam scanning PT (PBS). We utilized the cumulative incidence method to estimate local control (LC), freedom from distant metastases (FFDM), freedom from progression (FFP), and overall survival (OS). Treatment related toxicity was assessed according to the CTCAEv5. Median follow up was 6.2 years (Range: 0.9-15.2 years). The 10-year KM estimates for LC, FFDM, FFP, and OS were 100%, 100%, 100%, and 97% respectively. One patient developed acute lymphocytic leukemia 8 months after treatment and died. The sole ³ grade 3 radiation-related toxicity was a cavernoma with hemorrhage 3.6 years post treatment requiring surgical resection. The most common adverse events were hearing impairment requiring hearing aids (n = 3), hypersomnia requiring medication (n = 3), and new onset endocrinopathy (n = 1). Of the 22 evaluable patients ³18 years old at last follow up, 7 were high school graduates/in college, 8 college graduates, and 5 others gainfully employed. The use of proton therapy in the multi-modality approach to non-metastatic PG does not compromise local control. Although serious side effects are rare and treatment has minimal impact on survivors' early educational/career trajectory, the 100% cure rate supports further investigation into selective radiation dose and volume de-escalation.

Pre-Clinical Validation of Targeting Cholesterol Metabolism for Radiotherapy Sensitization in Head and Neck Cancer

This study examines the potential benefits of targeting cholesterol metabolism in combination with radiation therapy (RT) for treating Head and Neck Squamous Cell Carcinoma (HNSCC). Despite advancements in RT delivery, 60% of patients still experience a relapse within 2 years, highlighting the need for improved treatment strategies. Our hypothesis is that a specific, biological response to RT by a subset of tumor cells drives radio-resistance and recurrence, and that targeting cholesterol metabolism may be an effective way to improve RT efficacy. To identify radiation-induced biological responses, whole-genome RNA sequencing analysis was carried out on Cal27 (HPV-) and UMSCC47 (HPV+) cell lines after X-Ray (6MeV Linac) or proton (pencil beam scanning, center part of SOBP) radiations. Potential radio-resistance targets were identified by comparing RT-induced transcriptomic changes with publicly available GEO and TCGA transcriptomic data from human HNSCC samples. Cholesterol metabolism as a driver of HNSCC radio-resistance was validated in vitro by pharmacological and genetic approaches. Transcriptomic analysis revealed a significant increase in 11 genes related to sterol and cholesterol in both the Cal27 and UMSCC47 cell lines after radiation exposure (pathway functional enrichment analysis, p = 2.05e-05). High expression of these RT-induced cholesterol gene signatures correlated with decreased overall survival among HNSCC patients based on the TCGA database (log rank p = 0.022, HR = 1.4). Moreover, a Gene Set Enrichment Analysis of the GSE23036 dataset (pre-treatment HNSCC tumors: locoregional recurrence vs. nonrecurrence) also linked cholesterol metabolism and local recurrence (p = 0.002, FDR = 0.07, NES = 1.72). Single-cell RNA sequencing data from the GSE164690 dataset revealed high expression of cholesterol metabolism genes in human HNSCC epithelial and mesenchymal fractions. In vitro studies demonstrated that RT increased the nuclear translocation of the master transcription factors controlling cholesterol gene expression SREBP2 and increased the expression of the SREBP2 target gene SQLE at the protein level. In accordance with this result, RT increased cellular cholesterol content and uptake in human (p = 0.027) and murine (p = 0.015) HNSCC cell lines. Drug or genetic targeting of SREBP2 reduced cellular cholesterol levels in HNSCC cells and increased toxicity when combined with RT in both 2D and 3D cell cultures compared to RT alone. Targeting cholesterol metabolism in combination with RT could be beneficial for treating HNSCC. In vivo validation in immunocompetent mouse models and orthotopic patient-derived xenograft models is ongoing.

