Chang, John

Abstract Introduction: Radiotherapy (RT)-induced acute esophagitis (AE) is a common side effect in lung cancer patients receiving RT, which can significantly impact their quality of life. This highlights the need for a predictive model that can estimate AE risk in advance using pretreatment imaging data. However, collecting sufficient data for model development is often resource-intensive and costly. Additionally, acquiring a homogeneous training dataset (e.g., from a single center or modality) is not always feasible. This study aimed to develop an automated artificial intelligence-based 3D vision foundation model (VFM) combining standard-of-care planning computed tomography (pCT) and planned radiation dose maps to predict grade II or higher AE. Materials and methods: This study included 246 patients with non-small cell lung cancer who underwent either image-guided radiation treatment (IMRT) or proton-beam radiation therapy. The endpoint was grade two or higher AE (33% positive and 67% negative). The IMRT group consisted of 182 patients from a single center, whereas the proton therapy group included 64 patients from the Proton Collaborative Group trial from 11 different institutions. For each patient, pCT, dose maps, and radiotherapy segmentations were available. The VFM was created using a transformer pretrained using a large number of unlabeled volumetric 3D CTs from patients with varied diseases through a self-supervised learning approach, which extracts useful features directly from images. Next, the pretrained encoder was combined with fully connected classification layers and fine-tuned using stratified 5-fold cross-validation with 20 patients set aside for testing. We developed two models: a CT only and CT + dose model. Model performance was assessed using the area under the curve (AUC), specificity, and sensitivity. Results: In cross-validation, adding dose information to CT (CT+Dose) improved model performance, increasing AUC from 0.72 ± 0.08 to 0.76 ± 0.10 and specificity from 0.82 ± 0.09 to 0.87 ± 0.04, while sensitivity remained the same. On the independent test set, CT+Dose showed a marked improvement over CT only (AUC: 0.82 vs. 0.59; sensitivity: 0.57 vs. 0.14), with a slight decrease in specificity (0.77 vs. 1.00). Conclusion: A VFM model combining CT and radiation dose showed the capability to predict AE more accurately with higher specificity. Further studies on larger cohorts of testing patients are planned to assess model generalization. Citation Format: Chloe Min Seo Choi, Jue Jiang, Nikhil Mankuzhy, John Chang, Jing Zeng, Carlos Vargas, James Urbanic, Isabella Choi, Mark Mcdonald, James Gray, Joseph Deasy, Maria Thor, Charles Simone, Harini Veeraraghavan. Prediction of radiotherapy-Induced esophagitis in non-small cell lung cancer using a 3D vision foundation model [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Artificial Intelligence and Machine Learning; 2025 Jul 10-12; Montreal, QC, Canada. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(13_Suppl):Abstract nr B022.

Abstract Purpose: Embryonal tumor with multilayered rosettes (ETMR) is a rare pediatric CNS embryonal tumor with poor survival. The Pediatric Proton/Photon Consortium Registry (PPCR) was queried for outcomes data from prospectively consenting pediatric patients with ETMR treated with proton radiation therapy (RT). Methods: 20 patients (2013-2021) at 9 institutions had ETMR; 2 with prior RT were excluded from statistical analyses (PPCR ETMR, N=18). Overall Survival (OS) and Event Free Survival (EFS) analyses were performed using the Kaplan-Meier method and log-rank values. Median follow-up was calculated using the reverse Kaplan-Meier method. Results: Median age at RT was 3.0 years (1.7-12.2); median follow-up was 55.5 months (2.6-119.4). 8 patients (44%) expired and 6 patients (33%) are surviving ≥55 months. 11 (61%) patients received systemic therapy with stem cell support. The majority (89%) had focal RT, while 2 patients received craniospinal irradiation (CSI). 4-year OS and EFS were 59.6% and 54.2%, respectively. Local control (LC) at 4 years was 81%. No differences in OS or EFS were observed for receipt of systemic therapy with stem cell support (p=0.361, p=0.57), progression prior to RT (p=0.127, p=0.18), or surgery to RT ≥200 days (p=0.35, p=0.254). Symptomatic radionecrosis was not reported. Conclusion: Focal proton RT provided effective local control as part of multimodality therapy for ETMR, with encouraging survival for this rare and often infant age tumor. Outcomes for CSI were limited to 2 patients treated upfront, and 1 patient receiving salvage CSI for disseminated relapse after focal RT who is surviving >1 year. Trial: DFCI protocol 12-103, clinicaltrials.gov NCT01696721, date of registration 9/27/2012

