最新刊期
卷 35 ， 期 8 ， 2025
- 摘要：Background and purpose: Diffuse large B-cell lymphoma (DLBCL) is subclassified into germinal center B-cell-like (GCB) and non-GCB subtypes, which differ in prognosis and treatment response. However, current distinction still relies on invasive pathological assays. This study developed radiomics and deep-learning models based on multiparametric magnetic resonance imaging (MRI) to non-invasively differentiate the two subtypes preoperatively, thereby reducing dependence on histopathological examination. Methods: This study retrospectively included patients with pathologically confirmed DLBCL diagnosed at Huashan Hospital, Fudan University, and other institutions between March 2013 and December 2024. Using multiparametric MRI data, we developed DLBCL-subtype classification models that combined 4 radiomics-based machine-learning algorithms: support vector machine (SVM), logistic regression (LR), Gaussian process (GP) and Naive Bayes (NB), with 3 deep-learning architectures [densely-connected convolutional networks 121 (DenseNet121), residual network 101 (ResNet101) and EfficientNet-b5]. Additionally, two radiologists with different experience levels independently classified DLBCL on MRI in a blinded fashion. Model and radiologist performance were quantified using the area under the receiver operating characteristic curve (AUC), accuracy (ACC), and F1-score to evaluate their ability to distinguish GCB from non-GCB subtypes. This study was approved by the Ethics Committee of Huashan Hospital of Fudan University (No. KY2024-663), and all patients signed informed consents. Results: A total of 173 patients were enrolled (55 with GCB subtype and 118 with non-GCB subtype). Radiomics and deep learning methods effectively distinguished DLBCL subtypes. Among these, the GP radiomics model (based on T1-CE+T2-FLAIR+ADC sequences) and DenseNet121 deep learning model (based on T1-CE+T2-FLAIR+ADC sequences) demonstrated optimal performance. Both achieved excellent results on the internal validation set (GP: AUC=0.900, ACC=0.896, F1=0.840; DenseNet121: AUC=0.846, ACC=0.854, F1=0.774) and maintained robustness on the external validation set. Furthermore, the classification efficacy of the optimal AI model surpassed that of experienced radiologists (highest physician AUC=0.678). Conclusion: Radiomics and deep-learning models based on multiparametric MRI features can effectively differentiate GCB from non-GCB subtypes of DLBCL. Among them, GP and DenseNet121 exhibit outstanding performance, especially when integrating multi-sequence feature sets for classifying DLBCL subtypes on complex imaging data.关键词：Diffuse large B-cell lymphoma;Germinal center B-cell-like;Non-GCB;Radiomics;Deep learning17|2018|0更新时间：2025-12-31
- 摘要：Background and purpose: Relapsed or refractory diffuse large B-cell lymphoma (DLBCL) can be treated with chimeric antigen receptor T-cell (CAR-T) therapy. Imaging-based biomarkers may help identify patients likely to achieve clinical response to this immunotherapy. In this study, an 18-f1uoro-2-deoxy-D-glucose positron emission tomography-computed tomography (18F-FDG PET/CT) based model was developed to assess the progression-free survival (PFS) and overall survival (OS) of patients with relapsed or refractory (R/R) DLBCL who received CAR-T therapy. Methods: We retrospectively analyzed clinical and imaging data from patients with DLBCL who underwent CAR-T therapy at the First Affiliated Hospital of Soochow University between March 2017 and January 2022. Inclusion criteria: ① age ≥18 years old; ② pathologically confirmed R/R DLBCL; ③ 18F-FDG PET/CT performed before CAR-T cell therapy; ④ complete clinicopathologic data; ⑤ patients must have measurable lesions. Exclusion criteria: ① patients with incomplete clinical or imaging data; ② patients with other types of malignant tumors; ③ patients who have received granulocyte colony-stimulating factor treatment within 1 month prior to PET/CT scan. This study was reviewed by the Ethics Committee of the First Affiliated Hospital of Soochow University (ID: 2025256). Receiver operating characteristic (ROC) curves were used to determine the optimal thresholds for maximum standardized uptake value (SUVmax), tumor metabolic volume (MTV), and total glycolysis (TLG), and the patients were classified into high-risk and low-risk groups. Univariate and multivariate Cox regression analyses were used to identify potential prognostic factors and construct predictive models, which were visualized