Cardiopulmonary Toxicity from Intensity Modulated Proton Therapy for Thymic Malignancies

Use of radiation therapy for thymic malignancies is limited by excess dose to organs at risk (OARs) including heart, lung, and esophagus. Intensity Modulated Proton Therapy with Pencil Beam Scanning (IMPT/PBS) allows the conformality benefits of volumetric modulated arc therapy (VMAT) combined with dosimetry benefits of protons making it an exciting tool to treat thymic tumors. Very limited clinical data are reported with the use of IMPT/PBS to treat thymic malignancies. This study evaluates the incidence of acute and delayed toxicities among patients who underwent IMPT/PBS for thymic tumors. Our single center retrospective study identified 27 patients with diagnosis of either thymic carcinoma or thymoma who received IMPT/PBS between 2015 and 2022. Patient demographics, IMPT treatment details and clinical outcomes (toxicity, recurrence, and survival) were recorded. Frequency distributions are described for primary endpoints of acute (≤ 90 days) and late (>90 days) toxicity graded using CTCAE version 5.0. Specific toxicities assessed were dermatitis, esophagitis, pneumonitis, pulmonary fibrosis, and cardiac toxicity. Recurrence and survival data were analyzed as secondary endpoint using Kaplan-Meier method. Median follow-up was 22 months. Median age of the patients was 59 years (range, 30-87), predominantly female (55%), and white (66%), and stage ¾ (72%). Histologically showed mainly thymoma (59%) and Masaoka stage ¾ (70%). Surgery prior to IMPT was performed in 19 (70%) patients; of whom 5 patients had positive margins. Chemotherapy was used in 12 (44%) patients. Median IMPT dose was 50.4 GyE. Patients were primarily planned with 2 or 3 fields (81%), coplanar distribution (74%), using SFO technique (70%). Robust planning was performed accounting for 5 mm margin and 3.5-5% range uncertainty. All patients required use of range shifter ranging from 2-5 cm. Median of Heart-mean (10.3 GyE), Heart-max (54 GyE), Lung-mean (8.1 GyE), Lung V20 Gy (16.1%), Lung-V30 Gy (11.4%), Esophagus-mean (10.3 GyE) and Cord-Max (1.4 GyE). QACT was performed in 21 (77%) patients with replan needed in 5 of them. Only 1 (3.7%) patient had a grade 3+ acute toxicity (dermatitis) and only 2 (7.4%) patients had a grade 3+ late toxicity (both pulmonary). No patients had any acute or delayed cardiac-related adverse effects following PBT treatment. One (3.7%) patient had an infield recurrence of malignancy and 6 had out-of-field metastatic failure. Local control and overall survival were 74.1% and 85.2%, respectively. In this largest single-institution analysis of IMPT/PBS experience, we note extremely low incidence of grade 3+ acute or late toxicity with excellent local control and overall survival. No marginal failures were noted. In a patient population at high risk of cardiopulmonary radiation toxicities, IMPT/PBS should be strongly explored as a possible treatment option.

Five-Fraction Proton Therapy for Chordomas and Chondrosarcomas of the Skull Base: Preliminary Results of a Prospective Series

To describe the first series of patients with clival chordomas and chondrosarcomas treated with extremely hypofractionated proton therapy in a single institution with focus on acute tolerance and dosimetric parameters. A total of 11 consecutive patients with biopsy proven chordomas and chondrosarcomas were prospectively included in a five-fraction proton therapy protocol between June 2022 and January 2023. The inclusion criteria for this protocol were: Age > 18 yo., Karnofsky Performance Status (KPS) ≥ 70%, absence of metastases, no previous radiation courses, clinical target volume (CTV) up to 50 cc and CTV location relative to organs at risk that allowed compliance with the necessary dose restrictions for hypofractionated treatments in 5 fractions, according to the international guidelines. Treatment was delivered with a Proteus®ONE cyclotron using Pencil Beam Scanning (PBS) Intensity-Modulated Proton Therapy (IMPT). Patient setup was performed with CBCT imaging at isocenter and X-Ray obliques, allowing a robust planning with 1 mm setup uncertainty and 3.5% range uncertainty. IMPT plans were obtained with RayStation employing 4 to 6 non-coplanar beams. In 7 patients' apertures were used in 2 of the beams, to reduce lateral penumbra and to optimize dose gradient. A total of 11 patients (7 males and 5 females) with an age range between 29 and 79 yo, were included. 6 patients with classic chordomas and 5 patients with grade I-II chondrosarcomas were treated with a total dose of 37.5 GyRBE and 35 GyRBE, respectively, in 5 daily fractions. All the patients had received previous surgery with complete resection in 2 patients, subtotal resection in 7 patients and partial resection in 2 patients. Most common symptoms at diagnosis were diplopia and headache and baseline median KPS after surgery was 90% (70 - 100%). Median CTV was 30.62 cc (10.56 - 47.47 cc). Median CTV coverage was V95% = D95.5%. Median maximal dose for the brainstem was 26.5 GyRBE (30 - 24.7 GyRBE) and for the optic pathway was 20.7 GyRBE (8.9 - 23.6 GyRBE). Median follow-up was 3 months (8 - 1 months). Acute toxicity during follow-up was mild, with grade I - II headache (64%), grade I asthenia (45%), grade I nausea (27%), grade I dysphagia (18%) and reversible alopecia (45%). Only one patient showed nystagmus and mild loss of strength in left lower limb related with an ischemic brainstem lesion, in a low dose region, not clearly radiation related. 8 patients had a 3-month follow-up MRI with no signs of progression. Five-fraction proton therapy for the treatment of clival chordomas and chondrosarcomas is dosimetrically feasible and well tolerated for selected patients. Proton therapy remains a limited resource, consequently, reducing treatment time can have significant financial and psychosocial implications. Longer follow up is needed to validate these results.

