Proton therapy offers superior organ-at-risk sparing but is highly sensitive to anatomical changes, making accurate deformable image registration (DIR) across longitudinal CT scans essential. Conventional DIR methods are often too slow for emerging online adaptive workflows, while existing deep learning-based approaches are primarily designed for generic benchmarks and underutilize clinically relevant information beyond images. To address this gap, we propose a clinically scalable coarse-to-fine deformable registration framework that integrates multimodal information from the proton radiotherapy workflow to accommodate diverse clinical scenarios. The model employs dual CNN-based encoders for hierarchical feature extraction and a transformer-based decoder to progressively refine deformation fields. Beyond CT intensities, clinically critical priors, including target and organ-at-risk contours, dose distributions, and treatment planning text, are incorporated through anatomy- and risk-guided attention, text-conditioned feature modulation, and foreground-aware optimization, enabling anatomically focused and clinically informed deformation estimation. We evaluate the proposed framework on a large-scale proton therapy DIR dataset comprising 1,222 paired planning and repeat CT scans across multiple anatomical regions and disease types. Extensive experiments demonstrate consistent improvements over state-of-the-art methods, enabling fast and robust clinically meaningful registration.

Purpose:Intensity-modulated proton therapy (IMPT) enables highly conformal dose delivery for head and neck (H&N) cancers but is sensitive to inter-fractional anatomical changes, motivating online adaptive radiotherapy (oART). A major bottleneck in oART is fast and accurate Monte Carlo (MC) dose calculation for plan evaluation and re-optimization. While low-statistics MC simulations accelerate computation, they introduce substantial noise. This study investigates dose denoising to rapidly recover high-accuracy dose distributions from noisy MC simulations.Methods:A diffusion transformer(DiT)–based dose denoising framework was developed. IMPT plans and 3D CT images from 80H&N patients were used for training by generating paired low- and high-statistics MC dose maps. Data were divided into uniform chunks, normalized, and nonlinearly transformed to achieve a quasi-Gaussian voxel-wise dose distribution across the dataset, enabling stable diffusion training. An independent cohort of 40patients across H&N, lung, breast, and prostate(10per site) was used for testing with identical preprocessing. The model was trained using noisy dose maps and CT images as input and high-statistics dose maps as ground truth, with combined MSE, residual, regional MAE, and DVH-based loss functions. Performance was evaluated using MAE, 3D gamma passing rates, and clinically relevant DVH indices.Results:MAE was below 0.2Gy[RBE] for all test cases. Whole-body 3D gamma passing rates exceeded95% using a3%/2 mm criterion across all disease sites. DVH metrics showed strong agreement between predicted and ground-truth doses for clinical target volumes and organs at risk.Conclusion:A DiT–based framework was developed that accurately denoises low-statistics MC dose maps. Despite training exclusively on H&N, the model generalizes robustly across multiple disease sites.

