Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111293
Lone Hoffmann , Mai-Britt Linaa , Mai Lykkegaard Ehmsen , Maria Fuglsang Jensen , Sarah Eckholdt , Christina Larsen , Mikkel Drøgemüller Lund , Laura Patricia Kaplan , Morten Nielsen , Wiviann Ottosson , Cécile Peucelle , Arpit Saini , Hella Sand , Simon Nyberg Thomsen , Stine Fredslund , Torben Schjødt Hansen , Vladimira Horvat , Marianne Marquard Knap , Lotte Holm Land , Hanna Mortensen , Ane Appelt
Introduction
High-dose lung cancer reirradiation is promising but associated with high toxicity risk. Development of and adherence to consensus guidelines will support safe use.
Materials and methods
Literature review and live workshops were conducted to develop treatment planning guidelines in preparation for the Scandinavian CURE Lung trial. Relevant OARs, dose metrics, constraints, and priorities were considered. For six high-dose reirradiation lung cancer cases, the physical 3D dose distribution of previous treatment was mapped to current CT. The cases were distributed to eight radiotherapy centres, which optimised plans on current CT respecting equieffective cumulative dose constraints. Mapped previous dose and current dose were rescaled to EQD2Gy (α/β = 3 Gy, spinal cord: α/β = 2 Gy), summed by each centre, and reviewed centrally. After cases 1–4 were completed, prioritisation between OAR constraints and target coverage was clarified.
Results
Consensus agreement on guidelines for treatment planning, equieffective cumulative dose constraints, and priorities was established. For cases 1–3, centres complied with constraints. For case 4, covering PTV while respecting OAR constraints was difficult, and major variations were identified, underscoring the need for clearer prioritisation guidance. Consensus was reached to prioritise OAR constraints, and centres re-optimised case 4 accordingly. For cases 5 and 6, all centres underdosed PTV, to comply with OAR constraints. Generally, considerable variation in cumulative OAR doses was observed.
Conclusions
A treatment planning protocol was developed. The pre-trial multi-centre treatment planning study identified and resolved key missing guidance to facilitate common conception of consensus guidelines. Feasibility and compliance with the proposed equieffective cumulative dose constraints were established.
{"title":"Development of a treatment planning protocol for the multi-centre reirradiation CURE Lung trial","authors":"Lone Hoffmann , Mai-Britt Linaa , Mai Lykkegaard Ehmsen , Maria Fuglsang Jensen , Sarah Eckholdt , Christina Larsen , Mikkel Drøgemüller Lund , Laura Patricia Kaplan , Morten Nielsen , Wiviann Ottosson , Cécile Peucelle , Arpit Saini , Hella Sand , Simon Nyberg Thomsen , Stine Fredslund , Torben Schjødt Hansen , Vladimira Horvat , Marianne Marquard Knap , Lotte Holm Land , Hanna Mortensen , Ane Appelt","doi":"10.1016/j.radonc.2025.111293","DOIUrl":"10.1016/j.radonc.2025.111293","url":null,"abstract":"<div><h3>Introduction</h3><div>High-dose lung cancer reirradiation is promising but associated with high toxicity risk. Development of and adherence to consensus guidelines will support safe use.</div></div><div><h3>Materials and methods</h3><div>Literature review and live workshops were conducted to develop treatment planning guidelines in preparation for the Scandinavian CURE Lung trial. Relevant OARs, dose metrics, constraints, and priorities were considered. For six high-dose reirradiation lung cancer cases, the physical 3D dose distribution of previous treatment was mapped to current CT. The cases were distributed to eight radiotherapy centres, which optimised plans on current CT respecting equieffective cumulative dose constraints. Mapped previous dose and current dose were rescaled to EQD2Gy (α/β = 3 Gy, spinal cord: α/β = 2 Gy), summed by each centre, and reviewed centrally. After cases 1–4 were completed, prioritisation between OAR constraints and target coverage was clarified.</div></div><div><h3>Results</h3><div>Consensus agreement on guidelines for treatment planning, equieffective cumulative dose constraints, and priorities was established. For cases 1–3, centres complied with constraints. For case 4, covering PTV while respecting OAR constraints was difficult, and major variations were identified, underscoring the need for clearer prioritisation guidance. Consensus was reached to prioritise OAR constraints, and centres re-optimised case 4 accordingly. For cases 5 and 6, all centres underdosed PTV, to comply with OAR constraints. Generally, considerable variation in cumulative OAR doses was observed.</div></div><div><h3>Conclusions</h3><div>A treatment planning protocol was developed. The pre-trial multi-centre treatment planning study identified and resolved key missing guidance to facilitate common conception of consensus guidelines. Feasibility and compliance with the proposed equieffective cumulative dose constraints were established.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111293"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145574210","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111291
Longfei Diao , Xingyi Zhao , Chingyun Cheng , Tengda Zhang , Shouyi Wei , Dongdong Meng , Zhizhen Wei , Yangguang Ma , Kun Zhu , Hui Wu , Benjamin Durkee , Haibo Lin , Charles B. Simone , Xueqing Yan , Minglei Kang
Purpose
To evaluate the impact of key treatment planning parameters—including beam arrangement, minimum monitor unit (MMU) settings, and anatomical site variability—on the ability to achieve ultra-high dose rate (UHDR) delivery in Bragg peak FLASH radiotherapy.