Relationship Between Spatial Distribution Tumor Location and Dosimetric Outcomes in Patients Receiving Definitive Proton Therapy for Locally Advanced Lung Cancer

Intensity Modulated Proton Therapy with Pencil Beam Scanning (IMPT/PBS) has been postulated to have dosimetric benefits, especially for sparing of normal tissues in the treatment of thoracic malignancies. Distinguishing which patients may benefit most from IMPT/PBS treatment has important implications when counseling patients on optimal treatment modality, treatment approval turnaround time, and potential toxicities. We herein present a single institution experience assessing the relationship between spatial distribution of tumor location and dosimetric outcomes in patients receiving definitive IMPT/PBS for locally advanced lung cancer. Consecutive patients treated with IMPT/PBS for locally advanced lung cancer from 2016-2022 at an academic proton center were retrospectively analyzed. Treatment, tumor volume, and patient characteristics were collected. Distance from the posterior pleura to the anterior-most portion of the clinical tumor volume (CTV) of the tumor was measured. A ratio of this distance to the pleura-to-pleura distance at that corresponding lung level was used to classify anterior (>0.7) versus posterior tumors. Dosimetric details including beam angles, dose-volume parameters for the ipsilateral, contralateral, total lung and heart for each patient were recorded. Statistical differences between groups were determined using Wilcoxon Signed Rank test. Pearson's correlation coefficient was used to determine impact of CTV volume and the anterior-most distance of CTV on dosimetric parameters. Fifty-eight patients treated for Stage IIB-IIIC locally advanced lung cancer with a median CTV volume of 258 cc's were identified and included in this retrospective study. Median age was 68 years and there was an equal distribution of male (29, 50%) and female patients (29, 50%) as well as anterior (>0.7) and posterior lying tumors in this cohort. The majority of patients were white (42, 72%) Forty patients (69%) had right sided lung tumors. Median cumulative dose was 61.2 Gy (range 59.4-70 Gy) while median dose per fraction was 2 Gy (range 1.8-2 Gy). When compared to anterior tumors, posterior tumors had a lower Lung V20 (19.1% vs 21.6%, p = 0.1,), V5 (27.6% vs 32.4%, p = 0.03), Mean Lung (9.9 Gy vs 11.7 Gy, p = 0.04), Mean Heart (4.8 Gy vs 6.3 Gy, p = 0.1), and Heart V50 (3.1% vs. 4.6%, p = 0.15). Anterior-most distance of CTV from the posterior pleura was positively correlated with Lung V20 (r = .435, p = 0.001), Lung V5 (r = .390, p = 0.002), Mean Lung (r = .479, p = 0.001), Mean Heart (r = .304, p = 0.020), and Heart V50 (r = .322, p = 0.014). CTV volume was also significantly positively correlated (p <.01) with these parameters. Spatial distribution and treatment volume size are predictive of dosimetric outcomes for locally advanced lung cancer. Posterior lying tumors may derive greater dosimetric advantage when compared to anterior lesions when utilizing IMPT/PBS.