by drawing nomogram. Area under the ROC curve was used to assess the performance of each model. Results: A total of 61 patients (37 male patients and 24 female patients, aged 26-75 years) with DLBCL who underwent 18F-FDG PET/CT prior to CAR-T infusion were included. The median follow-up was 14 months; 36 patients (59.02%) had disease progression and 25 patients (40.98%) died. Multivariate analysis showed that grade of cytokine release syndrome (CRS) [Hazard ratio (HR)=3.671; P=0.003] and MTV (HR=0.171, P=0.004) were independent prognostic factors for OS; Eastern Cooperative Oncology Group (ECOG) score (HR=2.411, P=0.019), grade of CRS (HR=2.499; P=0.027), and MTV (HR=0.338, P=0.007) were independent prognostic factors for PFS. The combined model (MTV, ECOG score, grade of CRS) was better than the clinical model (ECOG score, grade of CRS), and metabolic parameter model (MTV) in predicting PFS and OS. Conclusion: 18F-FDG PET/CT metabolic parameter MTV in combination with traditional clinical risk factors (ECOG score, Grade of CRS) could identify patients with ultra-high risk of DLBCL.关键词：Diffuse large B-cell lymphoma;PET/CT;MTV;CAR-T;Prognosis5|465|0更新时间：2025-12-31
- 摘要：Background and purpose: Single-week ultra-hypofractionated whole breast irradiation (WBI) after breast-conserving surgery could shorten the treatment duration while ensuring efficacy and safety, making it a viable option for WBI. However, ultra-hypofractionated WBI requires daily image-guided radiotherapy (IGRT), and its impact on setup errors remains unclear. This study aimed to identify factors associated with set-up errors in ultra-hypofractionated WBI guided with daily cone-beam computed tomography (CBCT) and calculate margin expanded from clinical target volume (CTV) to planning target volume (PTV). Methods: This study included patients enrolled in a prospective trial that explored the safety of single-week ultra-hypofractionated WBI (NCT04926766) in Shanghai Ruijin Hospital, which was approved by Shanghai Ruijin Hospital Ethics Committee (No. 2020-352). All patients received CBCT1 after positioning. After correcting errors, patients received CBCT2. CBCT3 was conducted after radiotherapy was completed. The translational errors between CBCT1, CBCT2, and plan CT were initial and residual inter-fractional errors. The translational error between CBCT2 and CBCT3 was an intra-fractional error. The PTV margin was calculated according to the van Herk formula. Results: A total of 34 patients were enrolled in this study, and 510 CBCT images were collected. Daily CBCT significantly reduced set-up error in anterior-posterior (AP), superior-inferior (SI) and right-left (RL) directions (initial inter-fractional error vs residual inter-fractional error: AP, 2.8 mm vs 0.4 mm; SI, 1.6 mm vs 0.5 mm; RL, 1.8 mm vs 0.3 mm, all P<0.001). Higher CTV volume (>402.5 cm3 vs ≤402.5 cm3) was associated with larger residual inter-fractional error (0.5 mm vs 0.3 mm, P=0.023) and intra-fractional error (0.5 mm vs 0.2 mm, P=0.001) in AP direction. Higher CTV volume was also associated with larger residual inter-fractional error in the SI direction (0.6 mm vs 0.5 mm, P=0.037). Higher BMI (>23.2 kg/m2 vs ≤23.2 kg/m2) and larger weight (>60.0 kg vs ≤60.0 kg) were associated with larger intra-fractional error in AP direction: 0.7 mm vs 0.2 mm (P <0.001) and 0.5 mm vs 0.2 mm (P=0.033), respectively. Under guidance with daily CBCT, the recommended margins were 2.3 mm in AP direction, 2.8 mm in SI direction, and 2.0 mm in RL direction. However, in patients with CTV volume >402.5 cm3 and BMI>23.2 kg/m2, a larger margin was recommended in SI direction: 3.1 mm and 3.4 mm, respectively. Conclusion: The 3 mm margin was feasible under guidance with daily CBCT. The CTV to PTV margin should be larger in patients with higher BMI or CTV volume.关键词：Breast cancer;Ultra-hypofractionated radiotherapy;Whole breast irradiation;Set-up error;Margin12|285|0更新时间：2025-12-31