Methods
FLASH dose rate coverage can be assessed by dose-rate volume histogram (DRVH), thus objective functions based on DRVH can be constructed to optimize the dose rate distribution for individual regions of interest (ROIs). The optimization of each ROI, as part of the final objective function, is integrated into a multi-objective optimization problem that can be solved using a heuristic algorithm. A phantom-based study was conducted to investigate the effect of beam number on optimization performance in proton therapy planning. Treatment plans of 8 consecutive node-negative non-small cell lung cancer and 5 consecutive liver cancer patients were initially optimized, followed by optimizing the spot delivery sequence to enhance dose rate ratios without compromising dose performance. A thorough evaluation was conducted to assess the optimization of the scanning pattern in improving the FLASH ratio of critical OARs in Bragg peak FLASH-RT, considering beam currents, beam arrangement, MMU constraints, and anatomical sites in lung and liver cases.
Results
The phantom study demonstrated that the effectiveness of the spot pattern in dose rate depends on the number of beams and beam arrangement, and the 3-field arrangement can achieve better optimization effects. In lung cases, using a MMU of 600 (nozzle current of 252nA), scanning pattern optimization increased the average dose rate (V40Gy/s) for the esophagus, heart, spinal cord, and lung-GTV from 38.3 %, 62.8 %, 59.6 % and 61.9 % to 74.4 %, 85.5 %, 83.3 % and 78.6 %, respectively (all p-values < 0.001). When a higher MMU of 1200 (nozzle current of 504nA) was used, the benefits brought by optimization are not as obvious as the previous situation. For all liver cases with an MMU of 600, the average FLASH dose rate (V40Gy/s) for the esophagus, heart, spinal cord, and liver-GTV increased from 60.5 %, 52.7 %, 60.3 %, and 59.1 % to 75.1 %, 69.4 %, 80.2 %, and 75.9 %, respectively, after optimization (all p-values < 0.001). However, when a higher MMU of 1200 was used, the V40Gy/s for all four OARs increased from approximately 93.3 % to 97.0 %, showing only limited additional improvement.
Conclusion
This approach successfully optimized FLASH dose rate coverage for specific OARs, enhancing BP-FLASH effectiveness by improving OAR protection while maintaining dosimetric quality. However, the impact of spot pattern optimization is influenced by factors such as the number of beams, MMU constraints, and spot distribution, with limited effectiveness in significantly increasing the FLASH ratio.
{"title":"Assessing the potential and pitfalls of spot sequence optimization for OAR-specific dose rate control in proton PBS Bragg peak FLASH radiotherapy","authors":"Longfei Diao , Xingyi Zhao , Chingyun Cheng , Tengda Zhang , Shouyi Wei , Dongdong Meng , Zhizhen Wei , Yangguang Ma , Kun Zhu , Hui Wu , Benjamin Durkee , Haibo Lin , Charles B. Simone , Xueqing Yan , Minglei Kang","doi":"10.1016/j.radonc.2025.111291","DOIUrl":"10.1016/j.radonc.2025.111291","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the impact of key treatment planning parameters—including beam arrangement, minimum monitor unit (MMU) settings, and anatomical site variability—on the ability to achieve ultra-high dose rate (UHDR) delivery in Bragg peak FLASH radiotherapy.</div></div><div><h3>Methods</h3><div>FLASH dose rate coverage can be assessed by dose-rate volume histogram (DRVH), thus objective functions based on DRVH can be constructed to optimize the dose rate distribution for individual regions of interest (ROIs). The optimization of each ROI, as part of the final objective function, is integrated into a multi-objective optimization problem that can be solved using a heuristic algorithm. A phantom-based study was conducted to investigate the effect of beam number on optimization performance in proton therapy planning. Treatment plans of 8 consecutive node-negative non-small cell lung cancer and 5 consecutive liver cancer patients were initially optimized, followed by optimizing the spot delivery sequence to enhance dose rate ratios without compromising dose performance. A thorough evaluation was conducted to assess the optimization of the scanning pattern in improving the FLASH ratio of critical OARs in Bragg peak FLASH-RT, considering beam currents, beam arrangement, MMU constraints, and anatomical sites in lung and liver cases.</div></div><div><h3>Results</h3><div>The phantom study demonstrated that the effectiveness of the spot pattern in dose rate depends on the number of beams and beam arrangement, and the 3-field arrangement can achieve better optimization effects. In lung cases, using a MMU of 600 (nozzle current of 252nA), scanning pattern optimization increased the average dose rate (V40Gy/s) for the esophagus, heart, spinal cord, and lung-GTV from 38.3 %, 62.8 %, 59.6 % and 61.9 % to 74.4 %, 85.5 %, 83.3 % and 78.6 %, respectively (all p-values < 0.001). When a higher MMU of 1200 (nozzle current of 504nA) was used, the benefits brought by optimization are not as obvious as the previous situation. For all liver cases with an MMU of 600, the average FLASH dose rate (V40Gy/s) for the esophagus, heart, spinal cord, and liver-GTV increased from 60.5 %, 52.7 %, 60.3 %, and 59.1 % to 75.1 %, 69.4 %, 80.2 %, and 75.9 %, respectively, after optimization (all p-values < 0.001). However, when a higher MMU of 1200 was used, the V40Gy/s for all four OARs increased from approximately 93.3 % to 97.0 %, showing only limited additional improvement.</div></div><div><h3>Conclusion</h3><div>This approach successfully optimized FLASH dose rate coverage for specific OARs, enhancing BP-FLASH effectiveness by improving OAR protection while maintaining dosimetric quality. However, the impact of spot pattern optimization is influenced by factors such as the number of beams, MMU constraints, and spot distribution, with limited effectiveness in significantly increasing the FLASH ratio.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111291"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145574168","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111297
Junxian Li , Renhe Liu , Yuchen Xing , Ximin Gao , Qiang Yin , Qian Su
Objectives
This study aims to develop a self-supervised foundational model based on routine MRI and to evaluate its performance in tasks of brain tumor grading and pathological subtype classification.