Pencil Beam Scanning Proton Stereotactic Body Radiation Therapy (SBRT): A Robust Single Institution Experience

To describe the feasibility of treating a complex and diverse group of patients using pencil beam scanning (PBS) proton stereotactic body radiation therapy (SBRT: 5 or fewer fractions, with a fraction size of at least 5 Gy). Our center treats on average 105-120 PBS proton treatments daily, of which 9.5% of treatment courses are proton SBRT. Statistics of disease sites, treatment planning parameters (target volume, prescriptions, number of fields, SFO vs. MFO), and treatment efficiencies (scheduled time slots, actual treatment time) are presented for 305 consecutive SBRT patients receiving 1507 fractions in the past three years. Thermoplastic masks or Vacuum-lock bags are used to immobilize SBRT patients and index the patients' treatment position. Imaging guidance of orthogonal kV images and volumetric cone-beam CT is routinely used for patient setup. SBRT patients are grouped based on the target locations: pelvis (31%), liver (17%), thoracic (13%), spine (8%), abdominal (8%), brain (7%), non-spine bone (7%), ocular (6%), and head and neck (2%). Only 112 patients (37%) were receiving their 1st RT course, whereas 113 (37%) had one prior in-field RT course, and 80 (26%) had multiple prior in-field RT courses. The median [IQR] target volume was 65.4 [29.3, 168] cc (range: 0.3-2475 cc). 72% of cases were planned with SFO and 28% with MFO. On average, 3.76 fields (range: 2 to 12) were planned for each treatment. 44% of the treatments were planned with three or fewer fields, and 10% received more than five fields, most of which involved repainting for moving targets. Over 97% of treatments were delivered in 5 fractions, with ∼3% delivered in 3 fractions. The median [IQR] prescription per treatment was 8 [7, 10] Gy (range: 5-18 Gy per treatment). 85% (84%) of the SBRT treatments were scheduled (delivered) in a 45-minute or shorter slot, and 6% (7%) of treatments were scheduled (delivered) in over a one-hour slot, most commonly for multiple isocenter treatments. 93% of treatments were delivered within 15 minutes of the planned treatment time or shorter. Deep-inspiration breath-hold (DIBH) was applied to 45% of liver SBRT cases, with the remaining 55% planned on 4D CT with (14%) or without (86%) abdominal compression. DIBH was applied in 13% of lung SBRT cases. The application of other motion mitigation approaches, such as volumetric repainting, was determined by the target motion amplitude and whether the patient could tolerate DIBH. In the most diverse and largest proton SBRT experience delivered in the world over the past 3 years, over 300 patients were treated, demonstrating the feasibility and efficiency of delivering proton SBRT in a very busy center. The planning and treatment parameter statistics reported serve as a helpful reference for the proton community.

Comparison between Pre-Pectoral and Post-Pectoral Implant Reconstruction and Different Radiation Modalities on Breast Reconstruction Complications

Single Stage direct to implant (DTI) is a rising reconstruction approach. It has been speculated that prepectoral implant placement reduces reconstruction complications as well as using newer postmastectomy radiation (PMRT) modalities as protons. Largest series in this topic included 350 DTI patients with less than 30 patients receiving PMRT. We reviewed the charts of 2,187 patients (pts) who underwent mastectomy and reconstruction at our institution between 2000 and 2020. Pts only receiving DTI with and without PMRT were included. PMRT was delivered either with 3D conformal photon +/- chest wall electron boost (CWB) or proton therapy mainly with pencil beam scanning. All pts received conventional fractionation (50-50.4 Gy in 25-28 fractions). Patients on active protons clinical trials were excluded. Primary endpoints were reconstruction complications defined as infection/necrosis (I/N) requiring debridement; capsular contracture (CC) requiring capsulotomy and overall reconstruction failure (ORF: removal of permanent implant for any complication with and without salvage reconstruction). Subgroup analysis for pts receiving PMRT was done to explore impact of proton and photons on complications. Logistic and cox regression were used. Eight hundred nine pts received DTI, with an overall median follow-up of 6.2 years. Among the entire cohort, 78/809 (9.7%) had prepectoral implants while the rest had subpectoral implants with and without alloderm support. Of the entire cohort 391/809 (48.5%) received PMRT, among those 43/391 (11%) received protons. The 5 years cumulative incidence of CC was 8.5% and 6.6% among prepectoral vs subpectoral implants (HR = 1.2, p = 0.7). The 5 years cumulative incidence of ORF was 32.2% and 28.2% among prepectoral vs subpectoral implants (HR 1.1, p = 0.6). For pts receiving PMRT, the 5 years cumulative incidence of CC were 26.0% and 8.6% among protons vs photons (HR 3.7, p < 0.0005). The 5 years cumulative incidence of ORF were 38.2% and 30.1% among protons vs photons (HR 1.4, p = 0.2). Multivariable logistic regression analysis accounting for BMI, smoking history, diabetes, and PMRT showed no significant difference between prepectoral implants versus subpectoral coverage for CC (OR = 0.7, p = 0.5), and for ORF (OR = 0.9, p = 0.8). Subgroup multivariable analysis for those receiving PMRT only showed protons compared to photons significantly increased the risk of CC (OR = 5.3, p < 0.0001) and ORF (OR = 2.1, p = 0.03). No significant difference in I/N was noted between photons and protons or between prepectoral and subpectoral pts. For breast cancer patients receiving single stage direct to implant reconstruction with and without PMRT, prepectoral implant placement did not reduce the risk of complications. Proton therapy compared to photons significantly increased the risk of capsular contracture requiring capsulotomies and significantly increased overall reconstruction failures with implants removal.