Abstract BACKGROUND Standard of care treatment for recurrent malignant glioma after prior radiotherapy (RT) is not well established. Proton therapy (PT) is increasingly used for reirradiation (ReRT) at the time of recurrence; however, treatment outcomes, toxicity, and prognostic factors for patients treated with PT-ReRT remain poorly defined. METHODS The prospective, multi-institutional Proton Collaborative Group (PCG) registry was queried for patients with malignant glioma who previously received RT and underwent PT-ReRT between 7/2011-12/2023. Overall survival (OS) and progression-free survival (PFS) were assessed using the Kaplan-Meier method, and Cox proportional hazards regression was used for uni- and multivariate analysis (UVA and MVA). Dose was calculated using equivalent dose in 2-Gy fractions (EQD2). RESULTS 143 consecutive patients were identified. Median follow-up was 11.2 months and median time interval (TI) from prior RT (median 58.5 Gy) to PT-ReRT (median 44.6 Gy) was 42.4 months. Primary histology was glioblastoma (n=74), astrocytoma (n=35), and oligodendroglioma (n=34). Median PFS and OS were 8.1 and 11.2 months, respectively. On UVA, improved OS was associated with oligodendroglioma (HR 0.37) and astrocytoma histology (HR 0.53) compared to glioblastoma, TI >60 months (HR 0.2), CNS WHO Grade 2 compared to 4 (HR 0.42), and ECOG performance status 0 (HR 0.47). On MVA, improved OS remained associated only with oligodendroglioma (HR 0.42) and TI >60 months (HR 0.49). Acute and Late Grade 3 toxicity occurred in 7% and 4% of patients. Acute Grade 3 toxicity was associated with poor performance status (HR 16.15). CONCLUSIONS In the largest series of glioma PT-ReRT reported to date, retreatment appears well tolerated with variable outcomes based on clinical prognostic factors. Toxicity rates were acceptable compared to historical photon-based literature despite a high median prescription dose.

Abstract Purpose: The optimal salvage treatment for malignant glioma after prior radiotherapy (RT) is not well established. Proton therapy (PT) is an attractive treatment modality for reirradiation (ReRT) of recurrent disease; however, data are limited.Methods: The prospective, multi-institutional Proton Collaborative Group (PCG) registry was queried for patients with malignant glioma who previously received RT and underwent PT-ReRT between 7/2011-10/2018. Overall survival (OS) and progression-free survival (PFS) were assessed using the Kaplan-Meier method and Cox proportional hazards regression analysis. Results: 109 consecutive patients were identified. Median follow-up was 11.2 months and median time interval (TI) from prior RT (median 59.3 Gy) to PT-ReRT (median 49.6 Gy) was 45.7 months. Primary histology was glioblastoma (n=62), oligodendroglioma (n=26), and astrocytoma (n=20). Median PFS and OS was 6.1 and 11.1 months, respectively. On UVA, improved PFS was associated with oligodendroglioma (HR 0.39) and astrocytoma (HR 0.55) histologies, WHO Grade III (HR 0.36), TI (HR 0.989), surgical resection (HR 0.39), and PT-ReRT dose ≥50 Gy (HR 0.59) (all p<0.03), while improved OS was associated with oligodendroglioma histology (HR 0.35), WHO grade III (HR 0.33), ECOG 0 (HR 0.47), TI (HR 0.985), and PT-ReRT dose ≥50 Gy (HR 0.58) (all p<0.04). Late Grade 3 toxicities occurred in 5.6% of patients with no attributable acute or late Grade 4-5 events.Conclusion: In the largest series of glioma PT-ReRT reported to date, retreatment appears well tolerated with acceptable outcomes and favorable toxicities compared to photon historical controls. Dose escalation achievable with PT may improve outcomes for well-selected patients.