Methods
We developed the Unified Multimodal Brain Imaging Foundation (UMBIF) model by performing self-supervised learning on 51,029 MRI images from multiple institutions. The model was first trained using a contrastive masked image modeling task to extract robust feature representations. Next, UMBIF was fine-tuned for downstream tasks—glioma grading and histological classification—using multi-center cohorts. Finally, we compared UMBIF model with mainstream convolutional neural networks and machine learning algorithms, evaluating accuracy, sensitivity, specificity, and area under the curve (AUC).
Results
Compared to self-supervised pretraining methods applied to natural images or single large tumor region images, the UMBIF architecture effectively extracted more comprehensive feature representations, leading to superior model performance. The optimal classifier with pretrained weights demonstrated outstanding results on independent test datasets, achieving accuracies of 0.840 (AUC: 0.723) for grade II, 0.684 (AUC: 0.854) for grade III, 0.775 (AUC: 0.743) for grade IV gliomas and 0.903 (AUC: 0.966) for histological classification, respectively, highlighting its potential in clinical decision-making.
Conclusions
The UMBIF model demonstrated robust applicability across clinically relevant glioma-grading formulations and LGG/HGG subtype classification. By enhancing classification performance with pretrained weights and reducing reliance on annotated data, it holds strong clinical potential for improving diagnostic efficiency and decision-making.
{"title":"A foundation model for brain tumor MRI analysis: WHO grading and subtype classification","authors":"Junxian Li , Renhe Liu , Yuchen Xing , Ximin Gao , Qiang Yin , Qian Su","doi":"10.1016/j.radonc.2025.111297","DOIUrl":"10.1016/j.radonc.2025.111297","url":null,"abstract":"<div><h3>Objectives</h3><div>This study aims to develop a self-supervised foundational model based on routine MRI and to evaluate its performance in tasks of brain tumor grading and pathological subtype classification.</div></div><div><h3>Methods</h3><div>We developed the Unified Multimodal Brain Imaging Foundation (UMBIF) model by performing self-supervised learning on 51,029 MRI images from multiple institutions. The model was first trained using a contrastive masked image modeling task to extract robust feature representations. Next, UMBIF was fine-tuned for downstream tasks—glioma grading and histological classification—using multi-center cohorts. Finally, we compared UMBIF model with mainstream convolutional neural networks and machine learning algorithms, evaluating accuracy, sensitivity, specificity, and area under the curve (AUC).</div></div><div><h3>Results</h3><div>Compared to self-supervised pretraining methods applied to natural images or single large tumor region images, the UMBIF architecture effectively extracted more comprehensive feature representations, leading to superior model performance. The optimal classifier with pretrained weights demonstrated outstanding results on independent test datasets, achieving accuracies of 0.840 (AUC: 0.723) for grade II, 0.684 (AUC: 0.854) for grade III, 0.775 (AUC: 0.743) for grade IV gliomas and 0.903 (AUC: 0.966) for histological classification, respectively, highlighting its potential in clinical decision-making.</div></div><div><h3>Conclusions</h3><div>The UMBIF model demonstrated robust applicability across clinically relevant glioma-grading formulations and LGG/HGG subtype classification. By enhancing classification performance with pretrained weights and reducing reliance on annotated data, it holds strong clinical potential for improving diagnostic efficiency and decision-making.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111297"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145574217","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111295
Tae Hyun Kim , Jung Won Chun , Sang Myung Woo , Joo-Hyun Chung , Min Hee Lee , Sung-Sik Han , Sang-Jae Park , Sung Uk Lee , Yang-Gun Suh , Sung Ho Moon , Sang Soo Kim , Woo Jin Lee
Purpose
To evaluate the efficacy and safety of hypofractionated simultaneous integrated boost (SIB) − proton beam therapy (PBT) with systemic treatment and also to explore survival-related prognostic factors, including treatment sequencing and dose–response.
Methods
We reviewed 225 consecutive patients with LAPC treated with SIB-PBT (45–50 Gy[RBE] for PTV1 and 30 Gy[RBE] for PTV2 in 10 fractions). The patients were grouped by treatment sequence as follows: PBT after stable disease post-induction chemotherapy (Group I, 48.9 %), PBT after progression post-induction chemotherapy (Group II, 15.6 %), and PBT followed by maintenance chemotherapy (Group III, 35.6 %).
Results
Median overall survival (OS) from first treatment and PBT was 24.2 (95 % CI, 21.7–26.7) and 19.5 (95 % CI, 17.4–21.7) months, respectively. Group I had significantly longer OS than Groups II and III from first treatment (31.1 vs. 21.1 and 19.1 months) and PBT (23.7 vs. 9.9 and 19.1 months) (p < 0.05 each). Patients receiving 50 Gy(RBE) had significantly better OS from first treatment (25.6 vs. 19.3 months; p = 0.001) and a trend towards better OS from PBT (20.3 vs. 18.3 months; p = 0.062). Most acute adverse events were grade 1–2, with late grade ≥ 3 gastrointestinal events in 4 %.
Conclusion
This largest to-date study on PBT for LAPC suggests that hypofractionated SIB-PBT could be a safe and effective local treatment for LAPC. Favorable outcomes were observed with PBT after stable disease following induction chemotherapy, and a potential dose–response relationship was suggested. However, further prospective studies are required to validate our findings.
