Radiotherapy and Oncology最新文献

{"title":"Integrating radiomics and dosiomics with lung biologically equivalent dose for predicting symptomatic radiation pneumonitis after lung SBRT: A dual-center study","authors":"Yuxin Jiao ,&nbsp;Yawen Wen ,&nbsp;Shihong Li ,&nbsp;Hongbo Gao ,&nbsp;Di Chen ,&nbsp;Li Sun ,&nbsp;Guangwu Lin ,&nbsp;Yanping Ren","doi":"10.1016/j.radonc.2025.111129","DOIUrl":"10.1016/j.radonc.2025.111129","url":null,"abstract":"<div><h3>Background and purpose</h3><div>This study focused on developing and validating a composite model that integrates radiomic and dosiomic features based on a lung biologically equivalent dose segmentation approach to predict symptomatic radiation pneumonitis (SRP) following lung SBRT.</div></div><div><h3>Materials and methods</h3><div>A dual-centered cohorts of 182 lung cancer patients treated with SBRT were divided into training, validation, and external testing sets. Radiomic and dosiomic features were extracted from two distinct regions of interest (ROIs) in the planning computed tomography (CT) images and 3D dose distribution maps, which encompassed both the entire lung and biologically equivalent dose (BED) regions. Feature selection involved correlation filters and LASSO regularization. Five machine learning algorithms generated three individual models (dose-volume histogram [DVH], radiomic [R], dosiomic [D]) and three combined models (R + DVH, R + D, R + D + DVH). Performance was evaluated via ROC analysis, calibration, and decision curve analysis.</div></div><div><h3>Results</h3><div>Among the clinical and dosimetric factors, V_BED70 (α/β = 3 Gy) of the lung was recognized as an independent risk factor for SRP. BED-based radiomic and dosiomic models outperformed whole-lung models (AUCs: 0.806 vs. 0.674 and 0.821 vs. 0.647, respectively). The R + D + DVH trio model achieved the highest predictive accuracy (AUC: 0.889, 95 % CI: 0.701–0.956), with robust calibration and clinical utility.</div></div><div><h3>Conclusions</h3><div>The R + D + DVH trio model based on lung biologically equivalent dose segmentation approach outperforms other models in predicting SRP across various SBRT fractionation schemes.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111129"},"PeriodicalIF":5.3,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145045863","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}

{"title":"Efficacy and safety of twice-daily accelerated hyperfractionated re-irradiation for thoracic malignancies","authors":"Victoria L. Doss ,&nbsp;Cecilia F.P.M. de Sousa ,&nbsp;Elaina Hales ,&nbsp;Tinker Trent ,&nbsp;Esi Hagan ,&nbsp;Tsion Gebre ,&nbsp;Anas Obaideen ,&nbsp;Meti Negassa ,&nbsp;Dezhi Liu ,&nbsp;Chen Hu ,&nbsp;Akila N. Viswanathan ,&nbsp;Heng Li ,&nbsp;K.Ranh Voong ,&nbsp;Xun Jia ,&nbsp;Todd McNutt ,&nbsp;Rachel B. Ger ,&nbsp;Russell K. Hales","doi":"10.1016/j.radonc.2025.111130","DOIUrl":"10.1016/j.radonc.2025.111130","url":null,"abstract":"<div><h3>Introduction</h3><div>Balancing the demands for dose-escalated re-irradiation with toxicity to organs at risk (OARs) is a challenge. Twice-daily (BID) accelerated hyperfractionated re-irradiation (re-RT) is a potential solution, yet there is no guidance for its application for thoracic malignancies.</div></div><div><h3>Methods</h3><div>All BID thoracic re-RT cases at our institution since 2014 were reviewed. Prior radiation courses were deformably registered. Summary plan dosimetrics were extracted using per-voxel cumulative EQD2 (α/β = 3 for normal tissues). Maximum toxicity following BID re-RT was graded per CTCAE v5.0 and assessed for relationship to radiation. Local failure (LF), progression-free survival (PFS), and overall survival (OS) were measured and compared between subgroups based on histology, re-RT dose, and metastatic status.</div></div><div><h3>Results</h3><div>Sixty-four BID re-RT cases delivering at least 45 Gy in 30 fractions were reviewed. Fifty-eight patients (90.6 %) received a prior definitive course of radiation. The median first course EQD2 (α/β = 10) was 60.2 Gy. Median follow-up after re-RT was 14.2 months. Median OS was 18.5 months from completion of BID re-RT, and 59.0 % of patients had sustained local control at two years. Ten patients (15.6 %) experienced grade ≥3 toxicity. There were no grade 4 toxicities. There was one grade 5 event possibly attributable to re-RT. Of 24 OAR doses exceeding American Radium Society (ARS) and American College of Radiology (ACR) cumulative dose recommendations for re-RT, one case was associated with grade ≥3 toxicity.</div></div><div><h3>Conclusions</h3><div>Accelerated hyperfractionated BID re-RT is safe and effective for thoracic re-RT, providing durable local control and favorable survival outcomes with acceptable toxicity.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111130"},"PeriodicalIF":5.3,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145055666","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}

