International Journal Of Radiation Oncology, Biology, Physics[JOURNAL]
Sun
200 papers
Radiomics-Guided Automatic Delineation for Clinical Target Volume of Endometrial Cancer: Limited Sample Multicenter Study.
Qu A, Zhang X, Song Y … +12 more , Zhuo Y, Zhou X, Xiong W, Yang H, Xiao Z, Jiang W, Deng X, Jia L, Wang J, Xue X, Wei L, Jiang P
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41997452 · Publisher ↗
PURPOSE: The inconsistent delineation of the clinical target volume (CTV) in postoperative pelvic radiation therapy for endometrial carcinoma across centers poses a challenge for deep learning-based segmentation due to d...
PURPOSE: The inconsistent delineation of the clinical target volume (CTV) in postoperative pelvic radiation therapy for endometrial carcinoma across centers poses a challenge for deep learning-based segmentation due to differing definitions of the internal target volume. This study aimed to develop an effective method to address multi-institutional variations in CTV delineation, even under the constraints of limited data availability. METHODS AND MATERIALS: A total of 207 simulated computed tomography cases of patients with endometrial cancer across 5 centers were retrospectively collected. Within each center, the data were divided into support, query, and test sets. Each center was sequentially designated as the target center for fine-tuning and testing, while the remaining 4 centers were used for model training to validate the superiority of the proposed method. In addition, 26 cases from an external center were used exclusively for fine-tuning and testing. Radiomics features were extracted to analyze differences in CTV delineation and images across centers. A random forest classifier was trained to identify the most important radiomics features. Using these features as guidance, a model-agnostic meta-learning (MAML) strategy was applied to pretrain a 3-dimensional U-Net radiomics-guided MAML (MAML-r) model, which was subsequently fine-tuned on each target center's data. The performance of the proposed MAML-r method was compared with direct 3-dimensional U-Net training and transfer learning models. Evaluation metrics included the Dice similarity coefficient (DSC), the 95th percentile Hausdorff distance (HD95), and the average symmetric surface distance (ASSD), supplemented by qualitative assessments from clinical experts using a 4-point scoring system. RESULTS: Eight important features were identified from a total of 107 radiomics features, which showed significant differences across centers (P < .01). The MAML-r model yielded meaningful results, achieving a mean ± SD DSC of 0.818 ± 0.058, a mean ± SD HD95 of 9.314 ± 3.648 mm, and a mean ± SD ASSD of 2.772 ± 1.090 mm. The model also earned an average blinded expert evaluation score of 3.24, significantly outperforming all other models. Notably, improved performance was observed in the external test cohort, with corresponding mean ± SD values for DSC, HD95, and ASSD of 0.886 ± 0.012, 5.203 ± 1.435 mm, and 1.512 ± 0.334 mm, respectively. Furthermore, the MAML-r model achieved the shortest mean ± SD CTV modification time of 3.8 ± 1.2 minutes. Given the variations in CTV contouring styles across centers and the limited sample size, the MAML-r model demonstrated superior performance and adaptability compared with the other models. CONCLUSIONS: This study introduces a novel MAML-r framework for few-shot, multicentric CTV segmentation tasks in postoperative pelvic radiation therapy for endometrial carcinoma, significantly mitigating performance degradation caused by interinstitutional variations in delineation styles and data scarcity. The proposed approach offers a promising solution to these persistent clinical challenges.
Estimation of Detailed Cardiac Doses for Pediatric Radiation Therapy Patients in National Wilms Tumor Study.
Lee C, Mille MM, Griffin KT … +12 more , Rigsby C, Popescu A, Gopalakrishnan M, Leisenring W, Peterson S, Dome JS, Laurie F, Fitzgerald TJ, Bentzen S, Jung JW, Lee C, Kalapurakal JA
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41997451 · Full text
PURPOSE: Substructure-level heart dosimetry may improve the evaluation of long-term cardiac toxicity in childhood cancer survivors, but detailed pediatric heart models are limited. We developed age-specific heart models...
PURPOSE: Substructure-level heart dosimetry may improve the evaluation of long-term cardiac toxicity in childhood cancer survivors, but detailed pediatric heart models are limited. We developed age-specific heart models (1, 5, 10, and 15 years) using high-resolution imaging and integrated them into computational phantoms to estimate cardiac substructure doses in patients treated according to the National Wilms Tumor Study protocols and evaluate the impact of anatomic detail on radiation therapy dose estimates. METHODS AND MATERIALS: Heart models with detailed substructures, including chambers, myocardium, arteries, valves, and conduction nodes, were developed from pediatric magnetic resonance and adult computed tomography images. These were incorporated into a size-dependent phantom library representing a wide range of pediatric body sizes. Patient-specific radiation therapy plans were reconstructed using the Pinnacle treatment planning system, and heart doses were calculated using both treatment planning system and Monte Carlo (MC) methods. RESULTS: The developed heart models closely matched ICRP reference masses (within 2%). Treatment planning system and MC dose calculations showed strong agreement (median difference < 2%), so that MC-based doses were used for further analysis. Among 4,716 National Wilms Tumor Study patients treated with radiation therapy, the median whole heart dose was 4.2 Gy. Cardiac and substructure doses varied by treatment region, with the right atrium and left ventricular myocardium receiving higher doses (≤5.1 and 4.5 Gy), whereas coronary arteries and valves received lower doses (<1 Gy). In non-chest fields, substructure doses differed significantly from whole heart doses (P < .001), reflecting steep intracardiac dose gradients. Chest fields alone resulted in uniformly high cardiac doses with minimal variation. CONCLUSION: Our results demonstrate that relying solely on whole heart dose may obscure clinically relevant exposure to critical substructures. Detailed heart models enable more accurate dosimetry and support improved risk assessment and safer pediatric radiation therapy planning to reduce long-term cardiac toxicity.
Childhood Nasopharyngeal Carcinoma: A Rare Disease With Distinct Features and Favorable Prognosis.
Sun S, Huang X, Liu Y … +12 more , Yu Z, Wang K, Qu Y, Zhang Y, Wang J, Chen X, Ma Y, Zhang J, Luo J, Li YX, Wu R, Yi J
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41990887 · Publisher ↗
PURPOSE: This study aims to compare clinical features, treatment responses, radiosensitivity, and failure patterns between childhood and adult nasopharyngeal carcinoma (NPC) patients in the era of intensity modulated rad...