Dose Volume Thresholds Associated with Acute Skin Toxicities in Proton Beam Therapy

The depth-dose characteristics of proton beam therapy (PBT) mean that the skin-sparing effect is reduced with PBT, potentially leading to an increased incidence and severity of acute radiotherapy induced skin toxicities (RIST). Predictive factors of acute RIST in patients treated with PBT remain largely undefined. In this study, we retrospectively reviewed the acute RIST of patients treated with pencil beam scanning (PBS) PBT to identify dose-volume thresholds which are predictive of acute RIST. All patients treated with PBS-PBT at a single institution between December 2018-October 2022 were included in this study. Acute RIST were recorded as per RTOG grading scale and dichotomized to Grade (G) <2 vs ≥2. Anonymized demographics, clinical and dosimetric data were extracted from electronic patient records and a treatment planning system. Skin structure is defined as 5mm rind grown as an inner margin from the patient contour. The following skin dose-volume statistics were collected: Dmax (maximum dose to any pixel inside the skin contour) and dose to skin volumes in 5Gy increments (V5Gy, V10Gy etc.). Preliminary analyses of dosimetric data of patients with G0, G1 vs ≥G2 acute RIST are presented, with significance assessed at the 5% level using t-tests and univariate logistic regression models, and risk thresholds determined using receiver operating characteristic (ROC) curves. We report the data for 582 patients with extracted dosimetric data. The pediatric, teenage and young adult (TYA) and adult populations were 38%, 19% and 43% respectively. The three most common indications for PBT were head and neck cancers (HNC) (23%), sarcoma (21%), and chordoma (15%). Increasing age, HNC and sarcoma were associated with an increased risk of grade 2+ acute RIST. For patients who developed acute RIST of G2+, the median volume receiving 10Gy, 20Gy, 30Gy, 40Gy and 50Gy were significantly higher (P<0.0001) than patients with G0 and G1. The dose volume effect of acute RIST is greater at 30Gy and above. Similarly, median Dmax was significantly higher in the G2+ acute RIST group compared to G0 and G1 (P<0.0001) for all age groups. Using the ROC curve, we observed threshold volumes (in cm3) for V10Gy, V20Gy, V30Gy, V40Gy and V50Gy (Table 1). The volume of irradiated skin and Dmax are associated with the risk of developing acute RIST in patients treated with PBS PBT. Further work is being done to develop a model predictive of acute RIST in clinical setting.