{"title":"Impact of treatment sequence and dose response on outcomes in locally advanced pancreatic cancer treated with hypofractionated proton beam therapy using simultaneous integrated boost technique","authors":"Tae Hyun Kim , Jung Won Chun , Sang Myung Woo , Joo-Hyun Chung , Min Hee Lee , Sung-Sik Han , Sang-Jae Park , Sung Uk Lee , Yang-Gun Suh , Sung Ho Moon , Sang Soo Kim , Woo Jin Lee","doi":"10.1016/j.radonc.2025.111295","DOIUrl":"10.1016/j.radonc.2025.111295","url":null,"abstract":"<div><h3>Purpose</h3><div>To evaluate the efficacy and safety of hypofractionated simultaneous integrated boost (SIB) − proton beam therapy (PBT) with systemic treatment and also to explore survival-related prognostic factors, including treatment sequencing and dose–response.</div></div><div><h3>Methods</h3><div>We reviewed 225 consecutive patients with LAPC treated with SIB-PBT (45–50 Gy[RBE] for PTV1 and 30 Gy[RBE] for PTV2 in 10 fractions). The patients were grouped by treatment sequence as follows: PBT after stable disease post-induction chemotherapy (Group I, 48.9 %), PBT after progression post-induction chemotherapy (Group II, 15.6 %), and PBT followed by maintenance chemotherapy (Group III, 35.6 %).</div></div><div><h3>Results</h3><div>Median overall survival (OS) from first treatment and PBT was 24.2 (95 % CI, 21.7–26.7) and 19.5 (95 % CI, 17.4–21.7) months, respectively. Group I had significantly longer OS than Groups II and III from first treatment (31.1 vs. 21.1 and 19.1 months) and PBT (23.7 vs. 9.9 and 19.1 months) (<em>p</em> < 0.05 each). Patients receiving 50 Gy(RBE) had significantly better OS from first treatment (25.6 vs. 19.3 months; <em>p</em> = 0.001) and a trend towards better OS from PBT (20.3 vs. 18.3 months; <em>p</em> = 0.062). Most acute adverse events were grade 1–2, with late grade ≥ 3 gastrointestinal events in 4 %.</div></div><div><h3>Conclusion</h3><div>This largest to-date study on PBT for LAPC suggests that hypofractionated SIB-PBT could be a safe and effective local treatment for LAPC. Favorable outcomes were observed with PBT after stable disease following induction chemotherapy, and a potential dose–response relationship was suggested. However, further prospective studies are required to validate our findings.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111295"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145574237","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111286
Mette Skovhus Thomsen , Birgitte Vrou Offersen , on behalf of the authors
{"title":"Response to comment on “Quality assessment of 2705 treatment plans in the randomised Danish breast cancer Group Skagen trial 1”","authors":"Mette Skovhus Thomsen , Birgitte Vrou Offersen , on behalf of the authors","doi":"10.1016/j.radonc.2025.111286","DOIUrl":"10.1016/j.radonc.2025.111286","url":null,"abstract":"","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111286"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145564813","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111292
Martin J. Higgins , Philippa Johnstone , H. Miles Prince , Stephen Lade , Michael T. Fahey , Christopher McCormack , Carrie van der Weyden , Friyana Bhabha , Odette Buelens , Belinda A. Campbell
Background
Mycosis fungoides (MF) is the most prevalent subtype of primary cutaneous T-cell lymphoma. Large cell transformation of MF (LCTMF) is rare and confers a poor prognosis. Radiotherapy (RT) is an effective local treatment for LCTMF; however, dose–response data are limited.
Methods
Eligibility for this retrospective study required biopsy-proven LCTMF with clinico-pathological correlation, diagnosed 1/1/1990–1/10/2021, and managed at Peter MacCallum Cancer Centre.
Results
83 patients were eligible. Median age was 68 years, 63 (76 %) patients had cutaneous-only LCTMF at time of diagnosis. Median follow-up was 8.0 (95 % CI: 6–11) years.
Details of 155 irradiated LCTMF lesions (from 49 patients) were available: 150 cutaneous, 5 extra-cutaneous. For cutaneous LCTMF, median equivalent dose in 2.0 Gy per fraction (EQD2, α/β of 10) was 30.6 (range, 4.7–46.3) Gy. Dose-response data were available for 141 cutaneous LCTMF lesions. Overall response rate (ORR) was 94 % (60 % complete response (CR), 35 % partial response). For cutaneous lesions treated with >12.0 Gy, ORR was 100 %. Increasing doses of RT were associated with greater CR rates: doses >36.0 Gy achieved 100 % CR rate.
22 (27 %) patients had unifocal cutaneous-only LCTMF and were treated with local RT-alone, median EQD2 36.0 Gy (range, 17.3–46.3 Gy). CR rate was 100 %. 9 (41 %) patients remained relapse-free (median follow-up, 3.2 years). Only 1 patient experienced infield-only recurrence at first relapse.
Conclusion
LCTMF is radio-responsive, with EQD2 >12.0 Gy associated with 100 % ORR. Dose-response was observed, with EQD2 >36.0 Gy achieving 100 % CR rate. For unifocal cutaneous LCTMF, RT-alone achieved excellent infield control with possible curative potential in a proportion of patients.