- 摘要：Background and purpose: The aberrant activation of pyruvate dehydrogenase kinase 1 (PDK1) drives tumor microenvironment remodeling and metastasis through mediating the Warburg effect. As a critical tumor-suppressive phosphatase, phosphatase and tensin homolog deleted on chromoseme ten (PTEN) activates PDK1 via loss of expression to induce aerobic glycolysis and accelerate tumor progression. The molecular interplay between PDK1 and PTEN in kidney renal clear cell carcinoma (KIRC) urgently requires systematic elucidation. This study aimed to clarify how PTEN regulates PDK1 to inhibit malignant phenotypes in KIRC. Methods: Bioinformatics analysis was conducted to compare PTEN and PDK1 expression levels as well as their prognostic correlations in the Cancer Genome Atlas (TCGA)-KIRC datasets. KIRC cell models was established by either silencing PDK1 or enhancing its expression, subsequently evaluating their malignancy characteristics through cell counting kit-8 (CCK-8) proliferation, colony formation, cell migration, and invasion assays. To validate the regulatory interactions, we used PDK1-overexpressing cells treated with a PTEN-specific inhibitor. Western blot was used to dectect the protein expression. Results: The TCGA-KIRC analysis found significantly higher mRNA levels of PTEN and PDK1 in tumor tissues compared to normal controls (P<0.05), yet this high expression was associated with improved overall survival (P<0.01). Besides, a strong positive correlation was observed between PTEN and PDK1 expressions (r=0.52, P<0.001). Functional assays demonstrated that PDK1 knockdown markedly promoted cell proliferation, migration, and invasion, whereas PDK1 overexpression exhibited opposing effects. Mechanistically, inhibiting PTEN worsened malignant behaviors (P<0.01), however, these effects were reversed by overexpressing PDK1. Conclusion: This study presents the first evidence of the dual tumor-suppressive function of the PTEN-PDK1 biological axis in renal cancer, which supports the development of precision treatment strategies based on novel targets.关键词：Clear cell renal cell carcinoma;PDK1;PTEN;Molecular mechanisms;Prognosis11|509|0更新时间：2025-12-31
- 摘要：Background and purpose: Prostate cancer and breast cancer are highly prevalent malignant tumors, and there occurrence and development are related to the tumor suppressor genes phosphatase and tensin homolog deleted on chremosome ten (Pten) and the transformation related protein 53 gene (Trp53). The loss of function of Trp53 is closely related. The simultaneous loss of the two can accelerate the malignant progression of tumors and induce therapeutic resistance. The gene-edited spontaneous tumor model of mice based on the Cre-loxP system is a key tool for studying the mechanism of cancer. Studies have shown that prostate-specific promoter (probasin, Pbsn)-driven iCre recombinase (Pbsn-iCre) can induce spontaneous prostate cancer in male mice, but its role in female breast cancer and transgender expression characteristics have not yet been clarified. In this study, we constructed Ptenfl/fl;Trp53fl/fl; Pbsn-iCre+ transgenic mouse model which was designed to explore its spontaneous tumor phenotype in prostate cancer and breast cancer, and to verify the expression characteristics of Pbsn in breast tissue. Methods: The Ptenfl/fl;Trp53fl/fl; Pbsn-iCre+ mouse model was established using Cre-loxP system by hybridization and continuous backcross screening with Ptenfl/fl mouse, Trp53fl/fl mouse, and Pbsn-iCre+ mouse (Ethical No.: FUSCC-IACUC-2025115). Pten, Trp53 and Pbsn-iCre genotypes were verified by polymerase chain reaction and agarose gel electrophoresis. The incidence of tumor in transgenic mice was monitored, and the histopathological characteristics of tumor were evaluated by hematoxylin-eosin staining. The protein levels of Pten and p53 in prostate and breast tumor tissues were analyzed by immunohistochemistry, and the distributions of Pbsn in breast, prostate, ovary, heart, liver and kidney were detected. Results: Ptenfl/ fl;Trp53fl/fl;Pbsn-iCre+ male mouse developed spontaneous prostate tumor at age of 5 month, and female mouse developed spontaneous breast tumor at age of 6 months. The pathological manifestations of prostate cancer were invasive acinar adenocarcinoma structure with glandular structure disorder and basement membrane destruction. The pathological manifestations of breast cancer were invasive ductal carcinoma with ductal epithelial dysplasia and interstitial lymphocyte infiltration. Immunohistochemistry confirmed the complete deletion of Pten and p53 proteins in prostate and breast tumor tissues, which verified the prostate and mammary gland specific gene knockout effect. Immunohistochemistry also confirmed that Pbsn protein was specifically expressed in prostate acinar epithelial cells, ovarian tissue, and mammary duct epithelial cells, but not in heart, liver and kidney. Conclusion: Pbsn-iCre is functionally expressed in female mammary glands, and the simultaneous loss of Pten/Trp53 induced by Pbsn-iCre may drive the development of prostate cancer in male and breast cancer in female mouse.关键词：Transgenic mouse spontaneous tumor model;Pbsn-iCre; Pten/Trp53 double deletion;Prostate cancer;Breast cancer2|428|0更新时间：2025-12-31