PURPOSE: This study aims to compare clinical features, treatment responses, radiosensitivity, and failure patterns between childhood and adult nasopharyngeal carcinoma (NPC) patients in the era of intensity modulated radiation therapy. METHODS AND MATERIALS: A retrospective review was conducted on 140 childhood NPC patients, aged <21 years, treated at a single institution between January 2004 and November 2019. A propensity score matching method was used to select 280 matched adult NPC patients in a 1:2 ratio. Comparative analysis was performed between the childhood and adult cohorts. RESULTS: Childhood NPC is associated with more advanced clinical stages, heightened radiosensitivity, and improved prognostic outcomes compared to adult NPC. The 5-year rates for overall survival (OS), progression-free survival (PFS), locoregional relapse (LRR) were significantly higher in the childhood group compared with the adult group: 83.1% versus 71.9% (P = .001) for OS, 71.6% versus 60.9% (P = .023) for PFS, and 3.1% versus 12.6% (P = .002) for LRR, respectively. Childhood NPC exhibited heightened radiosensitivity, and high locoregional control can be achieved even in cases that showed no response to induction chemotherapy, which is unlike in adults. In children receiving induction chemotherapy, no significant differences were found in OS (84.0% vs 75.7%, P = .289), PFS (75.2% vs 57.7%, P = .123), LRR (4.4% vs 3.8%, P = .656), and distant metastasis (22.4% vs 38.5%, P = .122) between those with complete/partial response and those with stable/progression disease. Late recurrence after 2 years was less frequent in children than in adults (13.6% vs 29.2%, P = .039), and a higher proportion of childhood patients were successfully salvaged (42.1% vs 21.7%, P = .015). CONCLUSIONS: Childhood NPC often presents at an advanced stage but has a favorable prognosis with excellent locoregional control. Children's heightened radiosensitivity allows for potential dose reduction. Given the risk of late effects, individualized treatment and follow-up strategies are warranted for young patients.
Dosimetric Results for Organs at Risk From Radiation Therapy Dose De-Escalation to the Subclinical Region in Human Papillomavirus-Associated Oropharyngeal Cancer: A Propensity Score-Matched Analysis.
Miller DG, Yacoub I, Aliotta E … +13 more , Kallini D, Wu Y, Yu Y, Zakeri K, Chen L, Kang JJ, Shamseddine AA, Gelblum DY, McBride S, Zeng C, Aristophanous M, Riaz N, Lee NY
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41985762 · Publisher ↗
PURPOSE: The impact of radiation therapy (RT) dose de-escalation on organs at risk (OARs) for human papillomavirus-associated oropharyngeal cancer (OPC) is not known. This study retrospectively assessed OAR dosimetric ou...
PURPOSE: The impact of radiation therapy (RT) dose de-escalation on organs at risk (OARs) for human papillomavirus-associated oropharyngeal cancer (OPC) is not known. This study retrospectively assessed OAR dosimetric outcomes of de-escalated RT for OPC. METHODS AND MATERIALS: We conducted a retrospective analysis of dosimetric data for patients with T1 to 4, N0 to 3 human papillomavirus-associated OPC treated with intensity modulated RT at a single academic institution between August, 2014 and January, 2023. All patients received 70 Gy to gross disease. The low-dose group received 30 Gy to the subclinical neck, and high-dose patients received 50-60 Gy. Propensity score matching was performed to balance differences in clinical staging between groups. Mean OAR doses were collected and compared between the groups. RESULTS: A total of 575 patients were included. Of these, 226 (39.3%) had bilateral nodal disease. Significant reductions in mean doses were observed across all OARs. For matched 30 Gy patients with unilateral nodal disease, the contralateral parotid gland mean dose was 833 cGy, compared with 1391 cGy (P< .001). Other mean dose reductions included ipsilateral parotid gland (1984 vs 2769 cGy; P< .001), ipsilateral submandibular gland (5709 vs 6036 cGy; P= .008), contralateral submandibular gland (2832 vs 3734 cGy; P< .001), and larynx (2119 vs 3186 cGy; P< .001). Patients with bilateral nodal disease also had lower OAR doses in the 30 Gy group, including right parotid (1661 vs 2402 cGy; P< .001), left parotid (1725 vs 2443 cGy; P< .001), right submandibular gland (5385 vs. 6001; P= .002), left submandibular gland (5283 vs 5948 cGy; P= .005), and larynx (2341 vs 3846 cGy; P< .001). CONCLUSION: Subclinical neck RT dose de-escalation can generate low OAR doses well under established dose constraints. These reductions have the potential to mitigate toxicities such as xerostomia and dysphagia, and they may improve long-term patient quality of life. Ongoing prospective research evaluating toxicities and quality of life from RT de-escalation for OPC is warranted.
Is Deep Learning Ready for Abdominal Organ-at-Risk Segmentation in the Foundation Model Era: A Comprehensive Study of Challenging Clinical Cases.
Fu J, Li H, Luo Z … +6 more , He Y, Zhang S, Zou X, Wu Y, Wang G, Liao W
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41962748 · Publisher ↗
PURPOSE: The application of deep learning-based methods for accurate organ-at-risk segmentation in challenging clinical scenarios remains unexplored. This study aims to evaluate state-of-the-art fully supervised learning...
PURPOSE: The application of deep learning-based methods for accurate organ-at-risk segmentation in challenging clinical scenarios remains unexplored. This study aims to evaluate state-of-the-art fully supervised learning (FSL) methods and foundation model (FM)-based methods across challenging clinical scenarios, and to propose an effective solution to improve model robustness and reduce organ hallucination. METHODS AND MATERIALS: We retrospectively collected computed tomography (CT) scans from 413 patients across 2 institutions, divided into 3 cohorts based on treatment strategy. Seven FSL and 6 FM methods were comprehensively evaluated on an internal testing cohort (n = 67, without surgery), external testing cohort 2 (n = 22, partial organ resection surgery), and external testing cohort 3 (n = 74, whole organ resection surgery), as well as 3 public datasets. We further introduced an organ erasure augmentation (OEA) strategy to improve generalization and address hallucinations in missing organs. Quantitative metrics included Dice similarity coefficient, normalized surface Dice (NSD), and hallucination ratio. RESULTS: Two of three fine-tuned FM methods failed to produce any segmentation outputs for 5 and 6 of 19 organs, respectively. Prompt-based FM methods using tight bounding box prompts demonstrated stable performance but struggled with complex anatomy such as the intestine. Our proposed OEA method outperformed existing FM-based and FSL methods, achieving mean Dice similarity coefficient and mean normalized surface Dice of 87.29% and 87.84% on the internal testing cohort, 85.15% and 85.01% on the external testing cohort 2, and 82.29% and 81.81% on the external testing cohort 3, respectively. Compared with the best-performing method, our method reduced the mean hallucination ratio from 0.571 to 0.516 and demonstrated superior cross-dataset generalization with less performance degradation. CONCLUSIONS: Current FM-based and FSL methods remain insufficient for clinical use in cases involving irregular anatomy or significant distribution shifts. The proposed OEA strategy reduces hallucination and enhances segmentation robustness, offering a promising step toward reliable clinical application.
