Best Pract Res Clin Gastroenterol. 2025 Mar;75:102004. doi: 10.1016/j.bpg.2025.102004. Epub 2025 Mar 28.

ABSTRACT

Esophageal squamous cell carcinoma (ESCC) remains a significant global health burden with high incidence and mortality rates, particularly in developing regions. Early detection is crucial for improving patient survival, yet conventional screening methods such as endoscopy and non-endoscopic techniques face limitations in accuracy, cost, and dependency on clinician expertise. This review explores the transformative role of artificial intelligence (AI) in ESCC screening. AI technologies, including machine learning, deep learning, and transfer learning, demonstrate remarkable potential for early ESCC screening by targeting high-risk populations, optimizing screening modalities, refining screening intervals, and enhancing cost-effectiveness. AI-driven systems improve lesion detection, vascular pattern recognition, and risk prediction by integrating imaging, genomic, and clinical data. Additionally, AI applications in liquid biopsy analysis enable non-invasive detection of circulating tumor cells and DNA, further advancing early diagnosis. Despite these advancements, challenges such as dataset variability, model generalizability, algorithm transparency, and ethical and legal concerns require resolution to fully harness AI's capabilities. This paper highlights the current applications, persistent challenges, and future directions for AI in revolutionizing ESCC screening.

PMID:40451647 | DOI:10.1016/j.bpg.2025.102004

J Dent. 2025 May 30:105867. doi: 10.1016/j.jdent.2025.105867. Online ahead of print.

ABSTRACT

OBJECTIVE: Artificial intelligence (AI) is increasingly used in dental research for diagnosis, treatment planning, and disease prediction. However, many dental AI studies lack methodological rigor, transparency, or reproducibility, and no dedicated peer-review guidance exists for this field.

METHODS: Editors and reviewers from the ITU/WHO/WIPO AI for Health - Dentistry group participated in a structured survey and group discussions to identify key elements for reviewing AI dental research. A draft of the recommendations was circulated for feedback and consensus.

RESULTS: The consensus from editors and reviewers identified four key indicators of high-quality AI dental research: (1) relevance to a real clinical or methodological problem, (2) robust and transparent methodology, (3) reproducibility through data/code availability or functional demos, and (4) adherence to ethical and responsible reporting practices. Common reasons for rejection included lack of novelty, poor methodology, limited external testing, and overstated claims. Four essential checks were proposed to support peer review: the study should address a meaningful clinical question, follow appropriate reporting guidelines (e.g., DENTAL-AI, STARD-AI), clearly describe reproducible methods, and use precise, justified, and clinically relevant wording.

CONCLUSION: Editors and reviewers play a critical role in improving the quality of AI research in dentistry. This guidance aims to support more robust peer review and contribute to the development of reliable, clinically relevant, and ethically sound AI applications in dentistry.

PMID:40451605 | DOI:10.1016/j.jdent.2025.105867

Int J Pharm. 2025 May 30:125789. doi: 10.1016/j.ijpharm.2025.125789. Online ahead of print.

ABSTRACT

Integrating artificial intelligence (AI) into drug discovery has revolutionized pharmaceutical innovation, addressing the challenges of traditional methods that are costly, time-consuming, and suffer from high failure rates. By utilizing machine learning (ML), deep learning (DL), and natural language processing (NLP), AI enhances various stages of drug development, including target identification, lead optimization, de novo drug design, and drug repurposing. AI tools, such as AlphaFold for protein structure prediction and AtomNet for structure-based drug design, have significantly accelerated the discovery process, improved efficiency and reduced costs. Success stories like Insilico Medicine's AI-designed molecule for idiopathic pulmonary fibrosis and BenevolentAI's identification of baricitinib for COVID-19 highlight AI's transformative potential. Additionally, AI enables the exploration of vast chemical spaces, optimization of clinical trials, and the identification of novel therapeutic targets, paving the way for precision medicine. However, challenges such as limited data accessibility, integration of diverse datasets, interpretability of AI models, and ethical concerns remain critical hurdles. Overcoming these limitations through enhanced algorithms, standardized databases, and interdisciplinary collaboration is essential. Overall, AI continues to reshape drug discovery, reducing timelines, increasing success rates, and driving the development of innovative and accessible therapies for unmet medical needs.