- 摘要：Background and purpose: In breast cancer surgery, margin status assessment significantly impacts patient prognosis, with positive margins indicating higher recurrence and metastasis risks. Ensuring complete tumor resection is thus critical for surgical success. Indocyanine green (ICG) has garnered attention for its potential real-time imaging of breast cancer lesions under near-infrared light. This study employed ICG for intraoperative assessment of breast cancer lesion margin status and further explored the possibility of optimizing the safe margin distance surround the lesion in normal breast tissue. Methods: Clinical data of patients admitted to the Department of Thyroid and Breast Surgery, the Fourth Affiliated Hospital of Anhui Medical University (Affiliated Chaohu Hospital), from December 2021 to September 2022 were collected. A retrospective clinical study was conducted on breast cancer patients who were randomly assigned to either the ICG group or the conventional surgery group. Two to three hours before surgery, patients in the ICG group received a peripheral intravenous injection of 0.5 mg/kg ICG. Intraoperative fluorescence imaging was performed on the specimen before and after resection, as well as on the residual cavity. Near-infrared fluorescence imaging equipment was used to quantitatively measure fluorescence intensity of resected lesions at 4 directions (12, 3, 6, and 9 o'clock) and detect fluorescence in the residual cavity after lesion removal. Specimens were promptly sent to the pathology department for pathological examination, and safety margins of normal breast tissue in the 4 directions were recorded. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) checklist was followed for this study. This study was approved by the Ethics Committee of the Fourth Affiliated Hospital of Anhui Medical University (Affiliated Chaohu Hospital) (No. KYXM-202310-46). Results: This study included 50 breast cancer patients, with 24 in the ICG group and 26 in the traditional surgery group. In the ICG group, fluorescence signals were detected at all lesion sites. Specifically, fluorescence density values at the lesion center, margin, and surrounding normal breast tissue were measured as 251.08±10.73, 208.08±19.74, and 156.76±16.47, respectively, showing a gradual decrease from center outward with statistically significant differences (P<0.05). Additionally, fluorescence ratios between the lesion center and margin, and center and surrounding normal tissue, were 1.22±0.13 and 1.62±0.19, respectively. After resection, abnormal fluorescence was observed in 2 of 24 cases in the residual cavity, with 1 case being invasive carcinoma with ductal carcinoma in situ and the other normal breast tissue. Ultimately, this study demonstrated that ICG achieved a sensitivity of 95.9% and a specificity of 97.9% in margin assessment. After specimen resection, the safety margins of normal glandular tissue surrounding the lesion were measured. The safety widths for the ICG group and the concurrent breast cancer surgery group were (8.36±6.42) mm and (15.08±4.75) mm, respectively. This difference was statistically significant (P<0.05). Conclusion: ICG is a real-time, efficient, and cost-effective tracer that can be used to determine breast cancer margins, with excellent sensitivity and specificity. For early-stage breast cancer patients who are eligible for breast-conserving surgery, this tracer helps to reduce the amount of healthy breast tissue that is removed around the lesion.关键词：Indocyanine green;Breast cancer;Breast-conserving surgery;Surgical fluorescence imaging;Resection;Diagnosis;Safety range4|625|0更新时间：2025-12-31