{"title":"Radiotherapy dose–response of mycosis fungoides with large cell transformation","authors":"Martin J. Higgins , Philippa Johnstone , H. Miles Prince , Stephen Lade , Michael T. Fahey , Christopher McCormack , Carrie van der Weyden , Friyana Bhabha , Odette Buelens , Belinda A. Campbell","doi":"10.1016/j.radonc.2025.111292","DOIUrl":"10.1016/j.radonc.2025.111292","url":null,"abstract":"<div><h3>Background</h3><div>Mycosis fungoides (MF) is the most prevalent subtype of primary cutaneous T-cell lymphoma. Large cell transformation of MF (LCTMF) is rare and confers a poor prognosis. Radiotherapy (RT) is an effective local treatment for LCTMF; however, dose–response data are limited.</div></div><div><h3>Methods</h3><div>Eligibility for this retrospective study required biopsy-proven LCTMF with clinico-pathological correlation, diagnosed 1/1/1990–1/10/2021, and managed at Peter MacCallum Cancer Centre.</div></div><div><h3>Results</h3><div>83 patients were eligible. Median age was 68 years, 63 (76 %) patients had cutaneous-only LCTMF at time of diagnosis. Median follow-up was 8.0 (95 % CI: 6–11) years.</div><div>Details of 155 irradiated LCTMF lesions (from 49 patients) were available: 150 cutaneous, 5 extra-cutaneous. For cutaneous LCTMF, median equivalent dose in 2.0 Gy per fraction (EQD2, α/β of 10) was 30.6 (range, 4.7–46.3) Gy. Dose-response data were available for 141 cutaneous LCTMF lesions. Overall response rate (ORR) was 94 % (60 % complete response (CR), 35 % partial response). For cutaneous lesions treated with >12.0 Gy, ORR was 100 %. Increasing doses of RT were associated with greater CR rates: doses >36.0 Gy achieved 100 % CR rate.</div><div>22 (27 %) patients had unifocal cutaneous-only LCTMF and were treated with local RT-alone, median EQD2 36.0 Gy (range, 17.3–46.3 Gy). CR rate was 100 %. 9 (41 %) patients remained relapse-free (median follow-up, 3.2 years). Only 1 patient experienced infield-only recurrence at first relapse.</div></div><div><h3>Conclusion</h3><div>LCTMF is radio-responsive, with EQD2 >12.0 Gy associated with 100 % ORR. Dose-response was observed, with EQD2 >36.0 Gy achieving 100 % CR rate. For unifocal cutaneous LCTMF, RT-alone achieved excellent infield control with possible curative potential in a proportion of patients.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"215 ","pages":"Article 111292"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145573678","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111290
A. Bazani , M. Bagnalasta , G. Fontana , S. Lillo , V. Pavanello , L. Trombetta , E. Rossi , S. Molinelli , A. Iannalfi , M. Ciocca , E. Orlandi
Background and purpose
To investigate the combined effect of RBE-weighted dose (DRBE) and dose-averaged linear energy transfer (LETd) on the development of brain radionecrosis (BRN) in patients with intracranial meningioma and solitary fibrous tumors treated with intensity modulated proton therapy (IMPT).
Materials and methods
The study included 124 patients with a median follow-up of 47.4 months (range 5.8 ÷ 109.6). Dose prescriptions ranged between 54 and 66 Gy(RBE) (RBE = 1.1) in 28 ÷ 33 fractions. Patients reporting a BRN of grade ≥ 1 (CTCAE v5.0) at follow-up constituted the adverse event (AE) group (n = 26). The remaining 98 patients constituted the control group. The BRN volume was contoured on the follow-up MRI images that first revealed the presence of the AE. We extracted both DRBE and LETd distributions to build the cumulative dose LETd volume histogram (DLVH) of the healthy-brain. Univariable and multivariable logistic regressions were fitted to explore the impact of DLVH indexes on BRN (α = 0.01). Voxel-wise analysis was performed on the AE group using greedy random matching (caliper = 0.1) to match healthy-brain with AE voxels with a 5 Gy(RBE) dose-binning. A logistic mixed-effects model (random intercept) was fitted for each bin (α = 0.05).
Results
The best fitting DLVH index was selected as the median partial brain volume receiving a DRBE > 42.9 Gy(RBE) and LETd > 4.6 keV/µm (8.2 cc for the AE group and 1.1 cc for the control group). The effect of LETd reached the highest statistically significant impact in the dose bin 40–45 Gy(RBE) (OR: 2.24 [1.62; 3.20], q(False Discovery Rate-adjusted) = 0.01).
Conclusion
We identified a DLVH constraint, predictive of BRN, to be implemented in clinical practice aiming at a reduction of treatment toxicity.