8045 Background: Surgical resection is the primary therapy for stage I-II NSCLC but not performed in a subset of patients due to comorbidities or self-election, who instead undergo radiation with or without systemic therapy (RT±ST). Little is known about the demographic, tumoral, therapeutic, and quality-of-life (QOL) characteristics of this non-surgical group. Understanding their differences from the surgical group is essential for developing precision treatment strategies and tailored survivorship guidelines. Methods: In a prospectively enrolled (1997-2016) and followed (through 2025) cohort of 20,951 primary lung cancer patients, 6,486 (31%) had stage I-II NSCLC, including 5,359 (83%) treated surgically (Surgical) and 1,113 (17%) otherwise (Non-surgical). Preexisting diseases were grouped into cancer, lung, and other comorbidities. QOL data were available for 40% of all patients. Descriptive analysis of demographics, tumor features, treatment patterns, and QOL metrics, as well as survival analysis (median years and overall survival [OS] rates) were conducted. QOL measures, on overall and 15 symptom and functional domains, were scored 0 (worst) to 10 (best), with 1-unit difference considered clinically and statistically meaningful. Results: At time of diagnosis, compared to the Surgical, the Non-surgical patients were older (median 73 v 68 years), more likely to have smoked cigarettes (91% v 83%), squamous cell and unspecified NSCLC (53% v 28%), stage IIB (19% v 12%), and preexisting lung comorbidities (71% v 58%). Non-surgical patients more frequently received RT±ST than the Surgical (70% v 21%) and had shorter median survival (2.5 v 7.8) years. Within the Non-surgical, 28% had no cancer treatment (No-Tx), 44% RT-only, 7% ST-only, and 19% both (RT+ST); their respective median diagnosis ages were 74, 76, 68, 70 years; median survival were 1.7, 2.7, 2.2, 3.2 years; 5-year OS (95%CI) at 25% (21-31), 25% (21-29), 26% (17-38), 30% (25-37); and 10-year OS at 9% (7-14), 7% (5-10), 10% (5-21), 9% (6-14), respectively. RT-only group had highest rates of comorbidities (37-84%) and the No-Tx the lowest (31-66%). Although the overall QOL scored the same for all groups, RT-only patients reported the worst scores on multiple symptoms burden, and the No-Tx reported the least pain and lung cancer symptoms but the worst on mental well-being and social activities. Conclusions: Non-surgically managed stage I-II NSCLC cases present clinically distinct patients with the shortest median survival in No-Tx group and highest 5-year survival in RT+ST. RT-only patients were oldest, had the most comorbidities, and reported the worst symptom burdens. These differences highlight the need for more in-depth analyses and studies (considering age, histology, comorbidity, QOL, etc.) to support tailored therapeutic strategies and survivorship care for patients who decline or are not suitable for surgery.

Purpose:Intensity-modulated proton therapy (IMPT) is an advanced treatment modality for head and neck (H&N) cancer patients, offering precise tumor dose coverage while sparing surrounding organs at risk (OARs). However, IMPT is highly sensitive to inter-fractional anatomical changes, necessitating periodic adjustments through online adaptive radiation therapy (oART). But a significant bottleneck in the current oART workflow is the need for fast and accurate dose calculation using Monte Carlo (MC) simulations for plan quality assessment and re-optimization. Reducing the number of particles in MC-based simulations can accelerate dose calculation but at the cost of reduced accuracy. To address this, denoising noisy dose maps generated by low statistics MC simulations has been proposed as a method to rapidly and accurately generate high-accuracy dose maps. Methods:A diffusion transformer-based dose denoising framework was developed. IMPT treatment plans and 3D CT images from 80 H&N cancer patients were used to construct the training dataset by generating noisy dose maps and their corresponding high statistics dose maps using an open-source MC software, MCsquare, with a computation time of approximately 1 minutes and 10 minutes per plan, respectively. Each data sample was standardized into uniform chunks with zero-padding. Then, normalization and non-linear mapping were applied to the data samples to transform them toward a quasi-Gaussian distribution. The treatment plans and 3D CT images from another independent 10 H&N cancer patients, 10 prostate cancer patients, 10 lung cancer patients, and 10 breast cancer patients were used as the testing dataset, following the same preprocessing protocol as the training dataset. The proposed model was trained with noisy dose maps and 3D CT images as input and high statistics dose maps as the ground truth. The training was constrained by mean square error (MSE) loss, a residual loss that focused on reducing the difference between the predicted and ground truth dose maps and a regional mean absolute error (MAE) loss that specifically targeted voxels with the top 10% and bottom 10% dose values. Performance was evaluated using MAE. 3D Gamma passing rates and dose volume histogram (DVH) indices were calculated to assess differences between the predicted and ground-truth dose maps. Results:The proposed framework achieved MAE of 0.195 ± 0.112 Gy[RBE], 0.120 ± .054 Gy[RBE], 0 . 172 ± .096 Gy[RBE], and 0.376 ± 0.375 Gy[RBE] for H&N, lung, breast and prostate testing cases, respectively. The 3D gamma passing rate consistently exceeded 92% in the whole body using a 3%/2 mm criterion across all disease sites. DVH indices calculated from the ground truth and predicted dose distributions showed excellent agreement for both clinical target volumes (CTVs) and OARs. Conclusion:A diffusion transformer-based denoising framework was successfully developed. Although the denoising model was trained using only H&N data, it can accurately and robustly denoise noisy dose maps across different disease sites.

