Radiation Medicine and Protection最新文献

{"title":"Radiomic analysis of pre-treatment prostate-specific membrane antigen (PSMA) PET/CT for prognostic stratification in patients with oligometastatic castration-sensitive prostate cancer","authors":"Yufeng Cao ,&nbsp;Philip Sutera ,&nbsp;William Silva Mendes ,&nbsp;Zhuoran Jiang ,&nbsp;Amit Sawant ,&nbsp;Luigi Marchionni ,&nbsp;Nicole L. Simone ,&nbsp;Ozan Cem Guler ,&nbsp;Phuoc T. Tran ,&nbsp;Cem Onal ,&nbsp;Lei Ren","doi":"10.1016/j.radmp.2026.03.002","DOIUrl":"10.1016/j.radmp.2026.03.002","url":null,"abstract":"<div><h3>Objective</h3><div>To explore the imaging biomarkers obtained from baseline multi-modality imaging PET/CT before metastasis-directed therapy (MDT), which could offer early response prediction before MDT treatment, optimizing patient management and improving outcomes.</div></div><div><h3>Methods</h3><div>The study analyzed a multi-institutional cohort of 118 patients with oligometastatic castration-sensitive prostate cancer (omCSPC), including 34 from Johns Hopkins Hospital (JHH) and 84 from Baskent University (BU), USA, all treated with stereotactic ablative radiation therapy SABR-MDT. Before MDT, all patients underwent PSMA PET and CT imaging. For radiomics analysis, the gross tumor volume (GTV) was defined as zone 1, with an additional 5 mm peritumoral expansion designated as zone 2. From these regions, 1308 radiomics features were extracted. Feature selection was performed using a mutual information function, identifying the five most informative radiomics features from prostate-specific membrane antigen (PSMA) PET and CT. These were combined with five key clinical parameters—age, Gleason score, total number of lesions, number of untreated lesions, ADT, and pre-MDT prostate-specific antigen (PSA)—as model inputs. Multiple machine-learning algorithms, including random forest, decision tree, support vector machine, and naïve Bayes, were applied to predict 2-year metastasis-free survival (MFS). Model performance was evaluated using both leave-one-out and cross-institution validation.</div></div><div><h3>Results</h3><div>In a leave-one-out test with 93 patients, random forest achieved 78% accuracy and an AUC of 0.80 in predicting 2-year MFS. In cross-institution validation with 61 BU and 32 JHH patients, random forest correctly predicted 2-year MFS for 69% and 71% of patients, with AUC values of 0.71 and 0.73, respectively. Kaplan Meier curve comparison shows statistically significant separation between “rapid progressors” and “non-rapid progressors” patients stratified by the model in both leave one out and cross-institution validation tests.</div></div><div><h3>Conclusion</h3><div>This study provides evidence that pre-treatment multi-modality imaging biomarkers derived from PSMA PET and CT can serve as valuable predictors of metastasis-free survival (MFS) in patients with omCSPC.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 2","pages":"Pages 81-87"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147754055","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Association of irradiation dose and increased incidence of xerostomia inpatients with nasopharyngeal carcinoma","authors":"Dongling You ,&nbsp;Jie Zhou ,&nbsp;Guiquan Zhu ,&nbsp;En Long ,&nbsp;Peng Xu ,&nbsp;Yiming Li ,&nbsp;Quan Yao ,&nbsp;Jinyi Lang ,&nbsp;Shun Lu","doi":"10.1016/j.radmp.2026.02.003","DOIUrl":"10.1016/j.radmp.2026.02.003","url":null,"abstract":"<div><h3>Objective</h3><div>To identify risk factors for post-radiation xerostomia and to evaluate the tolerance to doses administered to the parotid gland in patients with nasopharyngeal carcinoma (NPC).</div></div><div><h3>Methods</h3><div>A total of 462 patients with NPC who received radiotherapy at Nanfang Hospital of Southern Medical University and Sichuan Cancer Hospital from 2012 to 2016 were retrospectively analyzed. Xerostomia was diagnosed according to the Radiation Therapy and Oncology Group (RTOG) grade. Medcalc software and SPSS software were used for statistical analysis, and MATLAB and R software was used for modeling.