{"title":"Outcomes of local therapies for recurrent and radiation-associated soft tissue sarcomas of the extremities and trunk: is there a role for re-irradiation?","authors":"Reinhardt Krcek ,&nbsp;Anthony M. Griffin ,&nbsp;Charles N. Catton ,&nbsp;Peter W.M. Chung ,&nbsp;Brian O’Sullivan ,&nbsp;Amy Parent ,&nbsp;Siyer Roohani ,&nbsp;David B. Shultz ,&nbsp;Philip Wong ,&nbsp;Peter C. Ferguson ,&nbsp;Kim Tsoi ,&nbsp;Jay S. Wunder ,&nbsp;David G. Kirsch","doi":"10.1016/j.radonc.2025.111136","DOIUrl":"10.1016/j.radonc.2025.111136","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Managing locally relapsed soft tissue sarcomas (STS) after prior radiotherapy (RT) and radiation-associated sarcomas (RAS) is challenging. Limb-sparing surgery without repeated RT may be suboptimal in anatomically constrained cases, while amputation compromises function and quality of life. This study evaluates outcomes of patients treated with re-irradiation plus surgery versus surgery alone.</div></div><div><h3>Material and methods</h3><div>This retrospective cohort study included patients treated at a single sarcoma centre (Princess Margaret Cancer Centre/Mount Sinai Hospital, Toronto) for locally relapsed STS after prior RT and surgery, or for RAS, diagnosed between October 1985 and August 2024. Treatment and outcome data were analyzed to compare local control, limb preservation, complications, distant failure, and overall survival between patients receiving re-irradiation plus surgery versus surgery alone.</div></div><div><h3>Results</h3><div>A total of 93 patients with recurrent STS and 13 with RAS were analyzed. RT for re-irradiation was mostly administered preoperatively using a twice-daily (BID) fractionation scheme. The 5-year local control rate for recurrent STS was 82.8 % in the re-irradiation plus surgery group and 67.9 % in the surgery-only group (p = 0.336). The 5-year limb preservation rate was significantly higher with re-irradiation (80.8 %) versus surgery alone (40.2 %) (p &lt; 0.001). Complications requiring intervention were more frequent with re-irradiation for relapses (13/43 vs. 3/50, p = 0.002). Freedom from distant failure and overall survival (OS) at 5 years were similar between the groups (65.1 vs. 71.7 % and 57.3 % vs. 55 %). The 5-year OS after RAS diagnosis was 75.2 %, with five complications, four local failures, and four patients developing distant failure during follow-up.</div></div><div><h3>Conclusion</h3><div>Re-irradiation combined with surgery represents a viable alternative to amputation or surgery alone in selected patients with locally relapsed STS or RAS, with high rates of limb salvage and acceptable toxicity.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111136"},"PeriodicalIF":5.3,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145055732","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}