Purpose: To develop a retrieval-augmented generation (RAG) system powered by LLaMA-4 109B for automated, protocol-aware, and interpretable evaluation of radiotherapy treatment plans. Methods and Materials: We curated a multi-protocol dataset of 614 radiotherapy plans across four disease sites and constructed a knowledge base containing normalized dose metrics and protocol-defined constraints. The RAG system integrates three core modules: a retrieval engine optimized across five SentenceTransformer backbones, a percentile prediction component based on cohort similarity, and a clinical constraint checker. These tools are directed by a large language model (LLM) using a multi-step prompt-driven reasoning pipeline to produce concise, grounded evaluations. Results: Retrieval hyperparameters were optimized using Gaussian Process on a scalarized loss function combining root mean squared error (RMSE), mean absolute error (MAE), and clinically motivated accuracy thresholds. The best configuration, based on all-MiniLM-L6-v2, achieved perfect nearest-neighbor accuracy within a 5-percentile-point margin and a sub-2pt MAE. When tested end-to-end, the RAG system achieved 100 Conclusion: Our findings highlight the feasibility of combining structured population-based scoring with modular tool-augmented reasoning for transparent, scalable plan evaluation in radiation therapy. The system offers traceable outputs, minimizes hallucination, and demonstrates robustness across protocols. Future directions include clinician-led validation, and improved domain-adapted retrieval models to enhance real-world integration.

Radiation oncology, like many specialties in medicine, would benefit from a learning environment cultural shift to one that is more supportive to the learner and their career goals. In order to accomplish this, a strategic plan to better develop mentorship skills in faculty is needed. Mentorship and teaching skill development should be built in to planned professional career development.

Purpose/Objective(s) NRG 0617 established that the standard dose of photon RT for unresectable NSCLC is 60 Gy in 30 fractions. However, no dose-escalating trials have been reported for modern active-scanning PBT. A signal-seeking Phase II randomized study (MC1623) was designed and conducted to address this. Materials/Methods A single-institutional study was performed. Inclusion criteria included patient (pt)’s age ≥ 18 years; tissue confirmation of NSCLC; PFT’s FEV1 ≥ 1 L; unresectable or medically inoperable stage II-III; ECOG PS 0-1; acceptable labs for concurrent chemo; and curative intent. Exclusion criteria included weight loss ≥ 10%; M1 cancer; uncontrolled illnesses; active second malignancy; and prior RT that would overlap with planned PBT. A brain MRI was required. The original design was to randomize pts into 60 Gy/30 fractions (fx) vs. 66 Gy/33 fx vs. 72 Gy/36 fx arms in a 1:1:1 fashion. Contouring included generations of GTV, iGTV, and CTV’s. SFO/MFO optimization methods were used for PBT. The primary endpoint was progression-free survival (PFS) improvement, with secondary endpoints being overall survival (OS), adverse events, locoregional and distant failure rates. Results Accrual was slower than expected, and the 66-Gy arm was first closed, followed by the entire study. Between Aug 2017 and June 2021, 20 pts met inclusion criteria (originally planned for 48 pts). 1 pt was denied by insurance for PBT; 2 (11%) withdrew their consent after randomization. The final analysis consisted of 17 pts. 2 (12%) pts were randomized to the 66-Gy arm, and they were included with the other 8 pts (60 Gy) as standard-dose arm. 7 (41%) pts received 72 Gy (high-dose). The Mayo Prognostic scores for comorbidities were balanced (P = 0.89). The mean age was 76.4 (standard) vs. 74.2 years (P = 0.50); 8 (47%) were male. 13 (77%) smoked in the past. 8 (47%) pts had squamous cell carcinoma; 8 (50%) pts had T3/4 tumors, and 14 (82%) with N2 nodes. At the end of follow-up (median 1.86 years), 6 pts were alive, and 11 deaths had occurred. There was no difference in progression-free (2.04 vs. 0.94 years, P = 0.71) nor overall (3.05 vs. 2.36 years, P = 0.74) survivals, for standard vs high dose arms, respectively. Female gender appeared to favor PFS (P = 0.03), and no nodal involvement for improved OS (P = 0.08). There was 1 pt with grade 3 pneumonitis in standard-dose arm, and 1 pt each with grade 3 myocardial infraction and pulmonary fibrosis in high-dose arm. A competing risk analysis is being planned. Conclusion While the study was too small to statistically determine a benefit of one dose arm vs. the other, numerically, there appeared no clear advantage to 72-Gy dose escalation in terms of PFS and OS which could be verified in a larger randomized trial. NRG 1308 should clarify if PBT is superior to photon-based RT. With standard use of immunochemotherapy in addition to CRT for locally advanced NSCLC, the increased chance of cardiopulmonary toxicities may be lessened with better tissue-sparing PBT. This, too, will require further prospective studies.

