BMC Cancer最新文献

{"title":"Construction and validation of a novel liver function-tumor burden-inflammation-nutrition (LTIN) score for HCC patients underwent hepatectomy.","authors":"Yuhao Su, Yuxin Liang, Deyuan Zhong, Hongtao Yan, Qinyan Yang, Jin Shang, Yahui Chen, Xiaolun Huang","doi":"10.1186/s12885-025-13867-w","DOIUrl":"10.1186/s12885-025-13867-w","url":null,"abstract":"<p><strong>Objective: </strong>Liver function, tumor burden, inflammation level, and nutritional status are critical factors influencing tumor onset, progression, and metastasis. This study sought to investigate the prognostic significance and clinical relevance of biomarkers associated with these factors to develop a novel liver function-tumor burden-inflammation-nutrition (LTIN) score for patients with hepatocellular carcinoma (HCC) who received hepatectomy.</p><p><strong>Methods: </strong>A retrospective analysis was conducted on 285 patients with HCC undergoing hepatectomy at two medical centers between July 2019 and July 2023. The patients were divided into a training set (n = 200) and a validation set (n = 85). The study evaluated the prognostic significance of eight relevant clinical indicators and developed an LTIN score using Least Absolute Shrinkage and Selection Operator (LASSO) regression. Kaplan-Meier survival curves and multivariate Cox regression analysis were utilized to determine the prognostic value of the LTIN score. Time-dependent receiver operating characteristic (ROC) analyses were used to compare the predictive performance of various prognostic factors.</p><p><strong>Results: </strong>The LTIN score, derived from the albumin-bilirubin (ALBI) grade, tumor burden score (TBS), prognostic nutritional index (PNI), and prognostic inflammatory index (PII), effectively classified patients into high- and low-risk groups based on the optimal cut-off value. Patients with low-risk scores exhibited significantly better overall survival (OS) and recurrence-free survival (RFS) than those with high-risk groups in both the training and validation sets (P < 0.001). Furthermore, the LTIN score was identified as a significant independent prognostic factor for both OS (P < 0.001) and RFS (P < 0.001). The LTIN score also exhibited superior prognostic capabilities compared to the other indicators, Tumor-Node-Metastasis (TNM) staging system, and Barcelona Clinic Liver Cancer (BCLC) staging system.</p><p><strong>Conclusion: </strong>Our findings indicated that the preoperative LTIN score has significant potential as a reliable predictor of OS and RFS for HCC patients underwent radical surgery. The LTIN score could further effectively guide treatment decisions and optimize follow-up strategies to enhance patients prognosis.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"504"},"PeriodicalIF":3.4,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11921615/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143662555","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Selumetinib in Combination with Anti Retroviral Therapy in HIV-associated Kaposi sarcoma (SCART): an open-label, multicentre, phase I/II trial.","authors":"Robin J Young, Amanda Kirkham, Joshua Savage, Charlotte Gaskell, Sarah Johnson, David H Dockrell, Mark Bower, Sarah Westwell, Christine Bowman, Michael Leahy, Penella Woll, Lucinda Billingham","doi":"10.1186/s12885-025-13890-x","DOIUrl":"10.1186/s12885-025-13890-x","url":null,"abstract":"<p><strong>Background: </strong>Kaposi sarcoma (KS) is the commonest HIV-associated malignancy. It is caused by co-infection with Kaposi sarcoma herpesvirus (KSHV), which upregulates the MAPK pathway. The aim of the SCART trial was to identify a safe dose for the MEK inhibitor selumetinib in combination with antiretroviral therapy (ART) and to establish evidence of the combination's efficacy.</p><p><strong>Methods: </strong>SCART was a prospective, single arm, open-label, multi-centre, phase I/II trial, recruiting from four UK centres. Eligible patients were HIV positive, established on an ART regimen ≥ 3 months, had HIV viral load ≤ 200/ml, and had histologically confirmed KS with progressive disease. Phase I primary outcomes were occurrence of dose limiting toxicity (DLT) to determine the maximum tolerated dose/recommended phase II dose (RP2D), and pharmacokinetic assessments of selumetinib and N-desmethyl metabolite. Phase II primary outcome was occurrence of objective response (OR) as defined by AIDS Clinical Trials Group (ACTG) criteria.</p><p><strong>Results: </strong>Between 15-Jun-2012 and 25-Sep-2018, 19 patients were recruited; three did not start treatment and were not included in the final analysis. Ten eligible patients were treated in phase I and an additional six in phase II. There was one DLT at the 75 mg bd dose, which was deemed to be the RP2D. Of those patients receiving the RP2D (six within phase I, six within phase II), one achieved a partial response (OR 8.3%, 90% confidence interval: 0.4, 33.9). Further to the DLT, two serious adverse reactions, one unrelated serious adverse event (AE), and six non-serious grade 3 AEs were reported, together with 360 AEs graded 1 or 2. No detrimental impact on ART drug levels or HIV viral load were observed, with improvements in CD4 count and evidence of response in Angiopoietin-2 demonstrated.