</div></div><div><h3>Results</h3><div>The cut-off values for <em>D</em>_mean (the average dose of parotid), <em>D</em>_mean-L (the average dose of left parotid) and <em>D</em>_50-R (dose received at 50% volume of right parotid) were 26.00, 28.30 and 23.93 Gy, above which the risk of long-time xerostomia occurs is increased (<em>P</em> &lt; 0.05). <em>D</em>₅₀, <em>D</em>_50-L, <em>D</em>_50-R, <em>D</em>_mean, <em>D</em>_mean-L and <em>D</em>_mean-R were 28.66, 23.12, 29.04, 28.00, 28.30 and 35.55 Gy, respectively, but the results showed that this dose was not associated with highest xerostomia grade (HXG) (<em>P</em> &gt; 0.05). The relevant factors were independent risk factors for the development of xerostomia, the same was true for age, T stage, and <em>D</em>-PTV (dose of PTV). Using patient's gender, radiation treatment, number of radiotherapy sessions, T stage, TNM stage, <em>D</em>_mean and <em>D</em>₅₀ (dose received at 50% volume of the parotid glands) to established a prediction model to predict HXG and long-term xerostomia grading (LTG), the results indicated that Training showed <em>R</em> = 0.82. The independent risk factors were included to establish the nomogram prediction model. The results showed that the <em>C</em>-index was 0.671.</div></div><div><h3>Conclusion</h3><div>It is feasible to establish a neural network model or a nomogram prediction model using parotid dose parameters to predict the incidence of radiation-induced xerostomia.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 2","pages":"Pages 73-80"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147754007","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Corrigendum to “Hexyl-pentynoic acid serves as a novel radiosensitizer for breast cancer by inhibiting UCHL3-dependent Rad51 deubiquitination” [Radiat. Med. Protect. (2023) 204–213]","authors":"Zuchao Cai ,&nbsp;David Lim ,&nbsp;Beidi Jia ,&nbsp;Guochao Liu ,&nbsp;Wenwen Ding ,&nbsp;Zhendong Wang ,&nbsp;Zhujun Tian ,&nbsp;Junxuan Peng ,&nbsp;Fengmei Zhang ,&nbsp;Chao Dong ,&nbsp;Zhihui Feng","doi":"10.1016/j.radmp.2025.12.006","DOIUrl":"10.1016/j.radmp.2025.12.006","url":null,"abstract":"","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 2","pages":"Page 118"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147754008","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Radiation medicine and sustainable networks for global health and safety: Institutional progress and perspectives from KIRAMS","authors":"Jin Kyung Lee,&nbsp;Kyeong Min Kim,&nbsp;Dong Ho Kim,&nbsp;Minsu Cho,&nbsp;Yong Jin Lee,&nbsp;Jung Young Kim","doi":"10.1016/j.radmp.2026.03.001","DOIUrl":"10.1016/j.radmp.2026.03.001","url":null,"abstract":"<div><div>Within the United Nations Sustainable Development Goals (UN SDGs), health and well-being are one of the most important goals. This perspective highlights recent institutional progress at Korea Institute of Radiological and Medical Sciences (KIRAMS) that bridges scientific innovation, clinical application, and sustainability, which are aligned with the global vision of the UN SDGs. The key activities include long-term research on second primary cancers to strengthen survivorship care, the clinical application of targeted alpha-particle therapy using Actinium-225 for advanced neuroendocrine and prostate cancers, and the expansion of radiopharmaceutical research and development. These activities demonstrate how advanced radiation technologies can be translated into tangible clinical benefits while enhancing patient safety and long-term outcomes. In addition, KIRAMS has leveraged its expertise in radiation emergency medicine to contribute to broader public health responses. Collectively, these efforts illustrate how a mission-driven medical institution can align specialized radiation medicine with global sustainability agendas and strengthen international networks.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 2","pages":"Pages 114-117"},"PeriodicalIF":0.0,"publicationDate":"2026-04-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147754004","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comparison of dose sparing between visual feedback guided- and active breathing coordinator-assisted- deep inspiration breath hold for left whole breast radiotherapy","authors":"Rencui Quan ,&nbsp;Dong Yang ,&nbsp;Hongtao Chen ,&nbsp;Xianming Li ,&nbsp;Zihuang Li ,&nbsp;Jingwen Liu ,&nbsp;Yucong Zhang ,&nbsp;Jianfeng Liu","doi":"10.1016/j.radmp.2026.01.004","DOIUrl":"10.1016/j.radmp.2026.01.004","url":null,"abstract":"<div><h3>Objective</h3><div>To compare the benefits of dosimetry between different deep inspiration breath hold (DIBH) techniques.</div></div><div><h3>Methods</h3><div>A total of sixty patients eligible for adjuvant left whole-breast radiotherapy were enrolled, with equal allocation (<em>n</em> = 30 per group) to either vDIBH or active breathing coordinator (ABC)-DIBH techniques. All patients received computed tomography (CT) scans in both free breathing (FB)and DIBH states. The target and crucial organs at risks (OARs) were precisely delineated on both breathing states of CT scans. The primary endpoints focused on assessing the irradiation exposure to OARs, including the heart, left anterior descending coronary (LAD) artery, left lung and stomach.