PMID:40451590 | DOI:10.1016/j.ijpharm.2025.125789

BMJ Open Ophthalmol. 2025 Jun 1;10(1):e002176. doi: 10.1136/bmjophth-2025-002176.

ABSTRACT

PURPOSE: Age-related macular degeneration (AMD) remains the leading cause of blindness in developed countries. There are many different intravitreal anti-vascular endothelial growth factor (VEGF) drugs available for the treatment of neovascular AMD (nAMD). Unfortunately, not all patients respond equally well to the drugs, and some show recurrences during treatment. Since 01/2024, aflibercept 8 mg represents an additional treatment option and contains a four times higher dosage than the already known aflibercept 2 mg.

METHODS: To evaluate the real-world efficacy of aflibercept 8 mg in refractory nAMD patients, focusing on changes in key optical coherence tomography biomarkers over a follow-up period of the first four aflibercept 8 mg injections using a deep learning-based semantic segmentation algorithm. Inclusion criteria were: switch to aflibercept 8 mg after insufficient response to aflibercept 2 mg, marked by persistent retinal fluid or inability to extend treatment beyond 6 weeks; completion of at least 3 months (90 days) follow-up under treat-and-extend treatment regime; and no confounding conditions like intraocular infection, uveitis or other retinal diseases.

RESULTS: 23 eyes of 21 patients with therapy-resistant nAMD were switched to aflibercept 8 mg. All patients had previously received aflibercept 2 mg, with an average of 30.7 previous anti-VEGF injections. Significant reductions in intraretinal fluid and fibrovascular pigment epithelial detachment at timepoint V3 were observed. The decrease in subretinal fluid and central retinal thickness at V3 was not significant. Treatment intervals extended significantly by 24%, from a baseline average of 34 days to 42 days. Best-corrected visual acuity remained stable throughout the study period.

CONCLUSIONS: Aflibercept 8 mg demonstrated significant efficacy and durability in reducing nAMD biomarkers and extending intervals in a real-world setting. The use of deep learning for biomarker quantification highlighted its potential for enhancing treatment monitoring and decision-making. Future studies with a larger patient cohort and prospective study setting should explore long-term outcomes and integration of artificial intelligence-driven analysis.

PMID:40451292 | DOI:10.1136/bmjophth-2025-002176

Can J Ophthalmol. 2025 May 29:S0008-4182(25)00248-0. doi: 10.1016/j.jcjo.2025.05.002. Online ahead of print.

ABSTRACT

OBJECTIVE: Microperimetry (MP) has emerged as a clinical functional endpoint in nonexudative age-related macular degeneration (AMD). In this study, we aim to provide reference values for test-retest outcomes on two MP devices in intermediate AMD (iAMD).

DESIGN: Prospective, cross-sectional study.

PARTICIPANTS: 3 600 stimuli from 20 eyes in 20 subjects.

METHODS: Patients diagnosed with iAMD underwent consecutive testing on MP-3 (NIKED, Gamagori, Japan) and MAIA (CenterVue Icare, Padova, Italy). The obtained point-wise sensitivity (PWS) measurements were superimposed with optical coherence tomography (OCT) (Spectralis, Heildelberg Engineering) acquired. Hyperreflective foci (HRF), drusen volume, ellipsoid zone (EZ)-thickness and outer nuclear layer (ONL)-thickness were quantified with deep-learning algorithms. Subretinal drusenoid deposits (SDD) were manually annotated. We assessed test-retest repeatability at the location of these biomarkers using Bland-Altmann coefficients of repeatability. Furthermore, interdevice correlation, fixation stabilities, and examination durations were evaluated.