</p><p><strong>Conclusions: </strong>SCART was closed early due to slow recruitment, partly due to the rarity of KS because of improvements in HIV care, but also due to patients' concerns about experiencing non-serious toxicity additional to those from ART. Although we cannot recommend the use of 75 mg bd selumetinib with ART in patients with HIV-associated KS, studies exploring selumetinib in combination with other agents including anti-angiogenic agents and/or immune checkpoint inhibitors are warranted.</p><p><strong>Trial registration: </strong>ISRCTN24921472.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"505"},"PeriodicalIF":3.4,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11921695/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143662562","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Fc fragment of IgG binding protein suppresses tumor growth by stabilizing wild type P53 in colorectal cancer cells.","authors":"Jiefu Wang, Ziqing Gong, Jia Liu, Wenpeng Wang, Kai Liu, Yanpeng Yang, Xinran Lu, Junfeng Wang","doi":"10.1186/s12885-025-13873-y","DOIUrl":"10.1186/s12885-025-13873-y","url":null,"abstract":"<p><p>The Fc fragment of IgG binding protein (FCGBP) exhibits differential expression across various tumor types. but its role in cancer progression remains underexplored. This research discovered that FCGBP is downregulated in colorectal cancer (CRC) cells and is negatively associated with poor prognosis. Overexpression of FCGBP inhibited the growth of P53 wild-type CRC cells both in vitro and in vivo. Mechanistically, immunoprecipitation experiments revealed that FCGBP competitively binds to MDM2, thereby attenuating the formation of the P53/MDM2 complex. This, in turn, reduces P53 ubiquitination and stabilizes the protein. Our findings reveal a novel mechanism through which FCGBP significantly inhibits CRC cell growth and propose a new targeted therapeutic strategy for CRC treatment.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"507"},"PeriodicalIF":3.4,"publicationDate":"2025-03-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11924790/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143662557","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Single-cell transcriptomic analysis reveals CD8 + T cell heterogeneity and identifies a prognostic signature in cervical cancer.","authors":"Rongbin Zhou, Yuli Xie, Zuheng Wang, Zige Liu, Wenhao Lu, Xiao Li, Chunmeng Wei, Xing Li, Fubo Wang","doi":"10.1186/s12885-025-13901-x","DOIUrl":"10.1186/s12885-025-13901-x","url":null,"abstract":"<p><strong>Background: </strong>In recent years, immunotherapy has made significant progress. However, the understanding of the heterogeneity and function of T cells, particularly CD8 + T cells, in cervical cancer (CESC) microenvironment remains insufficient. We aim to characterize the heterogeneity, developmental trajectory, regulatory network, and intercellular communication of CD8 + T cells in cervical squamous cell carcinoma and to construct a prognostic risk model based on the transcriptomic characteristics of CD8 + T cells.</p><p><strong>Methods: </strong>We integrated single-cell RNA sequencing data from CESC tumor samples with bulk transcriptome data from TCGA and GEO databases. We identified CD8 + T cell subsets in the CESC microenvironment, revealing significant interactions between CD8 + T cells and other cell types through intercellular communication analysis. Pseudotime trajectory analysis revealed dynamic transcriptional regulation during CD8 + T cell differentiation and functional acquisition processes. We constructed a transcriptional regulatory network for CESC CD8 + T cells, identifying key transcription factors. Based on CD8 + T cell-related genes, a prognostic risk model comprising eight core genes was developed and validated using machine learning.</p><p><strong>Results: </strong>We identified four distinct CD8 + T cell subsets, namely progenitor, intermediate, proliferative, and terminally differentiated, each exhibiting unique transcriptomic characteristics and functional properties. CD8 + T cell subsets interact with macrophages through different ligand-receptor networks, including the CCL-CCR signaling pathway and costimulatory molecules. Sorafenib was identified as a potential immunotherapeutic drug through drug screening. Experimental validation demonstrated that sorafenib enhances the cytotoxicity of CD8 + T cells by increasing the secretion of IFN-γ and TNF-α, thereby significantly inhibiting the invasiveness and survival of CESC cells.