{"title":"Investigation of the combined role of dose and dose-averaged LET in the occurrence of brain necrosis after intensity modulated proton therapy for meningioma and solitary fibrous tumours","authors":"A. Bazani , M. Bagnalasta , G. Fontana , S. Lillo , V. Pavanello , L. Trombetta , E. Rossi , S. Molinelli , A. Iannalfi , M. Ciocca , E. Orlandi","doi":"10.1016/j.radonc.2025.111290","DOIUrl":"10.1016/j.radonc.2025.111290","url":null,"abstract":"<div><h3>Background and purpose</h3><div>To investigate the combined effect of RBE-weighted dose (D<sub>RBE</sub>) and dose-averaged linear energy transfer (LET<sub>d</sub>) on the development of brain radionecrosis (BRN) in patients with intracranial meningioma and solitary fibrous tumors treated with intensity modulated proton therapy (IMPT).</div></div><div><h3>Materials and methods</h3><div>The study included 124 patients with a median follow-up of 47.4 months (range 5.8 ÷ 109.6). Dose prescriptions ranged between 54 and 66 Gy(RBE) (RBE = 1.1) in 28 ÷ 33 fractions. Patients reporting a BRN of grade ≥ 1 (CTCAE v5.0) at follow-up constituted the adverse event (AE) group (n = 26). The remaining 98 patients constituted the control group. The BRN volume was contoured on the follow-up MRI images that first revealed the presence of the AE. We extracted both D<sub>RBE</sub> and LET<sub>d</sub> distributions to build the cumulative dose LET<sub>d</sub> volume histogram (DLVH) of the healthy-brain. Univariable and multivariable logistic regressions were fitted to explore the impact of DLVH indexes on BRN (α = 0.01). Voxel-wise analysis was performed on the AE group using greedy random matching (caliper = 0.1) to match healthy-brain with AE voxels with a 5 Gy(RBE) dose-binning. A logistic mixed-effects model (random intercept) was fitted for each bin (α = 0.05).</div></div><div><h3>Results</h3><div>The best fitting DLVH index was selected as the median partial brain volume receiving a D<sub>RBE</sub> > 42.9 Gy(RBE) and LET<sub>d</sub> > 4.6 keV/µm (8.2 cc for the AE group and 1.1 cc for the control group). The effect of LET<sub>d</sub> reached the highest statistically significant impact in the dose bin 40–45 Gy(RBE) (OR: 2.24 [1.62; 3.20], q(False Discovery Rate-adjusted) = 0.01).</div></div><div><h3>Conclusion</h3><div>We identified a DLVH constraint, predictive of BRN, to be implemented in clinical practice aiming at a reduction of treatment toxicity.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111290"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145573568","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111289
Yuting Liu , Pieter Populaire , Camille Draguet , Karin Haustermans , Gilles Defraene
Purpose/Objective
To develop a general lung dose–response curve for postoperative pulmonary complications (PPCs) following neoadjuvant chemoradiotherapy (nCRT) and surgery in esophageal cancer (EC), and to assess the impact of including proton therapy (PT) data in normal tissue complication probability (NTCP) model development.
Materials and methods
A comprehensive PubMed search (2004–2024) identified studies reporting mean lung dose (MLD) and PPCs after esophagectomy with nCRT. Logistic and probit NTCP models were fitted to study-level or dose-bin data using weighted maximum likelihood optimization, with and without PT data (XT + PT and XT model, respectively). Meta-analysis evaluated other PPCs risk factors. A simulation study on 40 patients with comparative XT and PT plans assessed model-based patient selection, assuming PT referral when Δ NTCP ≥ 10 %.
Results
16 non-overlapping studies were included for review. PPCs incidence was significantly lower in PT versus XT cohort (15.6 % versus 30.7 %, p < 0.001). Both NTCP models showed increasing PPCs risk with rising MLD, with the XT model predicting higher PPCs rate in the low-dose region. Meta-analysis identified histology, surgery approach and pre-RT FEV1 as significant predictors (p < 0.05). In the simulation study, the XT + PT model predicted lower PPCs risk and subsequently higher Δ NTCP than the XT model (average 16.3 % versus 8.0 %, p < 0.001). Referral decisions differed in 15 of 40 patients (37.5 %), who were redirected to PT only by the XT + PT model.
Conclusion
Including PT dose–response data into thoracic NTCP models alters pulmonary complication model-based patient selection in EC.
{"title":"Integrating photon with proton dose–response data alters a pulmonary complication model-based patient selection in esophageal cancer","authors":"Yuting Liu , Pieter Populaire , Camille Draguet , Karin Haustermans , Gilles Defraene","doi":"10.1016/j.radonc.2025.111289","DOIUrl":"10.1016/j.radonc.2025.111289","url":null,"abstract":"<div><h3>Purpose/Objective</h3><div>To develop a general lung dose–response curve for postoperative pulmonary complications (PPCs) following neoadjuvant chemoradiotherapy (nCRT) and surgery in esophageal cancer (EC), and to assess the impact of including proton therapy (PT) data in normal tissue complication probability (NTCP) model development.</div></div><div><h3>Materials and methods</h3><div>A comprehensive PubMed search (2004–2024) identified studies reporting mean lung dose (MLD) and PPCs after esophagectomy with nCRT. Logistic and probit NTCP models were fitted to study-level or dose-bin data using weighted maximum likelihood optimization, with and without PT data (XT + PT and XT model, respectively). Meta-analysis evaluated other PPCs risk factors. A simulation study on 40 patients with comparative XT and PT plans assessed model-based patient selection, assuming PT referral when Δ NTCP ≥ 10 %.</div></div><div><h3>Results</h3><div>16 non-overlapping studies were included for review. PPCs incidence was significantly lower in PT versus XT cohort (15.6 % versus 30.7 %, p < 0.001). Both NTCP models showed increasing PPCs risk with rising MLD, with the XT model predicting higher PPCs rate in the low-dose region. Meta-analysis identified histology, surgery approach and pre-RT FEV<sub>1</sub> as significant predictors (p < 0.05). In the simulation study, the XT + PT model predicted lower PPCs risk and subsequently higher Δ NTCP than the XT model (average 16.3 % versus 8.0 %, p < 0.001). Referral decisions differed in 15 of 40 patients (37.5 %), who were redirected to PT only by the XT + PT model.</div></div><div><h3>Conclusion</h3><div>Including PT dose–response data into thoracic NTCP models alters pulmonary complication model-based patient selection in EC.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111289"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145573595","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-19DOI: 10.1016/j.radonc.2025.111296
Nicholas Summerfield , Qisheng He , Alex Kuo , Chase Ruff , Joshua Pan , Ahmed I. Ghanem , Simeng Zhu , Anudeep Kumar , Prashant Nagpal , Jiwei Zhao , Ming Dong , Carri Glide-Hurst
Purpose
Cardiac substructure delineation is emerging in treatment planning to minimize the risk of radiation-induced heart disease. Deep learning offers efficient methods to reduce contouring burden but currently lacks generalizability across different modalities and overlapping structures. This work introduces and validates a Modality-AGnostic Image Cascade (MAGIC) deep-learning pipeline for comprehensive and multi-modal cardiac substructure segmentation.