</div></div><div><h3>Results</h3><div>The dose sparing advantage was more pronounced with vDIBH compared to ABC-DIBH, correlating with its enhanced lung expansion capacity. The relative dose reduction of heart's mean dose (32 % ± 18 % <em>vs</em>. 22 % ± 15 %, <em>P</em> = 0.018) and LAD artery's mean dose (43 % ± 25 % <em>vs.</em> 30 % ± 16 %, <em>P</em> = 0.023) were improved with vDIBH comparing with ABC-DIBH. However, regardless of whether vDIBH or ABC-DIBH was employed, our study did not demonstrate the protective effect of DIBH on the left lung.</div></div><div><h3>Conclusion</h3><div>The vDIBH group demonstrated a remarkable dose reduction for OARs mentioned above compared to the ABC-DIBH group, attributed to greater lung inflation during vDIBH.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 1","pages":"Pages 43-47"},"PeriodicalIF":0.0,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147412425","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep DIA proteomic analysis of mouse serum in early-stage acute irradiation","authors":"Yuhong Xiang ,&nbsp;Meng Jia ,&nbsp;Ziteng Liu ,&nbsp;Na Huang ,&nbsp;Heming Wang ,&nbsp;Zhenhua Qi ,&nbsp;Liping Shen ,&nbsp;Qi Wang ,&nbsp;Yaqiong Li ,&nbsp;Zhidong Wang","doi":"10.1016/j.radmp.2026.01.003","DOIUrl":"10.1016/j.radmp.2026.01.003","url":null,"abstract":"<div><h3>Objective</h3><div>To investigate the early serum proteomic response in acute radiation syndrome (ARS) and identify potential biomarkers for early diagnosis and injury stratification using deep data-independent acquisition (DIA) proteomic.</div></div><div><h3>Methods</h3><div>24 male C57BL/6 J mice (8 weeks old) were randomly assigned to four groups for γ-ray whole-body irradiation at doses of 0, 2, 6.5 and 10 Gy, respectively. Serum samples were collected 6 h post-irradiation for deep DIA proteomic analysis combined with nanoparticle-based enrichment. Bioinformatics methods were employed for screening differentially expressed proteins (DEPs), functional enrichment and core protein set identification. Western blot was used to validate the dose and time effects of candidate biomarkers.</div></div><div><h3>Results</h3><div>Leveraging deep DIA proteomics, we established an early-stage radiation injury proteomic profile in mice, quantifying over 4500 serum proteins. Analysis revealed a distinct dose-dependent response pattern: at a dose of 2 Gy, DNA damage repair pathways were activated concurrently with significant downregulation of cardiovascular function-related proteins; when the irradiation dose reached 6.5 Gy, the response shifted to immune hyperactivation coupled with widespread impairment of fundamental cellular functions; at a cumulative radiation dose of 10 Gy, complete collapse of basic biological processes including protein synthesis was observed. Notably, neutrophil granule protein (NGP) demonstrated stable, dose-dependent elevation within 24 h post-irradiation. At 6 h post-exposure, NGP levels increased approximately 2-fold, 7.2-fold, and 7.5-fold at doses of 2 Gy (<em>t</em>=3.47, <em>P</em>&lt;0.05), 6.5 Gy (<em>t</em>=4.51, <em>P</em>&lt;0.05), and 10 Gy (<em>t</em>=15.91, <em>P</em>&lt;0.001), respectively, indicating its strong potential as an early biomarker for radiation injury.</div></div><div><h3>Conclusion</h3><div>This study elucidates a dose-dependent serum proteomic response pattern in early ARS, revealing that molecular alterations associated with long-term cardiovascular effects of radiation occur as early as the initial post-irradiation phase. The stable early response of NGP within 24 h post-exposure provides a potential molecular target and theoretical foundation for early warning and clinical intervention in ARS.