</p><p><strong>Conclusions: </strong>Our study provides valuable insights into the heterogeneity and functional diversity of CD8 + T cells in CESC. We demonstrate that a CD8 + T cell-related prognostic signature may serve as a potential tool for risk stratification in patients with CESC. Additionally, our finding suggests that sorafenib could be a promising therapeutic candidate for improving antitumor immunity in this patient population.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"498"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11916872/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143656117","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Immune modulation in solid tumors: a phase 1b study of RO6870810 (BET inhibitor) and atezolizumab (PD-L1 inhibitor).","authors":"Daniel Marbach, Jurriaan Brouer-Visser, Laura Brennan, Sabine Wilson, Iakov I Davydov, Nicolas Staedler, José Duarte, Iris Martinez Quetglas, Eveline Nüesch, Marta Cañamero, Evelyne Chesné, George Au-Yeung, Erika Hamilton, Stephanie Lheureux, Debra L Richardson, Iben Spanggaard, Bruno Gomes, Izolda Franjkovic, Mark DeMario, Martin Kornacker, Katharina Lechner","doi":"10.1186/s12885-025-13851-4","DOIUrl":"10.1186/s12885-025-13851-4","url":null,"abstract":"<p><strong>Purpose: </strong>Bromodomain and extra-terminal domain (BET) inhibitors (BETi) have demonstrated epigenetic modulation capabilities, specifically in transcriptional repression of oncogenic pathways. Preclinical assays suggest that BETi potentially attenuates the PD1/PD-L1 immune checkpoint axis, supporting its combination with immunomodulatory agents.</p><p><strong>Patients and methods: </strong>A Phase 1b clinical trial was conducted to elucidate the pharmacokinetic and pharmacodynamic profiles of the BET inhibitor RO6870810 as monotherapy and in combination with the PD-L1 antagonist atezolizumab in patients with advanced ovarian carcinomas and triple-negative breast cancer (TNBC). Endpoints included maximum tolerated dosages, adverse event profiling, pharmacokinetic evaluations, and antitumor activity. Pharmacodynamic and immunomodulatory effects were assessed in tumor tissue (by immunohistochemistry and RNA-seq) and in peripheral blood (by flow cytometry and cytokine analysis).</p><p><strong>Results: </strong>The study was terminated prematurely due to a pronounced incidence of immune-related adverse effects in patients receiving combination of RO6870810 and atezolizumab. Antitumor activity was limited to 2 patients (5.6%) showing partial response. Although target engagement was confirmed by established BETi pharmacodynamic markers in both blood and tumor samples, BETi failed to markedly decrease tumor PD-L1 expression and had a suppressive effect on antitumor immunity. Immune effector activation in tumor tissue was solely observed with the atezolizumab combination, aligning with this checkpoint inhibitor's recognized biological effects.</p><p><strong>Conclusions: </strong>The combination of BET inhibitor RO6870810 with the checkpoint inhibitor atezolizumab presents an unfavorable risk-benefit profile for ovarian cancer and TNBC (triple-negative breast cancer) patients due to the increased risk of augmented or exaggerated immune reactions, without evidence for synergistic antitumor effects.</p><p><strong>Trial registration: </strong>ClinicalTrials.gov ID NCT03292172; Registration Date: 2017-09-25.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"500"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11916277/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143655987","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Deep learning based on intratumoral heterogeneity predicts histopathologic grade of hepatocellular carcinoma.","authors":"Shaoming Song, Gong Zhang, Zhiyuan Yao, Ruiqiu Chen, Kai Liu, Tianchen Zhang, Guineng Zeng, Zizheng Wang, Rong Liu","doi":"10.1186/s12885-025-13781-1","DOIUrl":"10.1186/s12885-025-13781-1","url":null,"abstract":"<p><strong>Objectives: </strong>The potential of medical imaging to non-invasively assess intratumoral heterogeneity (ITH) is increasingly being recognized. This study aimed to investigate the value of the ITH-based deep learning model for preoperative prediction of histopathologic grade in hepatocellular carcinoma (HCC).</p><p><strong>Materials and methods: </strong>A total of 858 patients from primary cohort and two external cohorts were included. 3.0T or 1.5T axial portal venous phase MRI images were collected. We conducted radiomics feature-driven K-means clustering for automatic partition to reveal ITH. 2.5D and 3D deep learning models based on ResNet architecture were trained to extract deep learning hidden features of each subregion. The selected features were used to train Random Forest classifier, which constructed the feature-fusion model.</p><p><strong>Results: </strong>The extracted voxel-level radiomics features were unsupervised clustered by K-means to generate three subregions. In the 2.5D deep learning, the feature-fusion model based on ITH had superior predictive efficacy than the whole-tumor model (AUC: 0.82 vs. 0.72; p = 0.004). Even in the validation and external test sets, this model maintained a high AUC of 0.78-0.83, and net reclassification indices indicated that it could improve prediction by 25-28%. Regarding the prognostic value, overall survival (OS) and recurrence-free survival (RFS) could be significantly stratified by the 2.5D feature-fusion model, and multivariable Cox regressions indicated its signature was identified as a risk predictor for OS and RFS (p < 0.05).