Purpose: As patients with mediastinal lymphoma are typically young with curable disease, advanced radiation techniques such as proton therapy are often considered to minimize subacute and late toxicity. However, it is unclear which mediastinal lymphoma patients are treated with proton therapy. Within a prospective, multi-institutional proton registry, we characterized mediastinal lymphoma patients treated with proton therapy and assessed concordance with consensus recommendations published in 2018 by the International Lymphoma Radiation Oncology Group (ILROG). Methods: Eligible patients included those with lymphoma of the mediastinum treated exclusively with proton therapy for whom digital imaging and communications in medicine (DICOM) treatment data were available for review. Given the challenge with reliably visualizing the left mainstem coronary artery, the inferior-most aspect of the left pulmonary artery (PA) was used as a surrogate. Extent of disease was characterized as upper mediastinum (above level of left PA), middle mediastinum (below left PA but at or above level of T8), or low mediastinum (below T8). Results: Between November 2012 and April 2019, 56 patients were treated and met inclusion criteria. Patients treated with proton therapy were young (median, 24 y; range: 12 to 88), with over half being female (55%). Patients were most commonly treated at initial diagnosis (86%) and had Hodgkin lymphoma (79%). Most patients (96%) had mediastinal disease that extended down to the level of the heart: 48% had middle and 48% had low mediastinal involvement. Nearly all patients (96%) met the ILROG consensus recommendations: 95% had lower mediastinal disease, 46% were young females, and 9% were heavily pretreated. Heart (mean) and lung dose (mean, V5, V20) were significantly associated with lowest extent of mediastinal disease. Conclusions: Mediastinal lymphoma patients treated with proton therapy are typically young with lower mediastinal involvement. Within a prospective, multi-institutional proton registry, nearly all treated patients fit the ILROG consensus recommendations regarding which mediastinal lymphoma patients may most benefit from proton therapy.

Objective: The Pediatric Proton/Photon Consortium Registry (PPCR) is a comprehensive data registry composed of pediatric patients treated with radiation. It was established to expedite outcomes-based research. The attributes which allow the PPCR to be a successful collaboration are reviewed. Methods and materials: Current eligibility criteria are radiotherapy patients &lt; 22 years treated at one of the 15 US participating institutions. Detailed health and treatment data are collected about the disease presentation and treatment exposures, and annually thereafter, in REDCap (Research Electronic Data Capture). DICOM (Digital Imaging and Communications in Medicine) imaging and radiation plans are collected through MIM/MIMcloud. An optional patient-reported quality-of-life (PedsQL) study is administered at 10 sites. Results: Accrual started October 2012 with 2,775 participants enrolled as of 25 July 2019. Most patients, 62.0%, were treated for central nervous system (CNS) tumors, the most common of which are medulloblastoma (n = 349), ependymoma (n = 309), and glial/astrocytoma tumors (n = 279). The most common non-CNS diagnoses are rhabdomyosarcoma (n = 284), Ewing’s sarcoma (n = 153), and neuroblastoma (n = 130). While the majority of participants are US residents, 18.7% come from 36 other countries. Over 685 patients participate in the PedsQL study. Conclusions: The PPCR is a valuable research platform capable of answering countless research questions that will ultimately improve patient care. Centers outside of the USA are invited to participate directly or may engage with the PPCR to align data collection strategies to facilitate large-scale international research. Advances in knowledge: For investigators looking to carry out research in a large pediatric oncology cohort or interested in registry work, this paper provides an updated overview of the PPCR.