RESULTS: Comparable overall point-wise retest variances were detected for MP-3 (±4.54 dB) and MAIA (±5.24 dB). SDDs led to significantly worse repeatability in the MAIA device (p = 0.03). Drusen, HRF, EZ-thickness, and ONL thickness had no significant impact on test-retest variance. A good intradevice correlation (MP-3: 0.869 [0.851 - 0.886] MAIA 0.848 [0.827 - 0.867]), and a good mean interdevice correlation (0.841 [0.819 - 0.861]) was observed.

CONCLUSIONS: Intradevice and interdevice repeatability for MP examinations with MP-3 and MAIA in patients with iAMD can be considered as good. Biomarkers except for SDD show no significant impact in repeatability in both devices. This supports MP as a reliable functional endpoint in clinical trials in iAMD.

PMID:40451257 | DOI:10.1016/j.jcjo.2025.05.002

Artif Intell Med. 2025 May 27;167:103166. doi: 10.1016/j.artmed.2025.103166. Online ahead of print.

ABSTRACT

BACKGROUND: Occupational health assessment is critical for detecting respiratory issues caused by harmful exposures, such as cement dust. Quantitative computed tomography (QCT) imaging provides detailed insights into lung structure and function, enhancing the diagnosis of lung diseases. However, its high dimensionality poses challenges for traditional machine learning methods.

METHODS: In this study, Kolmogorov-Arnold networks (KANs) were used for the binary classification of QCT imaging data to assess respiratory conditions associated with cement dust exposure. The dataset comprised QCT images from 609 individuals, including 311 subjects exposed to cement dust and 298 healthy controls. We derived 141 QCT-based variables and employed KANs with two hidden layers of 15 and 8 neurons. The network parameters, including grid intervals, polynomial order, learning rate, and penalty strengths, were carefully fine-tuned. The performance of the model was assessed through various metrics, including accuracy, precision, recall, F1 score, specificity, and the Matthews Correlation Coefficient (MCC). A five-fold cross-validation was employed to enhance the robustness of the evaluation. SHAP analysis was applied to interpret the sensitive QCT features.

RESULTS: The KAN model demonstrated consistently high performance across all metrics, with an average accuracy of 98.03 %, precision of 97.35 %, recall of 98.70 %, F1 score of 98.01 %, and specificity of 97.40 %. The MCC value further confirmed the robustness of the model in managing imbalanced datasets. The comparative analysis demonstrated that the KAN model outperformed traditional methods and other deep learning approaches, such as TabPFN, ANN, FT-Transformer, VGG19, MobileNets, ResNet101, XGBoost, SVM, random forest, and decision tree. SHAP analysis highlighted structural and functional lung features, such as airway geometry, wall thickness, and lung volume, as key predictors.

CONCLUSION: KANs significantly improved the classification of QCT imaging data, enhancing early detection of cement dust-induced respiratory conditions. SHAP analysis supported model interpretability, enhancing its potential for clinical translation in occupational health assessments.

PMID:40450965 | DOI:10.1016/j.artmed.2025.103166

J Environ Manage. 2025 May 31;388:125858. doi: 10.1016/j.jenvman.2025.125858. Online ahead of print.

ABSTRACT

The compounded effect of heavy rainfall and high tide backwater significantly exacerbate the load on urban drainage systems in coastal cities, leading to an escalating risk of compound flood disasters. The formation mechanism of compound floods is of great complexity, and the research concerning it constitutes a highly challenging subject. While deep learning (DL) techniques have been increasingly applied in flood forecasting, their "black-box" nature often obscures the internal decision-making logic, limiting insights into the mechanisms driving compound flooding. To address this, our study proposes an explainable artificial intelligence (XAI) framework, utilizing a Long Short-Term Memory (LSTM) network integrated with a Multi-Head Attention (MHA) mechanism as a surrogate model for urban flood simulation. The SHapley Additive exPlanations (SHAP) method is employed to elucidate the model's decision-making process, uncovering critical driving factors and their interactions in compound flooding scenarios. Results demonstrate that the MHA mechanism enhances the model's ability to capture rainfall-tide interactions, with the LSTM-MHA model outperforming data-driven baseline models and achieving performance slightly below physics-based models, as evidenced by an R2 of 0.971, MAE of 0.040 m, and RMSE of 0.065 m. Furthermore, the LSTM-MHA model significantly improves computational efficiency, completing simulations 216 times faster than traditional physics-based models in the study case. SHAP analysis reveals consistent trends across typical scenarios, highlighting the dominant roles of rainfall and tidal factors across spatiotemporal scales and validating the surrogate model's decision-making rationality. By integrating XAI with SHAP, this study enhances both the accuracy and transparency of flood simulations, quantifying the relative contributions and interaction mechanisms of compound variable, and offering new perspectives for analyzing the underlying causes of compound flooding. This approach holds significant potential for developing more robust disaster mitigation systems and strengthening the resilience of coastal cities.