Importance:The optimal radiotherapy technique for unresectable locally advanced non-small cell lung cancer (NSCLC) is controversial, so evaluating long-term prospective outcomes of intensity-modulated radiotherapy (IMRT) is important. Objective:To compare long-term prospective outcomes of patients receiving IMRT and 3-dimensional conformal radiotherapy (3D-CRT) with concurrent carboplatin/paclitaxel for locally advanced NSCLC. Design, Setting, and Participants:A secondary analysis of a prospective phase 3 randomized clinical trial NRG Oncology-RTOG 0617 assessed 483 patients receiving chemoradiotherapy (3D-CRT vs IMRT) for locally advanced NSCLC based on stratification. Main Outcomes and Measures:Long-term outcomes were analyzed, including overall survival (OS), progression-free survival (PFS), time to local failure, development of second cancers, and severe grade 3 or higher adverse events (AEs) per Common Terminology Criteria for Adverse Events, version 3. The percentage of an organ volume (V) receiving a specified amount of radiation in units of Gy is reported as V(radiation dose). Results:Of 483 patients (median [IQR] age, 64 [57-70] years; 194 [40.2%] female), 228 (47.2%) received IMRT, and 255 (52.8%) received 3D-CRT (median [IQR] follow-up, 5.2 [4.8-6.0] years). IMRT was associated with a 2-fold reduction in grade 3 or higher pneumonitis AEs compared with 3D-CRT (8 [3.5%] vs 21 [8.2%]; P = .03). On univariate analysis, heart V20, V40, and V60 were associated with worse OS (hazard ratios, 1.06 [95% CI, 1.04-1.09]; 1.09 [95% CI, 1.05-1.13]; 1.16 [95% CI, 1.09-1.24], respectively; all P < .001). IMRT significantly reduced heart V40 compared to 3D-CRT (16.5% vs 20.5%; P < .001). Heart V40 (<20%) had better OS than V40 (≥20%) (median [IQR], 2.5 [2.1-3.1] years vs 1.7 [1.5-2.0] years; P < .001). On multivariable analysis, heart V40 (≥20%), was associated with worse OS (hazard ratio, 1.34 [95% CI, 1.06-1.70]; P = .01), whereas lung V5 and age had no association with OS. Patients receiving IMRT and 3D-CRT had similar rates of developing secondary cancers (15 [6.6%] vs 14 [5.5%]) with long-term follow-up. Conclusions and Relevance:These findings support the standard use of IMRT for locally advanced NSCLC. IMRT should aim to minimize lung V20 and heart V20 to V60, rather than constraining low-dose radiation bath. Lung V5 and age were not associated with survival and should not be considered a contraindication for chemoradiotherapy. Trial Registration:ClinicalTrials.gov Identifier: NCT00533949.