{"title":"A multidimensional deep ensemble learning model predicts pathological response and outcomes in esophageal squamous cell carcinoma treated with neoadjuvant chemoradiotherapy from pretreatment CT imaging: A multicenter study","authors":"Yunsong Liu ,&nbsp;Yang Su ,&nbsp;Jun Peng ,&nbsp;Wencheng Zhang ,&nbsp;Fangdong Zhao ,&nbsp;Yue Li ,&nbsp;Xinyun Song ,&nbsp;Zeliang Ma ,&nbsp;Wanting Zhang ,&nbsp;Jianrui Ji ,&nbsp;Ye Chen ,&nbsp;Yu Men ,&nbsp;Feng Ye ,&nbsp;Kuo Men ,&nbsp;Jianjun Qin ,&nbsp;Wenyang Liu ,&nbsp;Xin Wang ,&nbsp;Nan Bi ,&nbsp;Liyan Xue ,&nbsp;Wen Yu ,&nbsp;Zhouguang Hui","doi":"10.1016/j.radonc.2025.111133","DOIUrl":"10.1016/j.radonc.2025.111133","url":null,"abstract":"<div><h3>Purpose</h3><div>Neoadjuvant chemoradiotherapy (nCRT) followed by esophagectomy remains standard for locally advanced esophageal squamous cell carcinoma (ESCC). However, accurately predicting pathological complete response (pCR) and treatment outcomes remains challenging. This study aimed to develop and validate a multidimensional deep ensemble learning model (DELRN) using pretreatment CT imaging to predict pCR and stratify prognostic risk in ESCC patients undergoing nCRT.</div></div><div><h3>Methods</h3><div>In this multicenter, retrospective cohort study, 485 ESCC patients were enrolled from four hospitals (May 2009–August 2023, December 2017–September 2021, May 2014–September 2019, and March 2013–July 2019). Patients were divided into a discovery cohort (n = 194), an internal cohort (n = 49), and three external validation cohorts (n = 242). A multidimensional deep ensemble learning model (DELRN) integrating radiomics and 3D convolutional neural networks was developed based on pretreatment CT images to predict pCR and clinical outcomes. The model’s performance was evaluated by discrimination, calibration, and clinical utility. Kaplan-Meier analysis assessed overall survival (OS) and disease-free survival (DFS) at two follow-up centers.</div></div><div><h3>Results</h3><div>The DELRN model demonstrated robust predictive performance for pCR across the discovery, internal, and external validation cohorts, with area under the curve (AUC) values of 0.943 (95 % CI: 0.912–0.973), 0.796 (95 % CI: 0.661–0.930), 0.767 (95 % CI: 0.646–0.887), 0.829 (95 % CI: 0.715–0.942), and 0.782 (95 % CI: 0.664–0.900), respectively, surpassing single-domain radiomics or deep learning models. DELRN effectively stratified patients into high-risk and low-risk groups for OS (log-rank P = 0.018 and 0.0053) and DFS (log-rank P = 0.00042 and 0.035). Multivariate analysis confirmed DELRN as an independent prognostic factor for OS and DFS.</div></div><div><h3>Conclusion</h3><div>The DELRN model demonstrated promising clinical potential as an effective, non-invasive tool for predicting nCRT response and treatment outcome in ESCC patients, enabling personalized treatment strategies and improving clinical decision-making with future prospective multicenter validation.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111133"},"PeriodicalIF":5.3,"publicationDate":"2025-09-10","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145055563","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}

{"title":"Staged Gamma Knife radiosurgery for large brain metastases: local control and the influence of systemic treatment","authors":"R.G. Huisman ,&nbsp;P. van Schie ,&nbsp;L.G. Merckel ,&nbsp;T. Wiersma ,&nbsp;J.L. Knegjens ,&nbsp;E.P.M. Jansen ,&nbsp;A. Compter ,&nbsp;D. Brandsma ,&nbsp;M. Bot ,&nbsp;J. Hoogmoed ,&nbsp;P.C. de Witt Hamer ,&nbsp;G.R. Borst ,&nbsp;R. Post","doi":"10.1016/j.radonc.2025.111127","DOIUrl":"10.1016/j.radonc.2025.111127","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Staged Gamma Knife radiosurgery (SGKRS) delivers high-dose radiotherapy to large brain metastases (BM) in two or three fractions with a time interval of several weeks. Various systemic treatments have also demonstrated favorable intracranial responses. Therefore, the outcome of patients undergoing radiosurgery and systemic treatment for large BM is of high interest but unknown.</div></div><div><h3>Materials and methods</h3><div>A retrospective cohort study was conducted on patients with large BM treated with SGKRS without previous local treatment directed to the brain. The primary outcome measure was the probability of intracranial local control at 12 months, calculated by the Kaplan-Meier method. Univariable and multivariable Cox regression analyses were performed to identify variables associated with intracranial local control.</div></div><div><h3>Results</h3><div>295 patients were included. Intracranial local control probability at 12 months was 83 % and overall survival at 12 months was 39 %. In the multivariable Cox regression analysis, receiving any type of concurrent or adjuvant systemic treatment (adjusted hazard ratio [aHR] 0.30, 95 % confidence interval [CI] 0.15–0.61) and volume reduction between the first and second fraction (aHR 0.99, 95 % CI 0.98–0.998) were significantly associated with better intracranial local control. Larger total volume of all treated BM (aHR 1.02, 95 % CI 1.01–1.04) was significantly associated with worse intracranial local control. The probability of symptomatic cerebral radiation necrosis at 12 months was 26 %.</div></div><div><h3>Conclusion</h3><div>SGKRS results in high local control, with further improvement when systemic treatment is administered. However, overall survival remains limited, highlighting the importance of adequate patient selection.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111127"},"PeriodicalIF":5.3,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145034037","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}