PMID:40450943 | DOI:10.1016/j.jenvman.2025.125858

Int J Pharm. 2025 May 30:125789. doi: 10.1016/j.ijpharm.2025.125789. Online ahead of print.

ABSTRACT

Integrating artificial intelligence (AI) into drug discovery has revolutionized pharmaceutical innovation, addressing the challenges of traditional methods that are costly, time-consuming, and suffer from high failure rates. By utilizing machine learning (ML), deep learning (DL), and natural language processing (NLP), AI enhances various stages of drug development, including target identification, lead optimization, de novo drug design, and drug repurposing. AI tools, such as AlphaFold for protein structure prediction and AtomNet for structure-based drug design, have significantly accelerated the discovery process, improved efficiency and reduced costs. Success stories like Insilico Medicine's AI-designed molecule for idiopathic pulmonary fibrosis and BenevolentAI's identification of baricitinib for COVID-19 highlight AI's transformative potential. Additionally, AI enables the exploration of vast chemical spaces, optimization of clinical trials, and the identification of novel therapeutic targets, paving the way for precision medicine. However, challenges such as limited data accessibility, integration of diverse datasets, interpretability of AI models, and ethical concerns remain critical hurdles. Overcoming these limitations through enhanced algorithms, standardized databases, and interdisciplinary collaboration is essential. Overall, AI continues to reshape drug discovery, reducing timelines, increasing success rates, and driving the development of innovative and accessible therapies for unmet medical needs.

PMID:40451590 | DOI:10.1016/j.ijpharm.2025.125789

J Evid Based Med. 2025 Jun;18(2):e70035. doi: 10.1111/jebm.70035.

ABSTRACT

OBJECTIVES: To evaluate the relationship between socioeconomic status (SES), lifestyle factors, and their combined impact on chronic respiratory diseases (CRDs).

METHODS: Participants were from the UK Biobank and were categorized into SES groups using latent class analysis based on family income, education, and employment status. Lifestyle factors were assessed via 24-hour dietary recalls and structured questionnaires. Each criterion scored 1 (healthy) or 0 (unhealthy), creating a total score from 0 to 4. Multivariable Cox proportional hazards models, interaction analyses, and mediation analyses were conducted.

RESULTS: Among 296,731 participants, 12,128 (4.1%) participants were diagnosed with CRDs. Among low SES groups, healthy lifestyle groups with scores 2, 1, and 0 showed significantly increased hazard ratios of 1.32 (95% CI: 1.21-1.44), 1.77 (95% CI: 1.63-1.93) and 2.36 (95% CI: 2.15-2.60) compared with the healthy lifestyle scores ≥3. The combined effect of SES and healthy lifestyle increased the risk of CRDs by 15% over the risk expected from simply adding their respective effects. The proportion of SES on CRDs incidence mediated by healthy lifestyle factors was statistically significant (p < 0.001), accounting for about 2%.

CONCLUSIONS: The risk of incident CRDs in the low SES population with an unhealthy lifestyle increased by 32%-136%. Unhealthy lifestyles significantly affect the incidence of CRDs in different SES subgroups. About 2% of the risk between SES and incident CRDs was mediated by lifestyle factors. These findings highlight the importance of addressing socioeconomic disparities and unhealthy lifestyle behaviors in public health strategies aimed at preventing CRDs.