Materials and methods
MAGIC is implemented through replicated encoding and decoding branches of an nnU-Net backbone to handle multi-modality inputs and overlapping labels. First benchmarked on the multi-modality whole-heart segmentation (MMWHS) dataset including cardiac CT-angiography (CCTA) and MR modalities, twenty cardiac substructures (heart, chambers, great vessels (GVs), valves, coronary arteries (CAs), and conduction nodes) from clinical simulation CT (Sim-CT), low-field MR-Linac, and cardiac CT-angiography (CCTA) modalities were delineated to train semi-supervised (n = 151), validate (n = 15), and test (n = 30) MAGIC. For comparison, fourteen single-modality comparison models (two MMWHS modalities and four subgroups across three clinical modalities) were trained. Methods were evaluated for efficiency and against reference contours through the Dice similarity coefficient (DSC) and two-tailed Wilcoxon Signed-Rank test (p < 0.05).
Results
Average MMWHS DSC scores across CCTA and MR inputs were 0.88 ± 0.08 and 0.87 ± 0.04 respectively with significant improvement (p < 0.05) over unimodal baselines. Average 20-structure DSC scores were 0.75 ± 0.16 (heart, 0.96 ± 0.01; chambers, 0.89 ± 0.05; GVs, 0.81 ± 0.09; CAs, 0.60 ± 0.13; valves, 0.70 ± 0.18; nodes, 0.66 ± 0.12) for Sim-CT, 0.68 ± 0.21 (heart, 0.94 ± 0.01; chambers, 0.87 ± 0.05; GVs, 0.72 ± 0.18; CAs, 0.50 ± 0.18; valves, 0.62 ± 0.16; nodes, 0.52 ± 0.16) for MR-Linac, and 0.80 ± 0.16 (heart, 0.95 ± 0.01; chambers, 0.93 ± 0.04; GVs, 0.84 ± 0.06; CAs, 0.77 ± 0.12; valves, 0.68 ± 0.23; nodes, 0.72 ± 0.11) for CCTA. Furthermore, > 80% and > 70% reductions in training time and parameters were achieved, respectively.
Conclusions
MAGIC offers an efficient, lightweight solution capable of segmenting multiple image modalities and overlapping structures in a single model without compromising segmentation accuracy.
{"title":"Modality-AGnostic image Cascade (MAGIC) for multi-modality cardiac substructure segmentation","authors":"Nicholas Summerfield , Qisheng He , Alex Kuo , Chase Ruff , Joshua Pan , Ahmed I. Ghanem , Simeng Zhu , Anudeep Kumar , Prashant Nagpal , Jiwei Zhao , Ming Dong , Carri Glide-Hurst","doi":"10.1016/j.radonc.2025.111296","DOIUrl":"10.1016/j.radonc.2025.111296","url":null,"abstract":"<div><h3>Purpose</h3><div>Cardiac substructure delineation is emerging in treatment planning to minimize the risk of radiation-induced heart disease. Deep learning offers efficient methods to reduce contouring burden but currently lacks generalizability across different modalities and overlapping structures. This work introduces and validates a Modality-AGnostic Image Cascade (MAGIC) deep-learning pipeline for comprehensive and multi-modal cardiac substructure segmentation.</div></div><div><h3>Materials and methods</h3><div>MAGIC is implemented through replicated encoding and decoding branches of an nnU-Net backbone to handle multi-modality inputs and overlapping labels. First benchmarked on the multi-modality whole-heart segmentation (MMWHS) dataset including cardiac CT-angiography (CCTA) and MR modalities, twenty cardiac substructures (heart, chambers, great vessels (GVs), valves, coronary arteries (CAs), and conduction nodes) from clinical simulation CT (Sim-CT), low-field MR-Linac, and cardiac CT-angiography (CCTA) modalities were delineated to train semi-supervised (n = 151), validate (n = 15), and test (n = 30) MAGIC. For comparison, fourteen single-modality comparison models (two MMWHS modalities and four subgroups across three clinical modalities) were trained. Methods were evaluated for efficiency and against reference contours through the Dice similarity coefficient (DSC) and two-tailed Wilcoxon Signed-Rank test (p < 0.05).</div></div><div><h3>Results</h3><div>Average MMWHS DSC scores across CCTA and MR inputs were 0.88 ± 0.08 and 0.87 ± 0.04 respectively with significant improvement (p < 0.05) over unimodal baselines. Average 20-structure DSC scores were 0.75 ± 0.16 (heart, 0.96 ± 0.01; chambers, 0.89 ± 0.05; GVs, 0.81 ± 0.09; CAs, 0.60 ± 0.13; valves, 0.70 ± 0.18; nodes, 0.66 ± 0.12) for Sim-CT, 0.68 ± 0.21 (heart, 0.94 ± 0.01; chambers, 0.87 ± 0.05; GVs, 0.72 ± 0.18; CAs, 0.50 ± 0.18; valves, 0.62 ± 0.16; nodes, 0.52 ± 0.16) for MR-Linac, and 0.80 ± 0.16 (heart, 0.95 ± 0.01; chambers, 0.93 ± 0.04; GVs, 0.84 ± 0.06; CAs, 0.77 ± 0.12; valves, 0.68 ± 0.23; nodes, 0.72 ± 0.11) for CCTA. Furthermore, > 80% and > 70% reductions in training time and parameters were achieved, respectively.</div></div><div><h3>Conclusions</h3><div>MAGIC offers an efficient, lightweight solution capable of segmenting multiple image modalities and overlapping structures in a single model without compromising segmentation accuracy.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111296"},"PeriodicalIF":5.3,"publicationDate":"2025-11-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145573576","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2025-11-15DOI: 10.1016/j.radonc.2025.111284
Eline H. Huele , Renée Hovenier , Bas B.J. Bindels , Anouk van Oss , Rene H. Kuijten , Roxanne Gal , Bart J. Pielkenrood , Jorrit-Jan Verlaan , Nicolien Kasperts , Wietse S.C. Eppinga , Yvette M. van der Linden , Joanne M. van der Velden , Helena M. Verkooijen
Background
Bone metastases often cause pain and interfere with quality of life (QoL). Survival of patients with bone metastases is increasing and there is a paucity of data about QoL in long-term survivors after external beam radiotherapy.