- 摘要：Background and purpose: In the quality control of mammography, the signal-to-noise ratio (SNR) refers to the ratio of the useful signal intensity to the background noise in the image, which is one of the important indicators for measuring the quality of the image. The coefficient of variation (CoV) is a commonly used indicator to describe the consistency and repeatability of SNR. This study aimed to assess the stability and repeatability of mammographic device performance by analyzing the changes in SNR CoV in two-dimensional (2D) images and tomosynthesis images (referred to as Tomo images) under different exposure modes using three mammographic devices from different manufacturers. Methods: A polymethylmethacrylate (PMMA) phantom designed for mammography quality control was used to perform automatic exposure detection at PMMA thicknesses ranging from 20-80 mm, with actual compression thickness equivalent to the average density of the breast compressed to 21-103 mm under full-field digital mammography (FFDM), low-dose mammography and digital breast tomosynthesis (DBT) exposure modes. The CoV of SNR in 2D images and tomosynthesis images was calculated for different mammographic devices under different exposure modes and compression thicknesses. Results: Between the compression thicknesses equivalent to the average density of the breast from 21 mm to 103 mm under FFDM, low-dose mammography, and DBT exposure modes, the differences in SNR CoV of 2D images under different exposure modes among mammographic devices 1, 2, and 3 were statistically significant only in the DBT exposure mode (P=0.003), with SNR CoV ranging from 0.188% to 0.720%, 0.368% to 1.073% and 0.402% to 1.662%, respectively. There were no statistically significant differences in SNR CoV of 2D images among devices 1, 2, and 3 under FFDM and low-dose exposure modes (P=0.060). Under the DBT exposure mode, there were no statistically significant differences in the SNR CoV of the first projection image and the 0° projection image of tomosynthesis among devices 1 (2 angles), 2, and 3 (P=0.373, P=0.742, P=0.225, P=0.693, respectively). Conclusion: The SNR CoV in 2D images and tomosynthesis images varies under different mammographic devices and exposure modes, with no fixed or standard values, but all within the required range for mammographic device quality control. The stability and repeatability of 2D images of mammographic devices are better under FFDM and low-dose exposure modes; the SNR CoV values of the first projection image and the 0° projection image of tomosynthesis under the DBT exposure mode show no statistical differences, indicating good stability of the devices.关键词：Mammography;Signal-to-noise ratio;Coefficient of variation;Quality control1|434|0更新时间：2025-12-31
- 摘要：Background and purpose: Plasmacytoid dendritic cells are one of the key immune cells in the tumor microenvironment (TME), which can directly or indirectly regulate tumor related immune responses and play multiple roles in the development and metastasis of tumors. This study aimed to investigate the correlation between plasmacytoid dendritic cells in lymph nodes and lymph node metastasis in patients with advanced gastric cancer. Methods: Postoperative lymph node tissue specimens and clinicopathological data from advanced gastric cancer patients who underwent D2 radical surgery in the Department of General Surgery at the First Hospital of Dandong were gathered from January 2019 to December 2023. The lymph nodes were grouped based on pTNM staging and the diameter of metastatic lesions. This study was approved by the Medical Ethics Committee of Dandong First Hospital (No. DDSDYYY-LLSC-2025-02-18-019-01), and all patients signed informed consents. Immunohistochemical staining was used to detect the expression levels of CD123-positive plasmacytoid dendritic cells in different lymph node groups, and their correlation with lymph node metastasis in advanced gastric cancer was further analyzed. Results: Of the 116 patients, plasmacytoid dendritic cells infiltrate the lymph node tissues of gastric cancer patients. As tumor differentiation decreased and pT stage, pathological stage, lymphatic/vascular invasion, and perineural invasion increased, the mean number of CD123-positive pDC in metastatic lymph nodes rose significantly (P<0.05). The number of CD123-positive plasmacytoid dendritic cells was higher in metastatic lymph node-positive tissues than in metastatic lymph node-negative tissues, the number was higher in the macrometastasis group of pN1-3 staging than in the micrometastasis group, and the number was higher in the non-metastatic lymph node group of pN1-3 staging than in the pN0 staging lymph node group (P<0.05). In lymph node metastasis-positive cases, the number of CD123-positive plasmacytoid dendritic cells was higher in second-station lymph nodes than in first-station lymph nodes (P<0.05). Conclusion: The infiltration of CD123-positive plasmacytoid dendritic cells in lymph nodes of patients with advanced gastric cancer is closely associated with lymph node metastasis and may serve as a prerequisite for metastatic spread. Understanding the distribution of CD123-positive plasmacytoid dendritic cells in gastric cancer lymph nodes can help further explore their role in the immune microenvironment of gastric cancer. Targeted therapy focusing on CD123-positive plasmacytoid dendritic cells may become a new strategy for the treatment of advanced gastric cancer.关键词：Plasmacytoid dendritic cell;CD123;Advanced gastric cancer;Lymph node metastasis;pTNM staging6|259|0更新时间：2025-12-31