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 1","pages":"Pages 18-27"},"PeriodicalIF":0.0,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147412753","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"An orchestra of survival: Molecular mechanisms of radiation resistance across extremophiles","authors":"Ningang Liu ,&nbsp;Fengmei Cui ,&nbsp;Yang Jiao ,&nbsp;Kaixiao Zhou ,&nbsp;Jian Tong","doi":"10.1016/j.radmp.2026.01.001","DOIUrl":"10.1016/j.radmp.2026.01.001","url":null,"abstract":"<div><div>An orchestra of molecular mechanisms endows radiation-resistant extremophiles with extraordinary survival capabilities under extreme ionizing radiation. This review systematically explores the arsenal of key molecular strategies underpinning extreme radioresistance, including efficient DNA damage response and repair pathways, robust reactive oxygen species (ROS) scavenging systems, specialized protective proteins (e.g., Dsup, PprI), and coordinated regulatory networks. This review extensively examines species-specific adaptations in a diverse range of radioresistant organisms, such as <em>Deinococcus radiodurans</em>, archaea, lichens, bdelloid rotifers, <em>Drosophila melanogaster</em>, cockroaches, <em>Caenorhabditis elegans</em>, and tardigrades. Despite the conservation of core mechanisms such as DNA repair machinery, significant interspecies variations exist in antioxidant defense systems and cellular structure remodeling strategies. Based on the available data, this review further conducts an analysis of convergent <em>vs.</em> divergent evolution and quantitative efficiency metrics across species. Finally, translational applications of these mechanisms in radiobiology, medicine and space science are discussed, with future research directions highlighted that aim to exploit these natural radioprotective systems for technological and therapeutic innovations.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 1","pages":"Pages 1-17"},"PeriodicalIF":0.0,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147412755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Evaluation of the pulmonary perfusion generated by deep learning-based enhanced cone beam computed tomography","authors":"Yuyao Wu ,&nbsp;Yu-Lap Lam ,&nbsp;Ge Ren ,&nbsp;Jing Cai","doi":"10.1016/j.radmp.2026.01.002","DOIUrl":"10.1016/j.radmp.2026.01.002","url":null,"abstract":"&lt;div&gt;&lt;h3&gt;Objective&lt;/h3&gt;&lt;div&gt;To evaluate perfusion images generated by enhanced cone beam computed tomography (CBCT), quantify their correlation with CT-based perfusion images, and analyze the factors that may affect their clinical application.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Methods&lt;/h3&gt;&lt;div&gt;Thirty lung cancer patients with CBCT-CT pairs and single-photon emission computed tomography (SPECT) perfusion images were studied. A previously developed Cycle GAN CBCT enhancement model and a 3D CNN CT-to-perfusion model were utilized to generate perfusion images. Both CBCT images and SPECT perfusion images were registered to the corresponding CT images for further process. Registered CBCT images were input into the Cycle GAN model to produce enhanced CBCT (eCBCT) images. Subsequently, the lung mask generated from CBCT was applied to CBCT, eCBCT, CT, and SPECT images, generating CBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, eCBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, CT&lt;sub&gt;segCBCT&lt;/sub&gt;, and SPECT&lt;sub&gt;segCBCT&lt;/sub&gt;. Additionally, the CT lung mask was applied to CT and SPECT images for comparison, resulting in CT&lt;sub&gt;segCT&lt;/sub&gt; and SPECT&lt;sub&gt;segCT&lt;/sub&gt;. Then all 4 sets of CT images (CBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, eCBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, CT&lt;sub&gt;segCBCT&lt;/sub&gt;, CT&lt;sub&gt;segCT&lt;/sub&gt;) were translated into perfusion images using the 3D CNN network. Evaluation metrics, including Spearman's correlation coefficient &lt;em&gt;R&lt;/em&gt; and structural similarity index measure (SSIM) for voxel-wise agreement and dice similarity coefficient (DSC) for function-wise agreement, were calculated between the generated perfusion images and their corresponding ground truth SPECT images.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Results&lt;/h3&gt;&lt;div&gt;After enhancement, the SSIM between CBCT and CT has increased from 0.804 ± 0.072 to 0.824 ± 0.068, while the peak signal-to-noise ratio (PSNR) increased from 24.16 ± 2.82 to 25.04 ± 2.88. Furthermore, the mean absolute error (MAE) decreased from 47.75 ± 15.95 to 40.13 ± 14.61, showing that the eCBCT has higher voxel-wise agreement with CBCT. For CT-to-perfusion translation, the average Spearman correlation &lt;em&gt;R&lt;/em&gt; for CBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, eCBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, and CT&lt;sub&gt;segCBCT&lt;/sub&gt;, and CT&lt;sub&gt;segCT&lt;/sub&gt; was 0.64, 0.68, 0.73, 0.76, and the average SSIM was 0.62, 0.78, 0.76, 0.79, respectively. Function-wise evaluation of generated perfusion images showed a moderate agreement with the ground truth and the 4 sets had similar results, with DSC (High function) around 0.62 and DSC (Low function) around 0.69. In general, the order of performance from best to worse is: CT&lt;sub&gt;segCT&lt;/sub&gt;, CT&lt;sub&gt;segCBCT&lt;/sub&gt;, eCBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, and CBCT&lt;sub&gt;segCBCT&lt;/sub&gt;, as expected.