PMID:40450702 | DOI:10.1111/jebm.70035

J Ethnopharmacol. 2025 May 30:120071. doi: 10.1016/j.jep.2025.120071. Online ahead of print.

ABSTRACT

ETHNOPHARMACOLOGICAL RELEVANCE: Qingfeiyin decoction (QFY), a well-documented traditional Chinese medicine formula, is used to treat a variety of lung diseases. However, the effect of QFY on idiopathic pulmonary fibrosis (IPF) is still unclear.

AIM OF THE STUDY: This study aimed to investigate the effect of QFY on IPF, identify its bioactive components, and reveal the possible mechanism.

MATERIALS AND METHODS: A bleomycin-induced pulmonary fibrosis mouse model was established to evaluate the anti-IPF effects of QFY. The bioactive ingredients in QFY were identified by HPLC-MS, following which their in vitro effects were examined by the cell models of epithelial-to-mesenchymal transition, fibroblast-to-myofibroblast transition and macrophage polarization. Network pharmacology and clinical dataset (GSE110147) were integrated to predict the possible mechanism of the bioactive ingredients, which were verified by in vitro experiments, molecular docking and molecular dynamics simulation.

RESULTS: QFY demonstrated dose-dependent amelioration of IPF pathological phenotypes. Oroxylin A (OA), Geniposide (GDS), Baicalin (BCL), and Genipin-1-gentiobioside (GG) were identified with significant lung tissue enrichment, where OA and BCL exhibited obvious improvement in EMT, FMT, and macrophage polarization experiments. Network pharmacology and bioinformatics analyses revealed the association of OA/GDS/BCL/GG with the PI3K-AKT signaling pathway, which were confirmed by their effects on the migration and apoptosis of A549 and HFL1 cells. Concurrently, molecular docking and molecular dynamics simulations demonstrated strong binding affinities of BCL and OA for p110α and p110γ subunits of PI3K.

CONCLUSIONS: Our work demonstrates the potential of QFY in the treatment of IPF, which might due to the bioactive ingredients in the lung affecting the PI3K signaling pathway.

PMID:40451494 | DOI:10.1016/j.jep.2025.120071

FEBS J. 2025 Jun 1. doi: 10.1111/febs.70140. Online ahead of print.

ABSTRACT

The fruit fly Drosophila melanogaster has emerged as an important model organism to shed light on neurodegeneration. Parkinson's disease (PD) is the second most prevalent neurodegenerative disorder, the cause of which is still mostly unclear. The long-term use of available PD drugs may have major side effects, and they only target the symptoms without providing any effective cure for the disease. Therefore, in vivo and in silico approaches are extensively used to model PD-like phenotypes in Drosophila and investigate cellular alterations underlying PD pathogenesis. In vivo models are particularly crucial to provide insight into the PD-related molecular processes. It has been a preferred approach to investigate these models by collecting omics datasets, which can be further analysed using in silico modeling such as genome-scale metabolic models and artificial intelligence applications. This review aims to summarise in vivo and in silico modeling studies in the literature to illustrate the potential of the Drosophila in the characterisation of PD-related biological mechanisms towards providing early biomarkers and novel treatment options for PD.

PMID:40451947 | DOI:10.1111/febs.70140

Comput Methods Programs Biomed. 2025 May 23;269:108840. doi: 10.1016/j.cmpb.2025.108840. Online ahead of print.

ABSTRACT

BACKGROUND: Radiomic features, derived from a region of interest (ROI) in medical images, are valuable as prognostic factors. Selecting an appropriate ROI is critical, and many recent studies have focused on leveraging multiple ROIs by segmenting analogous regions across patients - such as the primary tumour and peritumoral area or subregions of the tumour. These can be straightforwardly incorporated into models as additional features. However, a more complex scenario arises, for example, in a regionally disseminated disease, when multiple distinct lesions are present.

AIM: This study aims to evaluate the feasibility of integrating radiomic data from multiple lesions into survival models. We explore strategies for incorporating these ROIs and hypothesize that including all available lesions can improve model performance.