Radiation therapy (RT) is a frontline approach to treating cancer. While the target of radiation dose delivery is the tumor, there is an inevitable spill of dose to nearby normal organs causing complications. This phenomenon is known as radiotherapy toxicity. To predict the outcome of the toxicity, statistical models can be built based on dosimetric variables received by the normal organ at risk (OAR), known as Normal Tissue Complication Probability (NTCP) models. To tackle the challenge of the high dimensionality of dosimetric variables and limited clinical sample sizes, statistical models with variable selection techniques are viable choices. However, existing variable selection techniques are data-driven and do not integrate medical domain knowledge into the model formulation. We propose a knowledge-constrained generalized linear model (KC-GLM). KC-GLM includes a new mathematical formulation to translate three pieces of domain knowledge into non-negativity, monotonicity, and adjacent similarity constraints on the model coefficients. We further propose an equivalent transformation of the KC-GLM formulation, which makes it possible to solve the model coefficients using existing optimization solvers. Furthermore, we compare KC-GLM and several well-known variable selection techniques via a simulation study and on two real datasets of prostate cancer and lung cancer, respectively. These experiments show that KC-GLM selects variables with better interpretability, avoids producing counter-intuitive and misleading results, and has better prediction accuracy.

Patients with metastatic epidural spinal cord compression (MESCC) and favorable survival prognoses may benefit from radiation doses exceeding 10 × 3.0 Gy. In a multi-center phase 2 trial, patients receiving 15 × 2.633 Gy (41.6 Gy10) or 18 × 2.333 Gy (43.2 Gy10) were evaluated for local progression-free survival (LPFS), motor/sensory functions, ambulatory status, pain, distress, toxicity, and overall survival (OS). They were compared (propensity score-adjusted Cox regression) to a historical control group (n = 266) receiving 10 × 3.0 Gy (32.5 Gy10). In the phase 2 cohort, 50 (of 62 planned) patients were evaluated for LPFS. Twelve-month rates of LPFS and OS were 96.8% and 69.9%, respectively. Motor and sensory functions improved in 56% and 57.1% of patients, and 94.0% were ambulatory following radiotherapy. Pain and distress decreased in 84.4% and 78.0% of patients. Ten and two patients experienced grade 2 and 3 toxicities, respectively. Phase 2 patients showed significantly better LPFS than the control group (p = 0.039) and a trend for improved motor function (p = 0.057). Ambulatory and OS rates were not significantly different. Radiotherapy with 15 × 2.633 Gy or 18 × 2.333 Gy was well tolerated and appeared superior to 10 × 3.0 Gy.

Purpose: We report 5-year oncologic outcomes of a prospective series of patients with prostate cancer treated with spot-scanning proton therapy (SSPT). Methods and Materials: A prospective registry identified fi ed patients with prostate cancer treated with SSPT between January 2016 and December 2018. Five-year overall survival, local control, biochemical failure, regional and distant failures, and adverse events (AEs) were assessed. Biochemical failure was defined fi ned as rise in prostate-specific fi c antigen >= 2.0 ng/mL above nadir prostate-specific fi c antigen. Baseline-adjusted toxicities were assigned using the Common Terminology Criteria for Adverse Events version 5.0. Results: With a median follow-up of 4.4 years, 284 patients with prostate cancer were treated with SSPT. Median total radiation dose was 79.2 Gy over 44 fractions, 70 Gy over 28 fractions, and 38 Gy over 5 fractions for conventional fractionation (CF), hypofractionation (HF), and stereotactic body radiation therapy (SBRT), respectively. Biochemical failure rate for all patients was 6.7%. Five-year local control rates for CF, HF, and SBRT were 100%, 100%, and 97.3%, respectively (P P = .07). Regional recurrences occurred in 12 (4.2%) patients: 8 treated with CF, 2 with HF, and 2 with SBRT (P P = .62). Distant failures occurred in 12 patients (4.2%): 5 treated with CF, 7 with HF, and none with SBRT (P P = .05). Five-year overall survival for patients treated with CF, HF, and SBRT SSPT were 88.1%, 86.1%, and 97.2%, respectively (P P = .1). Acute and chronic grade 2+ gastrointestinal AEs occurred in 8 (2.8%) and 51 (18.0%) patients, respectively. Acute and chronic grade 3+ gastrointestinal AEs occurred in 3 (1.1%) and 4 (1.4%) patients, respectively. Acute and chronic grade 2+ genitourinary-related AEs were observed in 71 (25%) and 63 (22.2%) patients, respectively. Acute and chronic grade 3+ genitourinary toxicity were observed in 3 (1.1%) and 6 (2.1%) patients, respectively. Conclusions: SSPT provides high local control rates and excellent oncologic outcomes across different fractionation schedules with low long-term AE rates. Published by Elsevier Inc. on behalf of American Society for Radiation Oncology. This is an open access article under the CC BY- NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