A Potential Cause for Unexplained Radiation Toxicity in Proton Therapy

Rare but devastating neurological tissue injury by proton therapy is poorly understood. One consideration is that the transition from passively scattered to pencil beam scanning (PBS) proton radiotherapy over the past decade has affected dose rate. We analyzed clinical data acquired at two institutions on a correlation of voxel-wise dose rates and injury of brain tissue (institution 1) and the optic apparatus (institution 2). All radiation-induced toxicities were delineated (Brain/optic apparatus) on follow-up MRIs. A method to calculate dose rates for PBS plans has been implemented at two institutions independently. Institution 1 reports beam-by-beam time-averaged maximum dose rates per voxel (VMDR) using a 100 ms time interval. Data from four patients who incurred brain tissue injury was included. Dose rate volume histograms were prepared for the target, a 1 cm outer ring to the target, and the necrotic region itself. Prescription doses ranged from 30.6 to 60 Gy delivering either 1.8 or 2.0 Gy per fraction. Tissue injury was generally observed in the dose fall-off at the edge of the target. Additionally, we performed LET weighted dose calculation, robustness analysis, and Monte Carlo calculations. Institution 2 presents voxel-wise dose delivered above a selected dose rate threshold, considering values between 30 and 180 Gy/min. The analysis was performed for 2 cases with optical apparatus injuries and one matching non-toxicity case. Both injuries were observed in the chiasma. Patients received 50.4, 50.4 and 55.8 Gy (non-toxicity) in 1.8 Gy fractions. D1 for chiasma was 51.1, 47.2 and 54 Gy, respectively. Mean and (max) LET in chiasma were 4.02 (4.72), 2.49 (3.16) and 3.44 (4.56) keV/µm accordingly. Institution 1 finds distinct dose rate hot spots in the contoured injured brain regions. Mean VMDR values across all beams are lower in targets than in the 1 cm outer ring (80 Gy/min and 82.8 Gy/min, respectively) and elevated to 113 Gy/min for necrotic regions. Neither of the additional computations showed significant increase of dose in the affected areas. Institution 2 observes that chiasmas, which developed radiation injuries, were exposed to higher dose levels delivered at elevated dose rates (> 72 Gy/min) than the chiasma that did not show any radiation injury. D95 above 72 Gy/min for the two volumes identified as radiation injury was 0.25 and 0.12 Gy/fx, while D98 to chiasmas minus radiation injury or chiasma without any injury for the matching case was 0.06, 0.03 and 0 Gy/fx respectively. Using unique formulations to assess voxel-wise dose rates to brain and optic chiasm, two independent institutions observed that regions with elevated dose rates correlated with radiographic radiation injury. An extended retrospective study including a larger patient cohort with radiation-induced toxicity and asymptomatic patients is under way to confirm or refute the presented observations.

Photon vs. Proton Radiotherapy in the Definitive Treatment of Nasopharyngeal Cancer: Single Institution Experience

Definitive therapy for nasopharyngeal cancer includes chemotherapy and radiation (RT). Common toxicities such as xerostomia, mucositis, and hearing loss are correlated with the RT dose delivered to associated organs at risk. We hypothesized that compared to our historical experience with IMRT, the implementation of proton therapy (PT) would reduce radiation dose to organs at risks without compromising oncologic outcomes. A retrospective review of all non-metastatic stage II-IV nasopharyngeal carcinoma (SCC, lymphoepithelioma, undifferentiated carcinomas) treated with definitive therapy at our institution from 2012-2022. Disease parameters and the mean dose to organs at risk were evaluated. Statistical comparison was made with the chi square test for categorical and Wilcoxon rank sum test for continuous variables. The Kaplan-Meier method was used to estimate overall survival (OS) and progression-free survival (PFS), using a log-rank test to compare IMRT and PT. PFS was defined as the time from the start of treatment to the first of either disease progression, relapse or death from any cause. PT was delivered with pencil-beam scanning in all patients. IMRT included multi-field treatment and volumetric-modulated arc therapy. A total of 80 patients were included in analyses: 48 treated with IMRT and 32 with PT. Comparing IMRT to PT cohorts, there was no difference in the median age of patients (51 vs 55 years, p = 0.73), nor the distribution by T stage (p = 0.57) or N stage (p = 0.34) or in the percentage of patients with ECOG 2/3 performance status at presentation (p = 0.11). All but one patient received concurrent systemic therapy and there was no difference in the use of concurrent cisplatin between cohorts (83% vs 78%, p = 0.57). The most common non-cisplatin concurrent regimen was weekly carboplatin and paclitaxel. Induction chemotherapy was more commonly used in patients treated with PT (10.4% vs 25%, p = 0.04) while there was no difference in the use of adjuvant chemotherapy (10.4% vs 9.4%, p = 0.88). Among 42 cases initiating treatment since the opening of our proton center, 32 (76%) have received PT. Comparing IMRT and PT dosimetry, patients treated with PT received significantly lower mean dose to the better spared parotid gland (32.8 vs 25.7 Gy, p = 0.001), lesser spared parotid gland (35.5 vs 31.1 Gy, p = 0.047), better spared cochlea (31.5 vs 25.5 Gy, p = 0.004), lesser spared cochlea (41.8 vs 33.2 Gy, p = 0.004), larynx (44.5 vs 21.7 Gy, p<0.001), and oral cavity (42.6 vs 17.0 Gy, p<0.001). After a median follow-up time of 30 months (45 mos IMRT, 23 mos PT) the estimated 2-year PFS was 63.9% with IMRT and 90.3% with PT (p = 0.047). The estimate of 2-year overall survival was 86.8% with IMRT and 96.8% with PT (p = 0.17). Comparing patients by radiation treatment modality, PT was associated with a statistically significant reduction in mean radiation dose to the parotid glands, cochlea, larynx, and oral cavity with excellent initial oncologic outcomes.