METHODS: While each lesion produces a feature vector, the desired result is a unified prediction. We propose methods to aggregate either the feature vectors to form a representative one or the modelling results to compute a consolidated risk score. As a proof of concept, we apply these strategies to predict distant metastasis risk in a cohort of 115 non-small cell lung cancer patients, 60% of whom exhibit regionally advanced disease. Two feature sets (radiomics extracted from PET and PET interpolated to CT resolution) are tested across various survival models using a Monte Carlo Cross-Validation framework.

RESULTS: Across both feature sets, incorporating all available lesions - rather than limiting analysis to the primary tumour - consistently improved the c-index, irrespective of the survival model used. The highest c-Index obtained by a primary tumour-only model was 0.611 for the PET dataset and 0.614 for the PET_CT dataset, while by using all lesions we were able to achieve c-Indices of 0.632 and 0.634.

CONCLUSION: Lesions beyond the primary tumour carry information that should be utilized in radiomics-based models to enhance predictive ability.

PMID:40451095 | DOI:10.1016/j.cmpb.2025.108840

Brain Behav Immun. 2025 May 30:S0889-1591(25)00204-1. doi: 10.1016/j.bbi.2025.05.027. Online ahead of print.

ABSTRACT

Disease-associated microglia (DAM), initially described in mouse models of neurodegenerative diseases, have been classified into two related states; starting from a TREM2-independent DAM1 state to a TREM2 dependent state termed DAM2, with each state being characterized by the expression of specific marker genes (Keren-Shaul, 2017). Recently, single-cell (sc)RNA-Seq studies have reported the existence of DAMs in humans (Pettas, 2022; Jauregui, 2023; Friedman, 2018; Mathys, 2019; Tuddenham, 2024); however, whether DAMs play beneficial or detrimental roles in the context of neurodegeneration is still under debate (Butovsky and Weiner, 2018; Wang and Colonna, 2019). Here, we present a pharmacological approach to mimic human DAM in vitro: we validated in silico predictions that two different histone deacetylase (HDAC) inhibitors, Entinostat and Vorinostat, recapitulate aspects of the DAM signature in two human microglia-like model systems. HDAC inhibition increases RNA expression of MITF, a transcription factor previously described as a regulator of the DAM signature (Dolan, 2023). This engagement of MITF appears to be associated with one part of the DAM signature, refining our understanding of the DAM signature as a combination of at least two transcriptional programs that appear to be correlated in vivo. Further, we functionally characterized our DAM-like model system, showing that the upregulation of this transcriptional program by HDAC inhibitors leads to an upregulation of amyloid β and pHrodo Dextran uptake - while E.coli uptake is reduced - and a specific reduction of MCP1 secretion in response to IFN-γ and TNF-α. Overall, our strategy for compound-driven microglial polarization offers potential for exploring the function of human DAM and for an immunomodulatory strategy around HDAC inhibition.

PMID:40451396 | DOI:10.1016/j.bbi.2025.05.027

BMC Cancer. 2025 Jun 1;25(1):979. doi: 10.1186/s12885-025-14391-7.

ABSTRACT

BACKGROUND: Wilms' tumor (WT) is a common renal malignancy in children. Although certain patient groups exhibit high survival rates, those experiencing recurrence, metastasis, or chemoresistance face significant challenges. The identification of reliable prognostic markers is essential for adapting treatment strategies to enhance survival rates and reduce chemotherapy-related adverse events (CRAEs).

METHODS: This study included patients diagnosed with WT at our institution. Inflammatory biomarkers were measured from pre-treatment blood tests, and their associations with event-free survival (EFS) and overall survival (OS) were evaluated using Kaplan-Meier and Cox regression analyses. The relationship between biomarkers and CRAEs was examined through logistic regression.