Split-Dose FLASH Irradiation to Investigate the Clinical Feasibility of Multifield Treatments: The Effect of Split Dose and Dose Rate on FLASH.
Erickson DPJ, Shaver BA, Grassberger C … +11 more , Morimoto A, Seitz Z, Cui S, Cao N, Saini J, Wong T, Mian O, Ford E, Rengan R, Zeng J, Schwarz M
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41962747 · Publisher ↗
PURPOSE: The "FLASH effect," a phenomenon, in which radiation-induced toxicity is diminished in healthy tissues when treated at ultrahigh dose rates, has been demonstrated in several experimental models. We aimed to eval...
PURPOSE: The "FLASH effect," a phenomenon, in which radiation-induced toxicity is diminished in healthy tissues when treated at ultrahigh dose rates, has been demonstrated in several experimental models. We aimed to evaluate the impact of split doses, the time interval between splits, and the effects of dose-rate variations below and above the 40 Gy/s threshold. METHODS AND MATERIALS: Mice received 16 Gy of pelvic radiation with a scattered or scanning beam. The conventional (CONV) dose rate was set at 0.5 to 1 Gy/s, while the FLASH effect was observed at dose rates ranging from 20 to 120 Gy/s. Mice were irradiated with a single dose or a split-dose regimen (1 or 2 splits), with pauses between treatments of either 30 seconds or 2 minutes. The endpoint was survival. RESULTS: The hazard ratio for single irradiation in FLASH mode versus CONV radiation was 0.31, confirming the presence of the FLASH effect. A split dose with a single 30-second or 2-minute pause reduced overall survival, with hazard ratios of 0.53 and 0.56, respectively. However, survival was still higher than with CONV radiation. Two 2-minute pauses were not significantly worse than one 2-minute pause. The lowest dose rate at which a FLASH effect was detected was 20 Gy/s; no benefit was observed at dose rates above 60 Gy/s. For the 50% survival rate endpoint, the FLASH modification factor was 0.91 and 0.96 for irradiations without and with 1 or 2 pauses, respectively. CONCLUSIONS: The FLASH effect is attenuated by the introduction of a split-dose regimen. In the clinical implementation of FLASH, the benefit of multiple fields should be weighed against the reduction in the FLASH effect. For our endpoint, the minimum dose rate for FLASH is 20 Gy/s, whereas an increase from 60 Gy/s to 120 Gy/s is not beneficial.
Patterns of Failure After Definitive Ablative 5-Fraction Stereotactic Body Radiation Therapy for Inoperable Pancreatic Ductal Adenocarcinoma.
Chuong MD, Herrera R, Extein JE … +17 more , Chundru SN, Luther N, Mittauer KE, Roy M, Carvallo N, Kotecha R, Hall MD, Lee YC, Bejarano T, Bassiri N, Gutierrez AN, Tolakanahalli R, Ucar A, DeZarraga F, Aparo S, Mehta MP, Kaiser A
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41956163 · Publisher ↗
PURPOSE: Definitive stereotactic body radiation therapy (SBRT) for inoperable pancreatic ductal adenocarcinoma (PDAC) is often delivered with a nonablative dose to gross tumor alone despite high rates of locoregional fai...
PURPOSE: Definitive stereotactic body radiation therapy (SBRT) for inoperable pancreatic ductal adenocarcinoma (PDAC) is often delivered with a nonablative dose to gross tumor alone despite high rates of locoregional failure (LRF). The purpose of this study was to characterize patterns of failure after definitive ablative SBRT and evaluate the impact of clinical target volume (CTV) design. METHODS AND MATERIALS: We performed a retrospective cohort study of nonmetastatic PDAC treated with definitive ablative SBRT on a 0.35 Tesla MR-Linac between 2018-2024. Patients who had surgery were excluded. The median prescribed gross tumor volume and CTV doses were 50 Gy and 33 Gy in 5 fractions, respectively. CTV coverage evolved from no CTV or limited perivascular coverage to larger anatomically derived volumes eventually including the "triangle volume." The first diagnostic scan showing LRF was registered to the simulation scan on which the recurrence was contoured. LRFs were classified as in-field, marginal, or out-of-field. RESULTS: Among 121 consecutive patients, 87.6% received induction chemotherapy, and 92.6% were treated with a CTV. Median follow-up after SBRT was 12.0 months. LRF occurred in 17 patients (14.0%) at a median of 14.3 months and no LRF was observed in patients treated to the "triangle volume." In-field failures were rare (2.5%) as were marginal (6.6%) and out-of-field (5.0%) failures. LRF involved the primary tumor (n = 5; 29.4%), paraaortic lymph nodes (n = 4; 23.5%), porta hepatis (n = 4; 23.5%), superior mesenteric artery (n = 3; 17.6%), or celiac artery (n = 1; 5.9%). Larger CTV size was associated with a nonstatistically significant increase in acute grade 1 to 2 nausea. CONCLUSIONS: This is the first study to characterize patterns of LRF following induction chemotherapy and definitive ablative SBRT for inoperable PDAC. Our findings suggest that routine use of an anatomically derived CTV should be considered including the "triangle volume."
Improved Progression-Free Survival Following Dose-Escalated Versus Standard-Dose Postoperative Radiation Therapy for High-Risk Meningiomas: An International Multicenter Individual Patient-Level Meta-Analysis (FIRESTORM).
Singh R, Koempel A, French B … +27 more , Crawford C, Perlow HK, Rauh S, Kutuk T, Beyer S, Grecula J, Blakaj D, Zhu S, Raval RR, Kotecha R, Chen WC, Raleigh DR, Frechette KM, Lehrer EJ, Khazanchi R, Chaliparambil RK, Sadagopan NS, Ho A, Youngblood MW, Sachdev S, Magill ST, Burgess L, Deng MY, Debus J, König L, Sahgal A, Palmer JD
Int J Radiat Oncol Biol Phys · 2026 Jul · PMID 41956162 · Publisher ↗
PURPOSE: We performed an individual patient-level meta-analysis of high-risk meningiomas to compare the outcomes of dose-escalated radiation therapy (DE-RT) versus standard-dose postoperative radiation therapy (SD-RT). M...