{"title":"Comment on: Radiographic classification of mandibular osteoradionecrosis: A blinded prospective multi-disciplinary interobserver diagnostic performance study”","authors":"Efsun Somay ,&nbsp;Erkan Topkan ,&nbsp;Ugur Selek","doi":"10.1016/j.radonc.2025.111131","DOIUrl":"10.1016/j.radonc.2025.111131","url":null,"abstract":"","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111131"},"PeriodicalIF":5.3,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145027398","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}

{"title":"A randomized Phase II trial of High Dose-Rate (HDR) and Low Dose-Rate (LDR) brachytherapy as monotherapy in localized prostate cancer: analysis of initial arms of Canadian cancer trials group PR19 (NCT02960087)","authors":"G. Morton ,&nbsp;E. Vigneault ,&nbsp;M. Barkati ,&nbsp;J. Helou ,&nbsp;T.M. Niazi ,&nbsp;J. Robinson ,&nbsp;D.A. Loblaw ,&nbsp;C.L. Tseng ,&nbsp;H.T. Chung ,&nbsp;G. Delouya ,&nbsp;C. Menard ,&nbsp;A.G. Martin ,&nbsp;P. Chung ,&nbsp;D. Batchelar ,&nbsp;M.D. Brundage ,&nbsp;K. Whelan ,&nbsp;K. Ding ,&nbsp;W. Parulekar","doi":"10.1016/j.radonc.2025.111124","DOIUrl":"10.1016/j.radonc.2025.111124","url":null,"abstract":"<div><h3>Purpose/objectives</h3><div>Low Dose-Rate Brachytherapy (LDR) and High Dose-Rate Brachytherapy (HDR) are options for favorable risk prostate cancer. We hypothesized that HDR provides comparable disease control with less urinary toxicity. Primary objective was to determine prostate cancer control at 48 months, defined as a PSA &lt; 0.4 ng/ml.</div></div><div><h3>Materials/methods</h3><div>Eligible patients had low and intermediate risk diseases. Randomization was to Arm 1 LDR with I-125 to 144 Gy, or Arm 2 HDR with 19 Gy x1 + intraprostatic boost. Follow-up included PSA, toxicity (CTCAE v 4.0), and Quality of Life (EPIC 26). Arm 2 was closed in May 2019 due to evidence of inferior outcomes, and a third arm (HDR 13.5 Gy x 2) was opened. We report outcomes of Arms 1 (LDR) and 2 (HDR single fraction) prior to study amendment.</div></div><div><h3>Results</h3><div>103 patients were randomized: 51 to LDR (Arm 1) and 52 to HDR (Arm 2). Median age 65 years; 76 % had Gleason 3 + 4, 90 % PSA &lt; 10, and 80 % stage T1c. Median follow-up 53 months. PSA control at 4 years was 78.4 % for LDR, and 21.2 % for HDR. Local progression rates: 2 % LDR, 13.5 % HDR. Grade 3 toxicity occurred in 6 patients for LDR and 2 patients Arm 2. Urinary irritative and bowel symptoms were worse in the first 6 months for LDR.</div></div><div><h3>Conclusions</h3><div>LDR brachytherapy has high prostate cancer disease control rate at 4 years, although with worse impact on urinary symptoms in the first 6 months. Single fraction HDR was associated with unacceptable cancer control.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111124"},"PeriodicalIF":5.3,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145034094","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}