We propose an oAPT workflow that incorporates all these functionalities and validate its clinical implementation feasibility with prostate patients. AI-based auto-segmentation tool AccuContourTM (Manteia, Xiamen, China) was seamlessly integrated into oAPT. Initial spot arrangement tool on the vCT for re-optimization was implemented using raytracing. An LET-based biological effect evaluation tool was developed to assess the overlap region of high dose and high LET in selected OARs. Eleven prostate cancer patients were retrospectively selected to verify the efficacy and efficiency of the proposed oAPT workflow. The time cost of each component in the workflow was recorded for analysis. The verification plan showed significant degradation of the CTV coverage and rectum and bladder sparing due to the interfractional anatomical changes. Re-optimization on the vCT resulted in great improvement of the plan quality. No overlap regions of high dose and high LET distributions were observed in bladder or rectum in re-plans. 3D Gamma analyses in PSQA confirmed the accuracy of the re-plan doses before delivery (Gamma passing rate = 99.57%), and after delivery (98.59%). The robustness of the re-plans passed all clinical requirements. The average time for the complete execution of the workflow was 9.12minutes, excluding manual intervention time. The AI-facilitated oAPT workflow was demonstrated to be both efficient and effective by generating a re-plan that significantly improved the plan quality in prostate cancer treated with PBSPT.

PURPOSE: We report 5-year oncologic outcomes of a prospective series of patients with prostate cancer treated with spot-scanning proton therapy (SSPT).METHODS AND MATERIALS: A prospective registry identified patients with prostate cancer treated with SSPT between January 2016 and December 2018. Five-year overall survival (OS), local control (LC), biochemical failure (BF), regional and distant failures, and adverse events (AEs) were assessed. Biochemical failure was defined as rise in PSA ≥ 2.0 ng/mL above nadir PSA. Baseline-adjusted toxicities were assigned using CTCAE v5.0.RESULTS: With a median follow up of 4.4 years, 284 prostate cancer patients were treated with SSPT. Median total radiation dose was 79.2 Gy over 44 fractions, 70 Gy over 28 fractions, and 38 Gy over 5 fractions for conventional fractionation (CF), hypofractionation (HF), and stereotactic body radiation therapy (SBRT), respectively. Biochemical failure rate for all patients was 6.7%. Five-year LC rates for CF, HF, and SBRT were 100%, 100%, and 97.3%, respectively (p = 0.07). Regional recurrences occurred in 12 (4.2%) patients: 8 treated with CF, 2 with HF, and 2 with SBRT (p = 0.62). Distant failures occurred in 12 patients (4.2%): 5 treated with CF, 7 with HF, and none with SBRT (p = 0.05). Five-year OS for patients treated with CF, HF, and SBRT SSPT were 88.1%, 86.1%, and 97.2%, respectively (p = 0.1). Acute and chronic grade 2+ GI AEs occurred in 8 (2.8%) and 51 (18.0%) patients, respectively. Acute and chronic grade 3+ GI AEs occurred in 3 (1.1%) and 4 (1.4%) patients, respectively. Acute and chronic grade 2+ GU-related AEs were observed in 71 (25%) and 63 (22.2%) patients, respectively. Acute and chronic grade 3+ GU toxicity were observed in 3 (1.1%) and 6 (2.1%) patients, respectively.CONCLUSIONS: SSPT provides high local control rates and excellent oncologic outcomes across different fractionation schedules with low long-term AE rates.