PURPOSE: We performed an individual patient-level meta-analysis of high-risk meningiomas to compare the outcomes of dose-escalated radiation therapy (DE-RT) versus standard-dose postoperative radiation therapy (SD-RT). METHODS AND MATERIALS: A total of 7 institutions participated. DE-RT was defined as treatment with a biologically effective dose of ≥79.2 Gy (equivalent of 66 Gy in 33 fractions). We compared progression-free survival (PFS) with DE-RT versus SD-RT via Kaplan-Meier analysis and log-rank t tests, a Cox proportional hazards multivariable model, and propensity score analyses with inverse probability of treatment weighting (IPTW). We also compared incidences of central nervous system radionecrosis (RN) with DE-RT versus SD-RT. RESULTS: The analysis included 248 patients with high-risk meningioma (59 received DE-RT and 189 received SD-RT). One hundred and eighty-eight cases (75.8%) were World Health Organization grade 2, and 103 cases (41.5%) were recurrent meningiomas. Extent of resection was subtotal resection in 182 of 248 (75.2%). Three- and 5-year PFS rates were 62.8% (95% CI, 55.8%-69.0%) and 45.0% (95% CI, 37.3%-52.3%), respectively. DE-RT was associated with superior PFS rates (P = .0022), with 3-year (86.4% vs 55.6%) and 5-year (65.8% vs 38.8%) PFS rates favoring DE-RT. On multivariable analysis, DE-RT was associated with superior PFS (hazard ratio, 0.40; 95% CI, 0.24-0.69; P = .001). On IPTW, DE-RT continued to be associated with superior PFS (hazard ratio, 0.45; 95% CI, 0.24-0.83; P = .01). A greater incidence of any grade RN was observed following DE-RT (20 of 59; 33.9%) versus SD-RT (25 of 189; 13.2%) (P = .001) but with similar grade 3 or greater RN events (DE-RT 5.1% vs SD-RT 3.2%). CONCLUSIONS: DE-RT resulted in superior PFS for patients with high-risk meningiomas over SD-RT without an increase in severe toxicities.
Multicenter End-to-End Evaluation of Single-Isocenter Multitarget Stereotactic Radiosurgery Approaches: Accuracy and Plan Quality.
Thomann B, Fechter T, Fischer J … +35 more , Runz A, Ludwig U, Sachpazidis I, Blanck O, Roers J, Grehn M, Grohmann M, Ziemann C, Judge M, Baus W, Grahle M, Walke M, Bathen B, Köhn J, Käthner P, Shariff M, Wegner N, Fleckenstein J, Karle H, Streller T, Howitz S, Priegnitz M, Weigel R, Konrad T, Schmitt D, Beck J, Machein M, Popp I, Reiner M, Moustakis C, Karger CP, Pappas E, Grosu AL, Bock M, Baltas D
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41956161 · Publisher ↗
PURPOSE: We conducted a multicenter study on single-isocenter multitarget stereotactic radiosurgery to dosimetrically assess end-to-end test results and identify approaches and techniques influencing spatial accuracy and...
PURPOSE: We conducted a multicenter study on single-isocenter multitarget stereotactic radiosurgery to dosimetrically assess end-to-end test results and identify approaches and techniques influencing spatial accuracy and treatment plan quality. METHODS AND MATERIALS: An anthropomorphic head phantom with radiochromic film and polymer gel inserts was used with a reference structure set of 5 brain metastases. End-to-end tests were performed on-site at 23 centers in Germany, Austria, and Switzerland, each following its own single-isocenter multitarget stereotactic radiosurgery protocol. Spatial accuracy was quantified by comparing planned and measured prescription isodose-volume centroids. Plan quality was assessed from treatment planning system calculations using the Paddick gradient index (GI) and Paddick conformity index. Statistical analyses, including a generalized linear model, correlated results with protocol parameters to identify favorable systems and techniques. RESULTS: The mean spatial offset between measured and calculated prescription isodose centroids across all centers and targets was 0.9 ± 0.4 mm. Offsets above 1 mm were observed in 33% of centers. Better imaging-to-radiation isocenter consistency (ICC) yielded significantly higher accuracy (P = .002): d = 0.6 ± 0.2 mm versus 1.1 ± 0.3 mm. Mean GI was 6.7 ± 3.3. Automated planning (AP) tools achieved significantly lower GI (4.9 ± 0.7) than conventional planning (8.1 ± 3.8, P = .015). Mean Paddick conformity index was 0.75 ± 0.17, with AP significantly improving conformity (0.83 ± 0.07 vs 0.68 ± 0.19; P = .028) and reducing variability in both indices. Target-to-isocenter distance had no significant influence on spatial accuracy, GI, or Paddick conformity index. CONCLUSIONS: Spatial accuracy in a static phantom was primarily determined by ICC and less by specific delivery infrastructure or techniques, emphasizing the importance of a precise imaging isocenter calibration. AP tools significantly improved and standardized treatment plan quality across centers.
A Learning-Driven Automatic Planning Framework for Proton Pencil Beam Scanning Treatments of Head and Neck Cancers.
Wang Q, Xiao L, Chang C
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41956160 · Publisher ↗
PURPOSE: Proton pencil beam scanning treatment planning for head and neck cancers involves numerous conflicting objectives, requiring iterative objective parameter adjustments to balance multiple clinical goals. We propo...
PURPOSE: Proton pencil beam scanning treatment planning for head and neck cancers involves numerous conflicting objectives, requiring iterative objective parameter adjustments to balance multiple clinical goals. We propose a learning-driven inverse optimizer and integrate it into a proximal policy optimization (PPO)-based planning framework to automatically generate high-quality plans for patients with diverse treatment requirements. METHODS AND MATERIALS: The inverse optimizer is a learning-to-optimize (L2O) method that predicts update steps by learning from task-specific data distributions. For the first time, long-context processing techniques developed for large language models are used to address the scalability limitations of existing L2O methods, enabling simultaneous optimization over a substantially large set of variables. The PPO framework functions as an outer-loop virtual planner, autonomously adjusting objective parameters through a policy network, and the inner-loop L2O inverse optimizer computes machine-deliverable spot monitor unit values based on the PPO-refined objectives. Moreover, a Swin UNetR dose predictor is trained with prescription- and beam-specific information to estimate the initial objective parameters. In our experiments, a total of 97 patients with bilateral or ipsilateral head and neck cancers were included for training and testing. RESULTS: Compared with second-order gradient-based methods, our L2O optimizer improves the effectiveness and efficiency of the time-consuming inverse optimization by 22.97% and 36.41%, respectively. In conjunction with the PPO-based virtual planner, plans are generated within clinically acceptable times, that is, 2.55 hours on average, and show improved or comparable organs at risk sparing with superior target coverage compared with human-generated plans. CONCLUSIONS: The proposed inverse optimizer is the first L2O model applied to radiation therapy treatment planning and achieves promising performance. The high-quality plans generated for patients with variable prescription dose levels, multiple target volumes, and patient-specific beam angles highlight the strong potential of the proposed automatic planning framework for practical clinical use.