{"title":"Intraoperative radiotherapy for resectable brain metastases: a systematic review and meta-analysis","authors":"Cas Stefaan Dejonckheere ,&nbsp;Matthias Schneider ,&nbsp;Anna-Laura Potthoff ,&nbsp;Motaz Hamed ,&nbsp;Davide Scafa ,&nbsp;Thomas Zeyen ,&nbsp;Lea L. Friker ,&nbsp;Molina Grimmer ,&nbsp;Fabian Kugel ,&nbsp;Stephan Garbe ,&nbsp;Alexander Radbruch ,&nbsp;Hartmut Vatter ,&nbsp;Frank Anton Giordano ,&nbsp;Ulrich Herrlinger ,&nbsp;Eleni Gkika ,&nbsp;Gustavo Renato Sarria ,&nbsp;Julian Philipp Layer","doi":"10.1016/j.radonc.2025.111128","DOIUrl":"10.1016/j.radonc.2025.111128","url":null,"abstract":"<div><h3>Background</h3><div>In recent years, intraoperative radiotherapy (IORT) with low-energy X-rays is emerging as an alternative to postoperative stereotactic radiotherapy (SRT) of the resection cavity in patients with resectable brain metastases (BMs).</div></div><div><h3>Methods</h3><div>We performed a systematic review of the MEDLINE, Embase, and Scopus databases, including all original articles on IORT for resectable BMs from 2015 to 2025. Data on safety, local control, and survival outcomes were collected.</div></div><div><h3>Results</h3><div>Ten records (5 prospective single-arm trials) were included, representing 261 patients (49 % lung primary) with a median follow-up (range) of 14 (0–79) months. 77 % of patients had a solitary BM at the time of surgery and IORT. The median applicator size was 2.0 cm and the median prescribed dose (range) 22.3 (20–30) Gy. The 1-year local control rate was 93 % and the 1-year distant brain control rate 48 %. Median overall survival was 19 months. Only 6 % of patients developed leptomeningeal disease and the cumulative rate of radiation necrosis was 2.6 % (grade 1 in 56 % of cases). The median time to next treatment beyond BM therapy (range) was 31 (1–136) days. This was significantly shorter compared to SRT control collectives.</div></div><div><h3>Conclusions</h3><div>IORT for patients with BMs has a favorable toxicity profile and yields excellent local control. A potential advantage is the rapid completion of interdisciplinary BM treatment, allowing a swift transition to subsequent cancer treatments. A planned registry and a prospective randomized phase 3 trial will establish the preferred radiotherapy modality in the context of resectable BMs.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111128"},"PeriodicalIF":5.3,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145034122","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}

{"title":"Minimum sample size calculation for radiomics-based binary outcome prediction models: Theoretical framework and practical example","authors":"Qian Cao ,&nbsp;Zhaoyu Jiang ,&nbsp;Zhixiang Wang ,&nbsp;Leonard Wee ,&nbsp;Andre Dekker ,&nbsp;Zhen Zhang ,&nbsp;Ji Zhu","doi":"10.1016/j.radonc.2025.111134","DOIUrl":"10.1016/j.radonc.2025.111134","url":null,"abstract":"<div><h3>Background and purpose</h3><div>Determining the appropriate sample size for developing robust radiomics-based binary outcome prediction models and identifying the maximum number of predictors safely allowable within a fixed dataset size remain critical yet challenging tasks. This study aims to propose and demonstrate a structured method for addressing these issues, enhancing methodological rigor and practicality in radiomics research.</div></div><div><h3>Materials and methods</h3><div>We introduce a comprehensive sample size calculation framework for binary outcome prediction models in radiomic studies. The proposed approach integrates three key criteria: (1) maintaining a global shrinkage factor (<em>S</em>) ≥ 0.9 to control model overfitting, (2) ensuring a minimal absolute difference between apparent and adjusted performance metrics, and (3) precisely estimating the overall outcome risk. Additionally, we develop an accessible online calculation tool enabling researchers to efficiently determine either the minimum sample size or the maximum number of predictors permissible, based on clearly defined statistical parameters.</div></div><div><h3>Results</h3><div>The presented method systematically addresses model overfitting by integrating a global shrinkage factor into the calculation, providing robust estimates compared with traditional heuristic approaches (“rules of thumb”). Practical examples demonstrate that this structured method effectively balances predictive accuracy and generalizability, while the online tool provides researchers with a user-friendly platform to perform the necessary calculations.</div></div><div><h3>Conclusion</h3><div>Clear justification of sample size decisions is essential for developing reliable predictive models in radiomics research. By adopting a structured and rigorous calculation method, researchers can effectively minimize overfitting, ensure accurate risk estimation, and substantially enhance the reliability and validity of their predictive models.</div></div>","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111134"},"PeriodicalIF":5.3,"publicationDate":"2025-09-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145034068","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}

{"title":"In Response to Efsun Somay et al. “Integration of dosimetric parameters into radiographic classification of Osteoradionecrosis”","authors":"Zaphanlene Kaffey,&nbsp;Clifton D. Fuller,&nbsp;Amy C. Moreno,&nbsp;Laia Humbert-Vidan","doi":"10.1016/j.radonc.2025.111132","DOIUrl":"10.1016/j.radonc.2025.111132","url":null,"abstract":"","PeriodicalId":21041,"journal":{"name":"Radiotherapy and Oncology","volume":"212 ","pages":"Article 111132"},"PeriodicalIF":5.3,"publicationDate":"2025-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"145030525","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}