</p><p><strong>Conclusion: </strong>The ITH-based feature-fusion model provided a non-invasive method for classifying tumor differentiation in HCC, which may serve as a promising strategy for stratification management.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"497"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11917083/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143656009","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Establishment of CD34 + hematopoietic stem cell-derived xenograft model of hyperleukocytic acute myeloid leukemia.","authors":"Yanxia Jin, Yuxing Liang, Balu Wu, Sanyun Wu, Xiaoyan Liu, Fuling Zhou","doi":"10.1186/s12885-025-13907-5","DOIUrl":"10.1186/s12885-025-13907-5","url":null,"abstract":"<p><strong>Background: </strong>Hyperleukocytic acute myeloid leukemia (HLL) is marked by high early mortality and presents significant therapeutic challenges. Research on HLL is still in its infancy, and comprehensive development of patient-derived xenograft (PDX) models, especially CD34 + hematopoietic stem cell-derived models, remains limited.</p><p><strong>Methods: </strong>We evaluated the establishment of the HLL model through blood examinations, smear analysis, bone marrow biopsy, flow cytometry, and mutation analysis. Correlation between survival times in mice and patients was assessed using linear regression.</p><p><strong>Results: </strong>In the HLL PDX mouse model, leukocyte counts could reach up to 37.35^10⁹/L, and immunophenotyping revealed the presence of hCD45+, hCD15+, and hCD33 + cells in both peripheral blood (PB) and bone marrow (BM) following inoculation with PB-derived cells for the establishment of the HLL PDX model. Similar results were observed with cells derived from the patient's BM. In the CD34 + hematopoietic stem cell-derived xenograft model, extensive infiltration of CD34 + cells into the BM, liver, and spleen was observed. Additionally, human WT1 and NRAS mutations were identified in the liver, spleen, and BM of the mice. A comparative analysis of multiple experiments revealed that shorter survival times were observed in mice receiving a higher irradiation dose of 2.5 Gy and a greater number of cells derived from PB. Additionally, shorter survival times were observed in model mice injected with cells carrying NRAS, DNMT3A, FLT3, or NPM1 gene mutations. Correlation analysis indicated that the survival times of the mice were significantly associated with the survival status of the patients.</p><p><strong>Conclusions: </strong>We successfully established a CD34 + hematopoietic stem cell-derived xenograft model of HLL, providing a valuable tool for mechanistic research, drug screening, individualized therapy, and precision medicine.</p><p><strong>Trial registration: </strong>Not application.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"499"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11917077/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143655945","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Prognostic evaluation and treatment strategies for cervical cancer in pregnancy: a systematic review and meta-analysis.","authors":"Siyuan Zeng, Xin Li, Simin Xiao, Peina Yang, Changsheng Lin, Huiling Chen, Hu Zhao, Xue Xiao","doi":"10.1186/s12885-025-13827-4","DOIUrl":"10.1186/s12885-025-13827-4","url":null,"abstract":"<p><strong>Abtstrac: </strong>OBJECTIVE: This study was conducted to evaluate the prognosis of cervical cancer in pregnancy (CCIP) and analyze the clinicopathological factors affecting the prognosis of this cancer.</p><p><strong>Data sources: </strong>The studies published through July 2024 were systematically retrieved from PubMed, Embase, Web of Science, and Cochrane Library.</p><p><strong>Study eligibility criteria: </strong>The cohort studies, case-control studies, randomized controlled trials, and non-randomized controlled trials involving CCIP patients with data on 5-year overall survival (OS) were included in this study.</p><p><strong>Study appraisal and synthesis methods: </strong>The quality of the included studies was assessed using the Newcastle-Ottawa Scale (NOS). A meta-analysis was performed using Stata 15.0, focusing on the 5-year OS and relevant clinicopathological factors.</p><p><strong>Results: </strong>The results demonstrated that the 5-year OS of patients with CCIP was similar to that of non-pregnant patients with cervical cancer (RR = 1.00, 95% CI: 0.94-1.06, P = 0.978). The subgroup analysis results revealed that tumor size (≥ 4 cm), International Federation of Gynecology and Obstetrics (FIGO) stage (≥ IB2), and timing of diagnosis (postpartum) were prognostic factors with statistical significance (P < 0.05). However, such factors as pregnancy termination and timing of delivery did not significantly affect the 5-year OS (P > 0.05). The delivery mode required further validation despite its borderline significance (P = 0.05).