High-Quality Four-Dimensional Magnetic Resonance Fingerprinting Reconstruction for Liver Cancer Radiation therapy.
Liu C, Li T, Wang L … +8 more , Liao W, Wang X, Zeng Z, Teng X, Wong YL, Ho-Fun Lee V, Cao P, Cai J
Int J Radiat Oncol Biol Phys · 2026 May · PMID 41956159 · Publisher ↗
PURPOSE: To develop a high-quality 4-dimensional magnetic resonance fingerprinting (HQ-4DMRF) framework with temporal low-rank-constrained motion compensation for precise tumor motion management in liver radiation therap...
PURPOSE: To develop a high-quality 4-dimensional magnetic resonance fingerprinting (HQ-4DMRF) framework with temporal low-rank-constrained motion compensation for precise tumor motion management in liver radiation therapy. METHODS AND MATERIALS: HQ-4DMRF integrated 4 key innovations: (1) an automated internal respiratory navigator to track organ motion without external sensors; (2) a results-driven phase-sorting algorithm to dynamically redistribute magnetic resonance fingerprinting (MRF) dynamics across respiratory phases; (3) a novel temporal low-rank-constrained 4-dimensional (4D) registration algorithm to simultaneously compute all interphase deformation vector fields by leveraging low-rank respiratory motion properties and enforcing spatiotemporal regularization; and (4) an iterative motion-compensated optimization algorithm to reconstruct motion-resolved 4D tissue maps. HQ-4DMRF was validated in 24 patients with hepatocellular carcinoma. All patients underwent a free-breathing abdominal MRF scan using a multislice 2-dimensional fast acquisition with steady-state precession sequence. The motion measurement accuracy of HQ-4DMRF was assessed through interphase structural repeatability. Interphase structural repeatability quantified the structural consistency in tissue maps across motion phases using the structural similarity index, local cross-correlation, and textural feature intraclass correlation coefficient for tumors. RESULTS: The HQ-4DMRF demonstrated superior precision in motion measurement versus conventional 4DMRF techniques (P < .001), with interphase structural repeatability-structural similarity index/-local cross-correlation/-textural feature intraclass correlation coefficient of 0.82 ± 0.06/0.36 ± 0.07/0.75 ± 0.20 for T1, 0.89 ± 0.05/0.29 ± 0.06/0.84 ± 0.24 for T2, and 0.80 ± 0.06/0.38 ± 0.06/0.91 ± 0.12 for proton density maps. Compared with using conventional pair-wised registration methods, the temporal low-rank-constrained 4D registration improved motion measurement accuracy by an average of 8.5% to 12.5% (structural similarity index), 9.1% to 36.2% (local cross-correlation), and 8.2% to 17.1% (textural feature intraclass correlation coefficient). The respiratory curve derived from automated internal respiratory navigator showed strong agreement with manual measurements (Pearson correlation coefficient = 0.90 ± 0.12) and demonstrated consistent performance across different anatomic regions (Pearson correlation coefficient = 0.83 ± 0.13). The result-driven phase sorting enhanced the 4DMRF performance by 7.2%. CONCLUSIONS: The HQ-4DMRF framework presents a comprehensive solution to critical challenges in 4DMRF. Clinical validation in patients with hepatocellular carcinoma demonstrates significant improvements in liver tumor motion characterization. These advances not only enhance the precision of radiation therapy planning through more accurate motion modeling but also establish HQ-4DMRF as a promising platform for 4D quantitative magnetic resonance imaging in oncologic applications.
Alternative End Joining Dependency Imposed by miR-21-5p Defines Radiation Resistance and a Targetable Vulnerability in Oral Squamous Cell Carcinoma.
Ma L, Tayier N, Hu W … +9 more , Chen J, Wen L, Li Y, Shi X, Li B, Luo Y, Deng Y, Li X, Liu Q
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41946423 · Publisher ↗
PURPOSE: Clinical control of oral squamous cell carcinoma (OSCC) is constrained by heterogeneous radiosensitivity driven by divergent DNA damage response programs. The architecture and functional contribution of alternat...
PURPOSE: Clinical control of oral squamous cell carcinoma (OSCC) is constrained by heterogeneous radiosensitivity driven by divergent DNA damage response programs. The architecture and functional contribution of alternative end joining (Alt-EJ), an error-prone DNA double-strand break (DSB) repair pathway frequently upregulated in cancer, to radiation resistance remains poorly defined. METHODS AND MATERIALS: We profiled microRNAs in radioresistant OSCC clones and performed multiomic integration across an institutional OSCC cohort, an external OSCC cohort from the Gene Expression Omnibus, The Cancer Genome Atlas pan-cancer tumors, and cell lines characterized by Sanger Genomics of Drug Sensitivity in Cancer to infer DNA damage response characteristics, genomic scar features, drug sensitivity, and radiation therapy outcomes. DSB repair capacity and pathway usage were validated using functional assays, including Alt-EJ reporters and droplet digital PCR quantification of microhomology-mediated repair events. Core Alt-EJ effectors such as PARP1 and POLQ were perturbed genetically and pharmacologically. Therapeutic efficacy of PARP or POLQ inhibition with or without irradiation was tested in a syngeneic OSCC model, followed by bulk tumor transcriptomics to assess pathway engagement. RESULTS: Upregulation of miR-21-5p was not only selectively detected in radioresistant OSCC, but also modulated radiosensitivity in vitro and in vivo, and was associated with inferior postradiation therapy survival. A calibrated miR-21-5p target-gene signature tracked Alt-EJ activity across patient and mouse tumors and cancer cell lines, correlated with microhomology-mediated indels and broader genomic scarring, and predicted sensitivity to clinically available PARP inhibitors. Functionally, enforced miR-21-5p expression increased Alt-EJ usage and accelerated DSB repair, whereas inhibition or depletion of key Alt-EJ effectors reduced repair efficiency and restored radiosensitivity. In vivo, Alt-EJ targeting with PARP or POLQ inhibitor abrogated miR-21-5p-driven radiation resistance; transcriptomic profiling supported suppression of Alt-EJ programs as the operative mechanism. CONCLUSIONS: These findings establish a mechanistic link between miR-21-5p activity and Alt-EJ dependence, provide a clinically deployable signature to identify Alt-EJ-dependent OSCC, and support rational combinations of Alt-EJ targeting agents with radiation therapy to overcome treatment failure and advance precision radiation oncology.