RESULTS: Multifactorial Cox regression analysis identified tumor stage (HR = 4.68, 95% CI: 1.58-13.87, p = 0.005), pan-immune-inflammation value (PIV) (HR = 3.94, 95% CI: 1.80-8.60, p < 0.001), and neutrophil-to-lymphocyte ratio (NLR) (HR = 0.40, 95% CI: 0.18-0.90, p = 0.027) as independent prognostic factors for EFS. Multivariate Cox regression revealed that stage IV (HR = 12.24, 95% CI: 1.56-95.85, p = 0.017) and PIV levels exceeding 246.4 (HR = 5.50, 95% CI: 2.13-14.19, p < 0.001) were significant predictors for OS. Additionally, high PIV (OR 2.32, 95% CI: 1.15-4.67, p = 0.018) independently predicted the occurrence of CRAEs.

CONCLUSION: WT patients with higher PIV levels showed significant associations with poorer EFS, worse OS, and an increased likelihood of developing CRAEs during treatment.

PMID:40452012 | DOI:10.1186/s12885-025-14391-7

Diabetes Metab. 2025 May 30:101668. doi: 10.1016/j.diabet.2025.101668. Online ahead of print.

ABSTRACT

AIMS: Limited data are available on the effectiveness of pharmacological treatments for diabetes secondary to disorders of the exocrine pancreas (DEP). This study evaluated the real-world effectiveness and safety of sodium-glucose cotransporter 2 (SGLT2) inhibitors in individuals with DEP.

METHODS: A retrospective cohort study was conducted using data from the Korean National Health Insurance Service database. Data on 66,120 individuals with DEP who initiated glucose-lowering drugs (GLDs) between September 2014 and December 2022 were analyzed. Patients initiating SGLT2 inhibitors were matched 1:1 with patients initiating other GLDs using propensity-score matching. The effectiveness outcomes included major adverse cardiovascular events (MACEs), heart failure, end-stage kidney disease (ESKD), and all-cause mortality. The safety outcomes included hypoglycemia, diabetic ketoacidosis, genital infections, urinary tract infections, fractures, and pancreatitis.

RESULTS: After matching, 4,128 SGLT2 inhibitor-other GLD user pairs were included in the analysis, with a mean follow-up of 2.3 years. Compared with use of other GLDs, use of SGLT2 inhibitors was associated with a significantly lower risk of MACE (hazard ratio [HR]: 0.69; 95% confidence interval [CI]: 0.51-0.93), hospitalization for heart failure (HR: 0.70; 95% CI: 0.51-0.95), ESKD (HR: 0.19; 95% CI: 0.06-0.61), and all-cause mortality (HR: 0.38; 95% CI: 0.27-0.53). SGLT2 inhibitor use was associated with a reduced risk of urinary tract infections (HR: 0.87; 95% CI: 0.78-0.96) and pancreatitis (HR 0.71; 95% CI 0.58-0.87).

CONCLUSIONS: SGLT2 inhibitors were associated with a reduced risk of adverse cardiorenal outcomes and all-cause mortality and were safely used in patients with DEP.

PMID:40451328 | DOI:10.1016/j.diabet.2025.101668

Int J Gynecol Cancer. 2025 May 10;35(7):101930. doi: 10.1016/j.ijgc.2025.101930. Online ahead of print.

ABSTRACT

OBJECTIVE: Chemoradiotherapy is currently the main treatment for locally advanced cervical cancer. Nevertheless, the survival profile of locally advanced cervical cancer patients remains unsatisfactory because of metastasis and recurrence. We aimed to assess the survival outcomes and safety of radiotherapy ± chemotherapy combined with nimotuzumab (a human monoclonal antibody against epidermal growth factor receptor that has anti-tumor activities) for patients with locally advanced cervical cancer.

METHODS: Patients with stage IIB to IVA (International Federation of Gynecology and Obstetrics 2018) pathological and diagnosed locally advanced cervical cancer from January 2021 to December 2022 were collected in this retrospective, multi-center, and single-arm study. All patients received platinum-based radiotherapy ± chemotherapy with nimotuzumab (200 mg once a week for 6 weeks). Primary end point was overall survival. Secondary end points were progression-free survival and safety. The adverse events were recorded. Statistical analysis was performed using Statistics Analysis System software (v 9.4).