&lt;/div&gt;&lt;/div&gt;&lt;div&gt;&lt;h3&gt;Conclusion&lt;/h3&gt;&lt;div&gt;In this study, pulmonary perfusion images were generated from CBCT images, demonstrated the potential application of CBCT-based functional images in functional lung avoidance radiotherapy (FLART). Although reducing noise and artifacts in CBCT images can enhance the accuracy of CT-to-perfusion translation and ","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 1","pages":"Pages 28-35"},"PeriodicalIF":0.0,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147412752","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Comment on “Normal cell responses to 3D-CRT, VMAT, and helical tomotherapy: A comparative study”","authors":"Prashant Ramdas Kokiwar ,&nbsp;Amit Kumar Gupta ,&nbsp;A. Kavya ,&nbsp;Archana Dhyani","doi":"10.1016/j.radmp.2025.12.007","DOIUrl":"10.1016/j.radmp.2025.12.007","url":null,"abstract":"","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 1","pages":"Page 54"},"PeriodicalIF":0.0,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147402338","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A comparison of dimensionality reduction approaches for pre-treatment PSMA-PET/CT radiomics in prostate adenocarcinoma outcome prediction","authors":"Apurva Singh ,&nbsp;William Silva Mendes ,&nbsp;Sang-Bo Oh ,&nbsp;Ozan Cem Guler ,&nbsp;Aysenur Elmali ,&nbsp;Birhan Demirhan ,&nbsp;Amit Sawant ,&nbsp;Phuoc Tran ,&nbsp;Cem Onal ,&nbsp;Lei Ren","doi":"10.1016/j.radmp.2025.12.008","DOIUrl":"10.1016/j.radmp.2025.12.008","url":null,"abstract":"<div><h3>Objective</h3><div>To compare dimensionality-reduction methods for building prognostic models predicting metastasis-free survival (MFS) in localized prostate adenocarcinoma (PCa) patients treated with androgen-deprivation therapy and external radiotherapy using clinical factors and prostate-specific membrane antigen (PSMA)-PET/CT radiomics from primary tumor and nodal volumes.</div></div><div><h3>Methods</h3><div>A total of 134 localized PCa patients (28 with nodal involvement) were analyzed. Gross tumor volumes for primary tumors (GTVp) and nodes (GTVn) were segmented on CT and PET scans; a 5-mm peritumoral ring was defined. Radiomic features were normalized and reduced using three techniques: principal component analysis (PCA), supervised, and unsupervised feature selection. Model 1 combined tumor and nodal radiomics via volume-weighted averaging and consisted of 12 predictors including clinical variables (age, Gleason score, initial PSA, PSA relapse) and radiomics from primary, nodal, and ring regions. Data imbalance (24 metastasis, 110 no metastasis) was addressed using a 70:30 train–test split with imbalance correction applied to train set. Univariate Cox regression (<em>P</em> &lt; 0.05) identified top predictors from train set; multivariate Cox regression was performed on corrected training data and applied to test data. Model 2 used clinical variables and radiomics from GTVp + ring; Model 3 used clinical data alone. Binary classification for five-year MFS was also evaluated.</div></div><div><h3>Results</h3><div>Supervised feature selection achieved highest performance. Model 1 test had c-score 0.71 (0.65–0.72). The five-year MFS test classification was sensitivity 80.1 %, specificity 85.4 %, and AUC 0.84. Unsupervised and PCA methods showed slightly lower results (test c-scores: 0.70 and 0.69, respectively). Model 1 consistently outperformed Model 2 with c-score 0.64 and AUC 0.79, as well as Model 3 with c-score 0.54 and AUC 0.68 across all dimensionality-reduction techniques.</div></div><div><h3>Conclusion</h3><div>Supervised feature selection yielded the highest c-scores and AUCs for the models. Integrating PSMA-PET/CT radiomics from primary, nodal, and peritumoral regions with clinical factors significantly improved MFS prediction, highlighting multi-regional radiomics as a promising biomarker for personalized therapy in prostate cancer.</div></div>","PeriodicalId":34051,"journal":{"name":"Radiation Medicine and Protection","volume":"7 1","pages":"Pages 36-42"},"PeriodicalIF":0.0,"publicationDate":"2026-02-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"147412754","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}