Simulation-Free Adaptive Prostate Stereotactic Body Radiation Therapy on Magnetic Resonance-Linac: Technical Feasibility and Clinical Experience.
Côté N, Sun L, Nagar H … +8 more , Brennan V, Chan CY, Maggia G, Mechalakos J, Aliotta E, Yu V, Lalonde A, Tyagi N
Int J Radiat Oncol Biol Phys · 2026 Apr · PMID 41932402 · Publisher ↗
PURPOSE: To develop and evaluate 2 simulation-free (SF) stereotactic body radiation therapy (SBRT) workflows for prostate cancer on magnetic resonance (MR)-Linac: one using a diagnostic MR reference scan and another usin...
PURPOSE: To develop and evaluate 2 simulation-free (SF) stereotactic body radiation therapy (SBRT) workflows for prostate cancer on magnetic resonance (MR)-Linac: one using a diagnostic MR reference scan and another using a preapproved template scan. METHODS AND MATERIALS: Population-based relative electron density (RED) values were derived from 100 prostate SBRT patients on the Elekta Unity MR-Linac and validated on a separate cohort of 10 patients for dosimetric equivalence. Two SF approaches were evaluated in another independent cohort of 10 patients using reference plans derived from: (1) a diagnostic MR scan (5 retrospective, 5 prospective), and (2) a digitally created pelvic phantom scan using population-based RED (all retrospective). The performance of artificial intelligence-generated contours was investigated during adaptive treatments using surface Dice (sDSC) with 1 mm tolerance and added path length. All adaptive plans, from both SF and the clinically delivered workflow, were compared in terms of dose metrics and monitor unit usage. A focused failure modes and effects analysis identified and mitigated diagnostic MR SF-specific risks by calculating the risk priority number (RPN). RESULTS: Population-based RED assignments resulted in mean dose differences under 1%. Artificial intelligence contours showed strong agreement with physician-approved contours, with sDSC and added path length of 0.96/26.2cm,0.98/17.2cm,0.99/0.4cm, and 0.99/0.01cm for bladder, rectum, penile bulb, and bone, respectively. Across both SF workflows, adaptive plans were clinically comparable to the standard workflow, with mean dose differences from -1.7% to 4.6% and mean monitor unit differences remaining under 2%. Failure modes and effects analysis identified 3 high-risk failure modes: mislabeling diagnostic MR scans (RPN=39), case tracking lapses (RPN=29), and incorrect RED assignment (RPN=22), all mitigated through workflow refinements and training. The diagnostic MR SF workflow has been clinically implemented, treating 22 patients without failures. CONCLUSIONS: Both SF workflows eliminated the need for simulation, streamlining adaptive prostate SBRT without compromising plan quality. The template-based workflow further improved patient access by removing imaging prerequisites.
Erratum to 'Evaluation of Radiation Therapy Treatment Plans in a Randomized Phase 2 Trial Comparing 2 Schedules of Twice-Daily Thoracic Radiation Therapy in Limited Stage Small Cell Lung Cancer.' International Journal of Radiation Oncology*Biology*Physics, Volume 120, Issue 2, 1 October 2024, Pages 332-342.
Levin N, Killingberg KT, Halvorsen TO … +2 more , Danielsen S, Grønberg BH
Int J Radiat Oncol Biol Phys · 2026 Jul · PMID 41931067 · Publisher ↗
Abstract loading — click title to view on PubMed.
Executive Summary of the American Radium Society Appropriate Use Criteria for Selective Use of Pelvic Radiation Therapy in Locally Advanced Rectal Cancer: A Consensus Statement.
Expert Panel for Selective Use of Pelvic Radiation in Locally Advanced Rectal Cancer:, Tchelebi LT, Jethwa KR … +17 more , Dozios E, Jin Z, Wilson G, Anker CJ, Abood G, Akselrod D, Attallah J, Codipilly DC, Hallemeier CL, Kennedy T, Lee P, Ling DC, Miller ED, Newman NB, Sharma N, Small W, Russo S
Int J Radiat Oncol Biol Phys · 2026 Mar · PMID 41895479 · Publisher ↗
Pelvic radiotherapy is commonly incorporated in the treatment of locally-advanced rectal cancer to reduce the risk of locoregional recurrence, but can be associated with substantial acute toxicity and long-term morbidity...
Pelvic radiotherapy is commonly incorporated in the treatment of locally-advanced rectal cancer to reduce the risk of locoregional recurrence, but can be associated with substantial acute toxicity and long-term morbidity. However, treatment may be personalized for selective use of radiotherapy in a subset of patients identified to be at low to intermediate risk of locoregional recurrence following resection. This multi-specialty-led committee included gastrointestinal radiation and medical oncology, gastroenterology, radiology, and colorectal surgery. Using the Population, Intervention, Comparator, Outcome, Timing and Study Design (PICOTS) framework, evidence regarding treatment outcomes was assessed using Cochrane and Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology. Eligible studies included prospective and retrospective (n ≥ 50) studies published between 1/1/2005 - 6/24/2025 from Embase, Medline and PubMed databases. Study type and quality were assessed. Well-established RAND corportaion/University of California Los Angeles (RAND-UCLA) consensus methodology (modified Delphi) was used to rate the appropriateness of the treatment options. Of the 110 articles identified using the search strategy, 35 were selected that met all inclusion criteria. Of the 35 references used as evidence, 35 are categorized as therapeutic including 14 well-designed studies (Phase II randomized and Phase III), 12 moderately well-designed studies that account for most common biases (matched cohort and Phase II studies), 7 studies with design limitations (retrospective reviews), and 3 meta-analyses. Variant cases were developed as examples to illustrate practical applications of consensus recommendations for when RT can be safely omitted. A treatment algorithm is also provided to assist with treatment decisions when considering selective use of pelvic RT.
Molecular Risk Stratification for Decision-Making Among Patients With Breast Cancer Eligible to Omit Radiation Therapy.
Rosenzweig SJ, Boe LA, Wen HY … +10 more , Mueller B, Roth O'Brien D, Naoum GE, Abou Yehia Z, Choi JI, Hahesy E, El-Tamer M, Powell SN, Khan AJ, Braunstein LZ
Int J Radiat Oncol Biol Phys · 2026 Mar · PMID 41895478 · Publisher ↗
PURPOSE: Radiation therapy (RT) following breast-conserving surgery reduces the risk of recurrence for early-stage breast cancer. A subset of patients with favorable clinical and pathologic characteristics can safely omi...