Purpose/Objective(s) Real-time cardiac side effects due to thoracic radiotherapy (RT) have never been previously characterized in patients (pts) with lung and esophageal cancers. We specifically designed and completed a prospective clinical trial to address this. Materials/Methods A planned cohort of 24 pts were accrued from Dec 2019 to Jan 2023. Two (8%) pts withdrew their consent prior to ICM insertion. All pts met eligibility criteria, which included ≥ 18 years; non-metastatic, de novo lung or esophageal cancer diagnosis; receiving standard-of-care curative RT or chemoRT with an anticipated heart dose V40 Gy ≥ 20 cc; and planned RT dose ≥ 40 Gy. Pts with any prior RT to the heart were excluded. The ICM provided continuous outpatient arrhythmia monitoring (24/7), and all cardiac rhythms were captured prior to RT, and 4 weeks, 3, 9, and 12 months after RT. The ICM was explanted at 12-month follow-up. The ICM automatically captured and alerted clinicians to the following cardiac events: bradycardia ≤ 40 bpm; asystole with pauses ≥ 3 seconds (s); high degree AV block ≤ 30 bpm lasting ≥ 8 seconds; symptomatic tachycardia ≥ 150 bpm for any duration; and atrial fibrillation. Clinical events such as heart failure were also recorded. The primary endpoint was pts’ 12-month cardiac event rate (both clinically and by ICM) after RT completion. Clopper-Pearson confidence intervals were used in the analysis. Results The final analysis included all 22 pts. Average age was 67 years; 13 pts were male. Eighteen pts finished 12-mo follow-up per protocol, 1 died during treatment, and 3 died during follow-up. At baseline, 4 (19%) reported arrhythmia, 2 (10%) had coronary artery disease, and 1 (5%) with prior myocardial infarction; 15 (68%) were past smokers. Seventeen (77%) received proton beam therapy, and 5 (23%) received photon therapy. The median RT dose was 50 Gy in 2 Gy/fraction. RT was interrupted in 2 (9%) pts, due to needing an ablation for atrial flutter and hospitalization. Cardiac changes as defined in the study were seen in 15 (68%) pts. ICMs detected events in 14 (64%) pts: 10 with atrial arrhythmias (fibrillation/flutter), 2 transient asystoles, 1 non-sustained ventricular tachycardia, and 1 paroxysmal 3rd degree AV block. These events led to 4 interventions including 2 atrial ablations, 1 pacemaker insertion, 1 aortic valve replacement (18% of pts, 95% CI = 5-40%). Conclusion With real-time cardiac monitoring, cardiac events were detected in 64% of pts within 12 months post-RT, leading to timely medical diagnoses and interventions, with potentially improved outcomes. This novel prospective data highlighted the possible benefits of close cardiac surveillance. Further prospective studies are warranted to study the broad impact of ICM-based cardiac evaluation in better characterizing the intricate effects of thoracic RT on the heart and the cardiac conduction system.

BACKGROUND: Prolonged survival of patients with metastatic disease has furthered interest in metastasis-directed therapy (MDT). RESEARCH QUESTION: There is a paucity of data comparing lung MDT modalities. Do outcomes among sublobar resection (SLR), stereotactic body radiation therapy (SBRT), and percutaneous ablation (PA) for lung metastases vary in terms of local control and survival? STUDY DESIGN AND METHODS: Medical records of patients undergoing lung MDT at a single cancer center between January 2015 and December 2020 were reviewed. Overall survival, local progression, and toxicity outcomes were collected. Patient and lesion characteristics were used to generate multivariable models with propensity weighted analysis. RESULTS: Lung MDT courses (644 total: 243 SLR, 274 SBRT, 127 PA) delivered to 511 patients were included with a median follow-up of 22 months. There were 47 local progression events in 45 patients, and 159 patients died. Two-year overall survival and local progression were 80.3% and 63.3%, 83.8% and 9.6%, and 4.1% and 11.7% for SLR, SBRT, and PA, respectively. Lesion size per 1 cm was associated with worse overall survival (hazard ratio, 1.24; P = .003) and LP (hazard ratio, 1.50; P < .001). There was no difference in overall survival by modality. Relative to SLR, there was no difference in risk of local progression with PA; however, SBRT was associated with a decreased risk (hazard ratio, 0.26; P = .023). Rates of severe toxicity were low (2.1%-2.6%) and not different among groups. INTERPRETATION: This study performs a propensity weighted analysis of SLR, SBRT, and PA and shows no impact of lung MDT modality on overall survival. Given excellent local control across MDT options, a multidisciplinary approach is beneficial for patient triage and longitudinal management.