Proton Single-Energy Bragg-Peak FLASH Using Clinical Systems Can Achieve IMPT-Equivalent Plan Quality for Breast and Prostate Cancers

Most current proton FLASH-RT studies focus on transmission proton techniques. In this study, we propose a novel method for achieving FLASH dose rate in hypofractionated proton radiotherapy using the Bragg peak of a single-energy proton beam. The dosimetric characteristics using this novel technique for proton pencil beam scanning (PBS) stereotactic body radiation therapy (SBRT) of prostate and breast cancers were first investigated based on the clinically available cyclotron beam parameters. This novel approach uses the distal tracking technique that enables PBS Bragg-peak of the highest proton energy to adapt to the target distally. Positioning of the Bragg peak at different depths is achieved using a universal range shifter and range compensator. To investigate the feasibility of this approach, we developed an in-house treatment planning platform for intensity-modulated proton therapy (IMPT) delivery and performed dosimetric studies on prostate and breast SBRT cases previously treated with conventional proton PBS technique. FLASH plans were generated using a similar clinical beam arrangement to deliver 40 Gy (RBE) in 5 fractions. Dose metrics were compared between the clinical and FLASH plans. Dose-rate volume histograms (DRVH) were also calculated to investigate the 40 Gy/s coverage (V40 Gy/s) of organs-at-risk (OARs) for FLASH plans. The distal tracking can precisely stop the Bragg peak at the target distal edge, and Bragg peak plans achieved tumor coverage and dose conformality equivalent to IMPT plans. The clinical IMPT plans yielded slightly superior target dose uniformity -CTV Dmax of FLASH plans was 10% higher for prostate and 2% higher for breast. There was no significant difference between the clinical and FLASH plans in dose metrics for major OARs, including rectum, large bowel, heart, and lung. Higher maximal doses to femoral heads (∼2 Gy) and urethra (∼6 Gy) were observed in prostate FLASH plans than in the clinical plans but were still within clinically accepted dose limits. The V40 Gy/s for OARs were >90% for prostate FLASH plans and >76.5% for breast FLASH plans. The proposed single-energy Bragg-peak FLASH technique eliminates exit dose associated with transmission proton FLASH and can still yield comparable plan quality and OAR sparing while preserve sufficient FLASH dose rate coverage for prostate and breast proton SBRT. This study demonstrates the potential application of Bragg peaks for highly conformal FLASH-RT using clinical cyclotron systems to treat prostate and breast cancer patients, which moves towards clinical application.

The Interplay Effect and Mitigations with Cyclotron and Linac Proton Beam Scanning for Lung SBRT