PURPOSE: Radiation therapy (RT) following breast-conserving surgery reduces the risk of recurrence for early-stage breast cancer. A subset of patients with favorable clinical and pathologic characteristics can safely omit RT with a modest increase in recurrences, yet without survival implications. We sought to evaluate whether molecular risk profiling among such patients could further stratify them into (1) a higher-risk subgroup that may be inappropriate for RT omission and (2) an exceedingly favorable subgroup that might more strongly consider RT omission. METHODS AND MATERIALS: Using a prospectively maintained institutional database, we evaluated patients aged ≥65 years who underwent breast-conserving surgery for early-stage, estrogen receptor-positive breast cancer (ie, patients putatively eligible for RT omission). Patients were stratified by clinicopathologic features and Oncotype DX Recurrence Score (RS; ≤25 vs >25) and were evaluated for oncologic outcomes and survival by receipt of RT. RESULTS: The overall cohort comprised 1587 patients: 92% had RS ≤25 (n = 1455) and 8% had RS >25 (n = 132). With a median follow-up of 74.4 months, the cumulative incidence of local recurrence was low throughout and did not significantly differ between those with RS ≤25 vs >25 (6-year rate 1.9% vs 0.8%; P = .5). Among those who did not undergo RT (n = 350), patients with RS >25 (n = 19) did not have a significantly higher incidence of local recurrence than patients with RS ≤25 (6-year rate 5.6% vs 4.2%; P = .5). CONCLUSIONS: Among patients ≥65 years of age who underwent breast-conserving surgery for early-stage estrogen receptor-positive breast cancer, Oncotype DX RS did not further refine recurrence risk estimates for RT decision-making beyond the findings of landmark RT omission trials. Although the number of patients with Oncotype DX RS >25 who omitted RT was limited in this study, this subgroup did not exhibit a significantly higher risk than their low-molecular-risk counterparts, suggesting that risk score need not necessarily disqualify select patients who seek to forgo adjuvant RT. Prospective analyses with larger sample sizes will further elucidate the role of molecular assays in guiding RT decision-making across risk strata.
Toward Optimal Fractionation Schemes in Nonuniform Radiation Therapy: Exploiting Helical Tomotherapy's Thread Effect for Gynecological Cancer Treatment.
Luo H, Yang M, Li J … +12 more , Qiu T, Hu W, Song Y, Wang T, Peng H, Yang X, Feng B, Chen L, Tan L, Liang B, Wang Y, Jin F
Int J Radiat Oncol Biol Phys · 2026 Mar · PMID 41887368 · Publisher ↗
PURPOSE: This study aimed to optimize dose fractionation for helical tomotherapy-based spatially fractionated radiation therapy (HT-SFRT) in gynecologic cancers by exploiting the "thread effect" to induce cohort effects....
PURPOSE: This study aimed to optimize dose fractionation for helical tomotherapy-based spatially fractionated radiation therapy (HT-SFRT) in gynecologic cancers by exploiting the "thread effect" to induce cohort effects. METHODS AND MATERIALS: HT-SFRT plans with varying parameters (pitch: 0.3-0.5; field width: 2.512 and 5.048 cm; modulation factor: 1.0-4.0) were applied to HeLa cells to identify the optimal configuration maximizing the peak-to-valley dose ratio. The cohort effect threshold was determined through quantification of DNA damage (γ-H2AX) and cell viability. Clinical validation involved 45 patients with gynecologic cancer, comparing 3 treatment approaches: conventional HT, stereotactic body radiation therapy, and SFRT (3 and 4 fractions). Comprehensive dosimetric analysis included evaluation of physical doses, biologically effective doses, tumor control probability, and normal tissue complication probability (NTCP) using the universal survival curve and Lyman-Kutcher-Burman models. RESULTS: Optimized HT-SFRT parameters (field width = 5.048 cm; pitch = 0.5; modulation factor = 2.0) achieved a peak-to-valley dose ratio of 1.15 at 10 cm off-axis. The 8 Gy/fraction dose threshold induced significant γ-H2AX expression and reduced cell viability by 15.41% (P < .05), establishing the cohort effect criterion. Clinically, the 32 Gy/4 fractions SFRT regimen demonstrated approximate tumor control (tumor control probability: 99.34% vs 100%) to conventional HT (50 Gy/25 fractions), whereas offering superior therapeutic advantages: (1) 84% reduction in treatment duration (4 vs 25 days); (2) significantly lower NTCP for bladder (6.5 × 10⁶ % vs 3.6 × 10⁴ %, P = .05); and (3) 42.9% faster fractional delivery than stereotactic body radiation therapy. Notably, SFRT exhibited elevated NTCP for rectum (3.99% vs 0.10%) and femoral head (0.16%-0.22% vs 2.9 × 10⁷ %-4.8 × 10⁷ %) structures. CONCLUSIONS: The thread effect in HT can be harnessed for SFRT, avoiding traditional mitigation strategies. The optimized HT-SFRT scheme (32 Gy/4 fractions) offers equivalent tumor control, shorter treatment duration, and acceptable normal tissue toxicity, providing a promising alternative for gynecologic cancer radiation therapy. Prospective clinical trials are warranted to validate these findings.
Assessment of Defacing Techniques on Medical Images for Radiation Therapy: Implications for Patient Privacy and Data Utility.
Wang D, Lee SH, Wang T … +1 more , Xiao Y
Int J Radiat Oncol Biol Phys · 2026 Mar · PMID 41876069 · Publisher ↗
PURPOSE: To assess the impact of defacing-based deidentification techniques on reidentification risk and data utility across multimodal imaging in radiation therapy. METHODS AND MATERIALS: We applied 4 defacing technique...