PURPOSE:To assess the clinical acceptability of a commercial deep-learning-based auto-segmentation (DLAS) prostate model that was retrained using institutional data for delineation of the clinical target volume (CTV) and organs-at-risk (OARs) for postprostatectomy patients, accounting for clinical and imaging protocol variations. METHODS AND MATERIALS:CTV and OARs of 109 prostate-bed patients were used to evaluate the performance of the vendor-trained model and custom retrained DLAS models using different training quantities. Two new models for OAR structures were retrained (n = 30, 60 data sets), while separate models were trained for a new CTV structure (n = 30, 60, 90 data sets), with the remaining data sets used for testing (n = 49, 19). The dice similarity coefficient (DSC), Hausdorff distance, and mean surface distance were evaluated. Six radiation oncologists performed a qualitative evaluation scoring both preference and clinical utility for blinded structure sets. Physician consensus data sets identified from the qualitative evaluation were used toward a separate CTV model. RESULTS:Both the 30- and 60-case retrained OAR models had median DSC values between 0.91 to 0.97, improving significantly over the vendor-trained model for all OARs except the penile bulb. The brand new 60-case CTV model had a median DSC of 0.70 improving significantly over the 30-case model. DLAS (60-case model) and manual contours were blinded and evaluated by physicians with contours deemed acceptable or precise for 87% and 94% of cases for DLAS and manual delineations, respectively. DLAS-generated CTVs were scored precise or acceptable in 54% of cases, compared with the manual delineation value of 73%. The 30-case physician consensus CTV model did not show a significant difference compared with the randomly selected models. CONCLUSIONS:Custom retraining using institutional data leads to performance improvement in the clinical utility and accuracy of DLAS for postprostatectomy patients. A small number of data sets are sufficient for building an institutional site-specific DLAS OAR model, as well as for training new structures. Data indicates the workload for identifying training data sets could be shared among groups for the male pelvic region, making it accessible to clinics of all sizes.

Purpose: Planning target volume (PTV) expansion for post-prostatectomy radiotherapy is typically ≥5 mm. Recent clinical trials have proved the feasibility of a reduced margin of 2–3 mm for treatments on MRI-linac. We aim to study the minimum PTV margin needed using iterative cone-beam CT (iCBCT) as image guidance on conventional linacs. Materials/Methods: Fourteen patients who received post-prostatectomy irradiation (8 with an endorectal balloon and 6 without a balloon) were included in this study. Treatment was delivered with volumetric modulated radiation therapy (VMAT). Fractional dose delivery was evaluated in 165 treatment fractions. The bladder, rectal wall, femoral heads, and prostate bed clinical tumor volume (CTV) were contoured and verified on daily iCBCT. PTV margins (0 mm, 2 mm, and 4 mm) were evaluated on daily iCBCT. CTV coverage and OAR dose parameters were assessed with each PTV margin. Results: CTV D100% was underdosed with a 0 mm margin in 32% of fractions in comparison with 2 mm (6%) and 4 mm (6%) PTV margin (p ≤ 0.001). CTV D95% > 95% was met in 93–94% fractions for all PTV expansions. CTV D95% > 95% was achieved in more patients with an endorectal balloon than those without: 0 mm—90/91 (99%) vs. 63/74 (85%); 2 mm—90/91 (99%) vs. 65/75 (87%); 4 mm—90/90 (100%) vs. 63/73 (86%). There was no difference in absolute median change in CTV D95% (0.32%) for 0-, 2-, and 4 mm margins. The maximum dose remained under 108% for 100% (0 mm), 97% (2 mm), and 98% (4 mm) of images. Rectal wall maximum dose remained under 108% for 100% (0 mm), 100% (2 mm), and 98% (4 mm) of images. Conclusions: With high-quality iCBCT image guidance, PTV margin accounting for inter-fractional uncertainties can be safely reduced for post-prostatectomy radiotherapy. For fractionated radiotherapy, an isotropic expansion of 2 mm and 4 mm may be considered for margin expansion with and without the endorectal balloon. Future application for margin reduction needs to be further evaluated and considered with the advent of shorter post-prostatectomy radiation courses.