Article
- 摘要：Breast cancer stands as the most prevalent malignancy and the primary cause of cancer-related mortality among women globally. The lymph node status is not only pivotal for accurate clinical staging of breast cancer but also significantly associated with patients’ prognosis. Magnetic resonance imaging (MRI) has advantages in evaluating lymph nodes status and the response effect of neoadjuvant therapy, serving as a valuable complement to other imaging modalities. Standardized scoring systems, such as Node Reporting and Data System (Node-RADS), integrate key features including lymph node size, margin characteristics, and enhancement patterns, effectively minimizing interobserver variability in evaluation. MRI radiomics, by extracting quantitative features at high throughput, converts medical images into mineable and analyzable data. Further integrating MRI radiomics, clinicopathological features and molecular subtype information to construct multi-omics models, can effectively predict axillary lymph node metastasis, thereby providing a biological basis for personalized treatment. Artificial intelligence (AI) leverages extensive search algorithms and parameter spaces to generate predictive models. AI-driven MRI analysis has proven effective in predicting lymph node metastasis and treatment responses. In the evaluation of neoadjuvant chemotherapy, the fully automated-integrated system based on deep learning (FAIS-DL) system, which combines multi-region dynamic contrast enhanced-MRI (DCE-MRI) and clinical data, can efficiently predict axillary pathological complete response. This innovation has substantially reduced the rate of unnecessary axillary lymph node dissection (ALND) from 47.9% to 6.8%. This article reviewed the prediction of lymph node status in breast cancer by MRI at different developmental stages, with the aim of enhancing the understanding of clinicians and radiologists regarding the application of MRI in the assessment of lymph node status in breast cancer and evaluating the efficacy of neoadjuvant therapy, and providing assistance for the construction of a model for accurately predicting lymph node status in breast cancer.关键词：Breast cancer;Neoadjuvant therapy;Lymph nodes;Magnetic resonance imaging;Radiomics;Artificial intelligence10|1235|0更新时间：2025-12-31
- 摘要：Radiotherapy-induced late effects (RILE) or delayed reactions have significant impacts on patients' long-term quality of life. However, current understanding of their developmental mechanisms remains limited, with a lack of effective risk prediction models and preventive interventions. Radiobiological studies and radiation-induced senescence models have revealed that radiotherapy can alter the epigenetic characteristics in late-responding tissue cells, leading to derepression of retrotransposable elements (particularly endogenous retroviral elements), subsequent activation of cytoplasmic nucleic acid sensor systems (cGAS-STING, MDA5/RIG-I-MAVS) and type Ⅰ interferon-mediated immune-inflammatory responses. This review summarized relevant research findings, proposing that the autoimmune-like inflammatory response induced by the 'radiotherapy-epigenetic alteration-retrotransposable element activation’ cascade is an underinvestigated mechanistic basis in the development of RILE. Constructing risk prediction models for late effects based on epigenetic signatures, cell type, and radiation dose, along with developing strategies to epigenetically suppress retrotransposable element expression, holds promise for preventing or mitigating RILE.关键词：Radiotherapy;Late effect;Late-responding tissues;Retrotransposable elements;Autoimmune inflammation6|1662|0更新时间：2025-12-31
Review
- Technical support is provided by Beijing Founder electronics co., LTD 京ICP备09064830号-19
京公网安备11010802024621 - It is recommended to read the content of this site in Chrome&IE9+. Please switch to extreme mode in browser 360.
- Cookies We use cookies to help provide and enhance our service and tailor content. By continuing, you agree to the use of cookies.
0