</p><p><strong>Conclusion: </strong>The results of this study suggest that pregnancy does not exert a significant adverse effect on the long-term survival of patients with cervical cancer. Tumor size (≥ 4 cm), FIGO stage (≥ IB2), and time of diagnosis (postpartum) are identified as unfavorable prognostic factors for CCIP patients, while delivery mode requires further investigation. These findings provide strong evidence to support the optimization of personalized treatment strategies for CCIP patients.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"502"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11917149/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143656051","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"A semi-supervised convolutional neural network for diagnosis of pancreatic ductal adenocarcinoma based on EUS-FNA cytological images.","authors":"Dong Fang, Yigeng Huang, Suwen Li, Chen Shi, Junjun Bao, Dandan Du, Lanlan Xuan, Leping Ye, Yanping Zhang, ChengLin Zhu, Hailun Zheng, Zhenwang Shi, Qiao Mei, Huanqin Wang","doi":"10.1186/s12885-025-13910-w","DOIUrl":"10.1186/s12885-025-13910-w","url":null,"abstract":"<p><strong>Background: </strong>The cytological diagnostic process of EUS-FNA smears is time-consuming and manpower-intensive, and the conclusion could be subjective and controversial. Moreover, the relative lack of cytopathologists has limited the widespread implementation of Rapid on-site evaluation (ROSE) presently. Therefore, this study aimed to establish an AI system for the detection of pancreatic ductal adenocarcinoma (PDAC) based on EUS-FNA cytological images.</p><p><strong>Methods: </strong>We collected 3213 unified magnification images of pancreatic cell clusters from 210 pancreatic mass patients who underwent EUS-FNA in four hospitals. A semi-supervised CNN (SSCNN) system was developed to distinguish PDAC from Non-PDAC. The internal and external verifications were adopted and the diagnostic accuracy was compared between different seniorities of cytopathologists. 33 images of \"Atypical\" diagnosed by expert cytopathologists were selected to analyze the consistency between the system and definitive diagnosis.</p><p><strong>Results: </strong>The segmentation indicators Mean Intersection over Union (mIou), precision, recall and F1-score of SSCNN in internal and external testing sets were 88.3%, 93.21%,94.24%, 93.68% and 87.75%, 93.81%, 93.14%, 93.48% successively. The PDAC classification indicators of the SSCNN model including area under the ROC curve (AUC), accuracy, sensitivity, specificity, PPV and NPV in the internal testing set were 0.97%, 0.95%, 0.94%, 0.97%, 0.98%, 0.91% respectively, and 0.99%, 0.94%, 0.94%, 0.95%, 0.99%, 0.75% correspondingly in the external testing set. The diagnostic accuracy of senior, intermediate and junior cytopathologists was 95.00%, 88.33% and 76.67% under the binary diagnostic criteria of PDAC and non-PDAC. In comparison, the accuracy of the SSCNN system was 90.00% in the dataset of man-machine competition. The accuracy of the SSCNN model was highly consistent with senior cytopathologists (Kappa = 0.853, P = 0.001). The accuracy, sensitivity and specificity of the system in the classification of \"atypical\" cases were 78.79%, 84.20% and 71.43% respectively.</p><p><strong>Conclusion: </strong>Not merely tremendous preparatory work was drastically reduced, the semi-supervised CNN model could effectively identify PDAC cell clusters in EUS-FNA cytological smears which achieved analogically diagnostic capability compared with senior cytopathologists, and showed outstanding performance in assisting to categorize \"atypical\" cases where manual diagnosis is controversial.</p><p><strong>Trial registration: </strong>This study was registered on clinicaltrials.gov, and its unique Protocol ID was PJ-2018-12-17.</p>","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"495"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11917044/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143655755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

{"title":"Correction: Exposure to major coronary heart disease events reduces lung cancer risk: a Mendelian randomization study based on a European population.","authors":"Dongming He, Hongting Lu, Xinhuai Ou, Tiaozhan Zheng, Zhiwen Zheng, Zhanyu Xu, Xiaohong Duan, Shikang Li","doi":"10.1186/s12885-025-13916-4","DOIUrl":"10.1186/s12885-025-13916-4","url":null,"abstract":"","PeriodicalId":9131,"journal":{"name":"BMC Cancer","volume":"25 1","pages":"501"},"PeriodicalIF":3.4,"publicationDate":"2025-03-18","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC11916949/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"143655814","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"医学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}