Methods
Patients with bone metastases treated with radiotherapy were identified from the PRospective Evaluation of interventional StudiEs on boNe meTastases (PRESENT+) cohort. Patient and treatment characteristics were collected before radiotherapy. Patient-reported QoL scores were collected up to 5 years, and included global QoL, physical and emotional functioning, symptoms including pain (EORTC QLQ-C15 PAL) and functional interference and psychosocial aspects (EORTC QLQ-BM22). Dutch reference QoL data were used for comparison.
Results
In total, 458 (23%) short-term (<3 months), 915 (46%) intermediate-term (≥3 months and <2 years), and 603 (31%) long-term (≥2 years) survivors were included. Median survival of the full cohort (n = 1,976) was 9.5 months (interquartile range, 3.3–30.5 months). Two third (n = 1,316, 67%) provided QoL scores. Long-term survivors reported a substantial increase in global QoL, emotional and physical functioning, and functional interference 4 weeks post radiotherapy; this improvement persisted over 5 years. Their QoL scores were substantially better than those of short-term and intermediate-term survivors, and somewhat lower than the reference population.
Conclusion(s)
Patients with bone metastases, referred for radiotherapy, represent a highly heterogeneous group. One third survives more than two years with QoL scores comparable to, or somewhat lower than the Dutch reference population. An individualized treatment approach is needed, addressing both short-term symptom palliation and, if applicable, long-term goals.
{"title":"Course of quality of life in short- and long-term survivors after radiotherapy for bone metastases","authors":"Eline H. Huele , Renée Hovenier , Bas B.J. Bindels , Anouk van Oss , Rene H. Kuijten , Roxanne Gal , Bart J. Pielkenrood , Jorrit-Jan Verlaan , Nicolien Kasperts , Wietse S.C. Eppinga , Yvette M. van der Linden , Joanne M. van der Velden , Helena M. Verkooijen","doi":"10.1016/j.radonc.2025.111284","DOIUrl":"10.1016/j.radonc.2025.111284","url":null,"abstract":"<div><h3>Background</h3><div>Bone metastases often cause pain and interfere with quality of life (QoL). Survival of patients with bone metastases is increasing and there is a paucity of data about QoL in long-term survivors after external beam radiotherapy.</div></div><div><h3>Methods</h3><div>Patients with bone metastases treated with radiotherapy were identified from the PRospective Evaluation of interventional StudiEs on boNe meTastases (PRESENT+) cohort. Patient and treatment characteristics were collected before radiotherapy. Patient-reported QoL scores were collected up to 5 years, and included global QoL, physical and emotional functioning, symptoms including pain (EORTC QLQ-C15 PAL) and functional interference and psychosocial aspects (EORTC QLQ-BM22). Dutch reference QoL data were used for comparison.</div></div><div><h3>Results</h3><div>In total, 458 (23%) short-term (<3 months), 915 (46%) intermediate-term (≥3 months and <2 years), and 603 (31%) long-term (≥2 years) survivors were included. Median survival of the full cohort (n = 1,976) was 9.5 months (interquartile range, 3.3–30.5 months). Two third (n = 1,316, 67%) provided QoL scores. Long-term survivors reported a substantial increase in global QoL, emotional and physical functioning, and functional interference 4 weeks post radiotherapy; this improvement persisted over 5 years. Their QoL scores were substantially better than those of short-term and intermediate-term survivors, and somewhat lower than the reference population.</div></div><div><h3>Conclusion(s)</h3><div>Patients with bone metastases, referred for radiotherapy, represent a highly heterogeneous group. One third survives more than two years with QoL scores comparable to, or somewhat lower than the Dutch reference population. An individualized treatment approach is needed, addressing both short-term symptom palliation and, if applicable, long-term goals.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"214 ","pages":"Article 111284"},"PeriodicalIF":5.3,"publicationDate":"2025-11-15","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145542404","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":1,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
扫码关注我们
求助内容：
应助结果提醒方式：
京公网安备 11010802042870号