To evaluate the impact of different types of pencil beam scanning proton accelerators and spot sizes on interplay effects, mitigations, and plan quality for lung cancer patients treated with SBRT. Twenty lung cancer patients (ten peripheral and ten central tumors) treated in our institution with photon SBRT were selected to represent varying tumor volumes and respiratory motion amplitudes for this retrospective study. The respiratory motion amplitude ranged from 0.1 to 1.0 cm with compression. For each patient, plans were created using: 1) cyclotron-generated proton beams (CPB) (σ: 2.7-7.0 mm); 2) linear accelerator proton beams (LPB) (σ: 2.9-5.5 mm); and 3) linear accelerator proton minibeams (LPMB) (σ: 0.9-3.9 mm). Plans were robustly optimized on the GTV using each individual 4DCT phase. Single-filed optimization (SFO) plans were the first attempt, and if the plan quality did not meet the dosimetric requirement, multi-field optimization (MFO) was used. MFO plans were created for all patients for comparison. For each patient, all plans were normalized to have the same dose to 99% of the GTV. Interplay effects were evaluated for ten scenarios of treatment delivery starting in ten breathing phases using machine generic time models and a constant breathing period of 4 seconds. Volumetric repainting (VR) was performed 2-6 times for each plan. To assess plan quality in the nominal scenario, we compared the conformity index (CI), R50, and the percentage of lung volume receiving 20 Gy (RBE) (V20Gy). CI is defined as the ratio of the 100% isodose volume to the GTV. R50 is defined as the 50% isodose volume divided by the GTV. Dmax and V18Gy of the proximal bronchial tree (PBT) were evaluated for central tumors. Twelve of 20 plans can be optimized sufficiently with SFO. In interplay effect evaluation, the mean V100%RX values of the GTV were 99.42±0.6%, 97.52±3.9%, and 94.49±7.3%for CPB, LPB, and LPMB plans respectively. After VR 2/3/5 times, the V100%RX values were improved (on average) by 0.13%/1.84%/4.63% for CPB/LPB/LPMB plans. The delivery time for VR plans was the lowest for LPB plans, while delivery time for LPMB was on average 1 minute longer than CPB plans. VR showed no effect on lung V20Gy, Dmax and V18Gy of the PBT. SFO plans were more robust against the interplay effect compared with MFO plans for LPB and LPMB. Average CIs of 1.88±0.4, 1.79±0.4, and 1.75±0.4; average R50s of 7.99±4.0, 6.68±3.0, and 5.70±2.6; and average lung V20Gy values of 2.81±1.5, 2.26±1.3, and 1.85±1.1 were obtained for CPB, LPB, and LPMB plans, respectively. Dmax and V18Gy of the PBT decreased with decreasing spot sizes. LPMB, with the smallest spot size, produced superior plan quality. In the absence of VR, proton machines with large spot sizes generated more robust plans against interplay effects. VR improved the plan robustness against interplay effects for modalities with small spot sizes and fast energy changes, preserving the low dose sparing aspect of the LPMB, even when motion is included.

Increased Complication Rates with Proton Therapy in Breast Cancer Patients with Immediate, Implant-Based Reconstruction: Single-Institution Comparative Effectiveness Analysis

To compare the impact of protons vs. photons on breast reconstruction complications, for patients (pts) receiving postmastectomy radiation (PMRT) with either single-stage direct-to-implant (DTI) or two-staged expander/implant (TE/I). We reviewed the charts of 578 pts who underwent immediate reconstruction followed by radiation at our institution between 2010 and 2020. Pts with implant-based reconstruction using either TE/I or DTI and PMRT delivery in the presence of the prosthesis were included. Pts enrolled in active ongoing clinical trials were excluded from the analysis. The photon group received 3D conformal or IMRT/VMAT treatment with a median dose of 50-50.4 Gy in 25 to 28 fractions. For proton pts, treatment was delivered mainly with pencil beam scanning technique (PBS); few pts received passively scattered proton spread-out Bragg peak (SOBP). The complications were defined as infection/skin necrosis (I/N) requiring operative debridement, capsular contracture (CC) necessitating capsulotomy, and overall implant failure (ORF) as the removal of the permanent implant irrespective of replacement outcomes (i.e., with and without salvage reconstruction). We fit inverse-probability weighted cumulative incidence curves to adjust for confounding and non-random loss to follow-up. Various sensitivity analyses were conducted. Four hundred ninety-five pts were available for the final analysis with an overall median follow-up of 55 months. 66 (13%) received protons, of which14 were treated with SOBP protons. 137 (28%) and 256 (56%) received photons with and without chest wall boost (CWB), respectively. The 5-year inverse probability-weighted risk of CC post-PMRT was 31% for protons vs. 10% for photons (RR:3.09, 95% CI: 1.77, 5.40). The 5 years ORF risk was 35.6% in protons compared to 22.7% in photons pts (RR: 1.57; 95% CI 1.0, 2.48). Hazard ratios from the adjusted Cox models were 3.79 (p<0.001) for CC and 2.05 (p<0.01) for ORF. No difference in I/N was noted between protons and photons pts. Sensitivity analysis showed that protons significantly increased CC risk vs photons both with CWB (HR:3.56, P<0.001) and without CWB (HR:3.9, p<0.001). Similar outcomes were observed with ORF, where protons increased the rate of ORF compared to photons, irrespective of CWB (HR 1.8, p = 0.038 with CWB; HR 2.4, p = 0.004 without CWB). No differences between PBS and SOBP proton techniques were noted. Compared to photons, proton therapy increases the risk of capsular contracture requiring surgical intervention and hence overall reconstruction failure. This data should inform discussions about the risks and benefits of protons in patients with reconstruction, while awaiting mature data from ongoing clinical trials (RADCOMP) utilizing protons for breast cancer.