PURPOSE: To assess the impact of defacing-based deidentification techniques on reidentification risk and data utility across multimodal imaging in radiation therapy. METHODS AND MATERIALS: We applied 4 defacing techniques: biometric_mask, quickshear, mri_reface, and Carina's deidentifier, to imaging from 88 brain patients (magnetic resonance imaging, computed tomography [CT], and RTDose) and 97 head and neck patients (positron emission tomography, CT, and RTDose) in The Cancer Imaging Archive. Reidentification risk was assessed using ArcFace, a deep learning-based facial recognition model, by measuring cosine similarity scores and conducting receiver operating characteristic analysis to distinguish between original and defaced images. Data integrity was evaluated by statistically comparing the volume and image intensity changes between the original and defaced images across 9 critical organs and gross tumor volume. RESULTS: Quickshear provides the highest privacy protection, achieving the lowest area under the curve across imaging modalities (area under the curve, 0.61-0.74), followed by Carina (0.59-0.80). Mri_reface showed moderate protection (0.70-0.91), whereas biometric_mask offered the least (0.76-0.94). Carina preserved structure volumes, and mri_reface produced minor volumetric changes in the eyes (5%) and lens (9%). In contrast, biometric_mask substantially affected the mandible (39%) and oral cavity (69%), whereas quickshear significantly altered multiple structures (10.0%-86.6%). Median changes in mean CT intensity after defacing were -48.3% (interquartile range [IQR], -65.9% to -33.9%) with biometric_mask and -77.7% (IQR, -89.4% to -51.7%) with quickshear in the oral cavity. For eyes, Carina and mri_reface produced changes of +24.5% (IQR, 11.6% to 38.2%) and +54.9% (IQR, 16.2% to 82.3%), respectively. In the brain data set, biometric_mask and quickshear decreased oral cavity D by 0.72 Gy (IQR, 0.33 Gy to 0.89 Gy) and 1.61 Gy (IQR, 0.72 Gy to 2.12 Gy), respectively. Carina and mri_reface reduced eyes D by 2.11 Gy (IQR, 0.00 Gy to 3.39 Gy) and 1.05 Gy (IQR, 0.21 Gy to 1.16 Gy), respectively. A similar trend was observed in the head and neck data set with larger deviations. CONCLUSIONS: Carina's deidentifier and mri_reface showed favorable privacy-utility trade-offs relative to facial removal; the optimal choice may vary by application priorities.
The Quasi-Adiabatic Graphite Calorimeter for Absolute Dosimetry in Ultrahigh Dose Rate FLASH X-Ray Radiation Therapy.
Huang J, Zhang G, Wang K … +11 more , Tian B, Yang Y, Wu D, Jin S, Wang Z, Qi Y, Zhang H, Zhang H, Wu H, Yan X, Subiel A
Int J Radiat Oncol Biol Phys · 2026 Mar · PMID 41876068 · Publisher ↗
PURPOSE: FLASH radiation therapy using high-energy x rays combines ultrahigh dose rate irradiation with the physical characteristics of high-energy x-ray beams, achieving a significant reduction in normal tissue biologic...
PURPOSE: FLASH radiation therapy using high-energy x rays combines ultrahigh dose rate irradiation with the physical characteristics of high-energy x-ray beams, achieving a significant reduction in normal tissue biological damage while maintaining sufficient tissue penetration, thereby presenting great potential for clinical translation. However, the absence of a traceable absolute dosimetry method for FLASH x-ray beams remains a substantial limitation to its clinical implementation. This study aims to establish a quasi-adiabatic water-controlled probe-type graphite calorimeter for the absolute measurement of absorbed dose to water in the 10 MV FLASH x-ray beam, to address the current lack of a traceable dosimetry standard for x-ray FLASH radiation therapy. METHODS AND MATERIALS: A probe-type graphite calorimeter was developed, employing thermally stabilized water as the thermal control medium to precisely regulate the thermal equilibrium of the graphite core. This quasi-adiabatic system is designed to facilitate accurate absolute dose measurements under ultrahigh dose rate conditions. RESULTS: The results indicate that for a single irradiation with a total dose exceeding 2 Gy, the mean type A relative uncertainty, determined from 5 repeated measurements using the sample standard deviation, is less than 0.2%. By deriving the necessary correction factors for determining the absolute dose (ie, in Gy) of FLASH photon radiation therapy, the uncertainty in water absorbed dose measurement is determined to be 1.0% (k = 1). CONCLUSIONS: This study develops a probe-type graphite calorimeter for the absolute measurement of absorbed dose to water in 10 MV FLASH x-ray beams. The system is designed to address the current lack of a traceable dosimetric standard for x-ray FLASH radiation therapy, thereby supporting its clinical translation and application.
A Longitudinal Assessment of Remote IMRT/VMAT Training for Medical Physicists in Low- and Middle-Income Countries.
Silverwood SM, Carlson CM, Zhu H … +13 more , Biancia CD, Dempsey C, Shulman A, Keiper T, Koufigar S, Yorke AA, Jalal B, Yu C, Dimitriadou D, Wilson L, Nano T, Gillespie EF, Li B
Int J Radiat Oncol Biol Phys · 2026 Mar · PMID 41871640 · Publisher ↗
PURPOSE: Equitable access to high-quality cancer care depends on scalable, sustainable training initiatives, particularly in radiation oncology. This study evaluated whether gains in treatment planning competency, confid...
PURPOSE: Equitable access to high-quality cancer care depends on scalable, sustainable training initiatives, particularly in radiation oncology. This study evaluated whether gains in treatment planning competency, confidence, and knowledge among medical physicists were maintained 2 years after a virtual volumetric modulated arc therapy/intensity modulated radiation therapy (VMAT/IMRT) training course. METHODS AND MATERIALS: Medical physicists who completed a 15-week virtual training in 2022 were reassessed 2 years later via remote submission of a head and neck VMAT/IMRT plan, a multiple-choice knowledge test, and a confidence survey. Plan quality was assessed using automated scorecards (25 points) and expert rubrics (14 points), whereas knowledge and confidence were measured via surveys. Wilcoxon signed-rank tests compared pre, post, and follow-up outcomes. Effect sizes were calculated using Cohen's d. RESULTS: Nineteen of the 40 invited participants who had completed the prior postcourse assignment (47.5%) were enrolled in the follow-up study, representing 15 countries across four continents. Seventeen participants had complete data, defined as submitting pre, post, and follow-up responses for both objective and subjective assessments. Objective scores improved from 10.5/25 (±8.1) pretraining to 16.4/25 (±6.8) posttraining and 18.6/25 (±6.4) at follow-up (P = .011 and P = .001), with no significant change from posttraining to follow-up (P = .26). Subjective scores declined posttraining (11.1/14 ± 5.9 to 8.5/14 ± 3.0; P = .021) but rebounded at follow-up (10.8/14 ± 3.4; P = .037), with no difference from baseline (P = .96). Among 18 participants with complete survey data, confidence improved from 3.18 ± 1.26 at baseline to 4.14 ± 0.95 posttraining and was sustained at 4.12 ± 0.74 at follow-up, with a significant improvement from baseline to follow-up (P = .0023). Knowledge increased from 65.3% ± 29.9 at baseline to 83.3% ± 17.1 posttraining and was sustained at 80.6% ± 20.2 at follow-up, with a significant improvement from baseline to follow-up (P = .026). CONCLUSION: Remote VMAT/IMRT training led to lasting improvements in treatment planning, confidence, and knowledge, with objective scores continuing to rise over time. These findings support remote education as a sustainable model for building radiation therapy capacity in low- and middle-income countries.
About
- Frequency
- Sun
- Papers found
- 200
- RSS feed
- Subscribe