RESULTS: A total of 60 patients were collected with a median follow-up of 17.4 months (95% CI 14.8 to 19.2). The median age was 58 years (range; 35-90). A total of 16 patients (26.7%) had stage II, 38 patients (63.3%) had stage Ⅲ, and 6 patients (10%) had stage Ⅳ. The median overall survival was not reached, and the median progression-free survival was 20.4 months (95% CI 16.3 to not evaluable). Radiotherapy ± chemotherapy with nimotuzumab achieved 90.7% 1- and 2-year overall survival. Moreover, 1-year progression-free survival was 82.1%, and the 2-year progression-free survival was 47.7%. The most common treatment-related grade 3 to 4 adverse events included neutropenia (15%), anemia (21.7%), and thrombocytopenia (5%). No drug-related severe adverse events or deaths occurred.

CONCLUSIONS: The addition of nimotuzumab to radiotherapy ± chemotherapy was associated with favorable oncologic outcomes for patients with locally advanced cervical cancer, and the toxicity was tolerable and manageable.

PMID:40450867 | DOI:10.1016/j.ijgc.2025.101930

ACS Chem Neurosci. 2025 May 31. doi: 10.1021/acschemneuro.5c00082. Online ahead of print.

ABSTRACT

Caspase-3-dependent apoptosis is believed to contribute to the brain injury of ischemic stroke, and a caspase-3 inhibitor has been repeatedly reported to reduce the brain injury of ischemic stroke. However, currently recognized caspase-3 inhibitors are still only used as a research tool, and none of them is available in the clinic to treat brain injury of ischemic stroke. Based on the concept of drug repositioning and bioinformatics techniques, we have identified Cerdulatinib, a multitargeted tyrosine kinase inhibitor to treat tumors and immune-related diseases in the clinic, as a potential caspase-3 inhibitor. This study aims to explore the effect of Cerdulatinib on brain injury from ischemic stroke and the underlying mechanisms. In mice with ischemic stroke, Cerdulatinib significantly decreased infarct volume and improved sensorimotor function, memory ability, and cognitive function. In nerve cells exposed to hypoxia, Cerdulatinib increased cell viability and decreased LDH release. Mechanistically, Cerdulatinib inhibited the protein level of cleaved caspase-3 and the activity of caspase-3, resulting in a decrease in brain cell apoptosis. Based on these results, we conclude that Cerdulatinib can protect the brain against ischemic injury by reducing apoptosis, which is related to the suppression of caspase-3 cleavage and caspase-3 activity. This study may extend the clinical indications of Cerdulatinib in the treatment of patients with an ischemic stroke.

PMID:40448621 | DOI:10.1021/acschemneuro.5c00082

J Cheminform. 2025 May 31;17(1):91. doi: 10.1186/s13321-025-01025-0.

ABSTRACT

Persistent air quality pollution poses a serious threat to human health, and is one of the action points that policy makers should monitor according to the Directive 2008/50/EC. While deploying a massive network of hyperlocal sensors could provide extensive monitoring, this approach cannot generate geospatial continuous data and present several challenges in terms of logistics. Thus, developing accurate and trustable expert systems based on chemistry transport models is a key strategy for environmental protection. However, chemistry transport models present an important lack of standardization, and the formats are not interoperable between different systems, which limits the use for different stakeholders. In this context, semantic technologies provide methods and standards for scientific data and make information readable for expert systems. Therefore, this paper proposes a novel methodology for an ontology driven transformation for CHIMERE simulations, a chemistry transport model, allowing to generate knowledge graphs representing air quality information. It enables the transformation of netCDF files into RDF triples for short term air quality forecasting. Concretely, we utilize the Semantic Web Integration Tool (SWIT) framework for mapping individuals using an ontology as a template. Then, a new ontology for CHIMERE has been defined in this work, reusing concepts for other standards in the state of the art. Our approach demonstrates that RDF files can be created from netCDF in a linear computational time, allowing the scalability for expert systems. In addition, the ontology complains with the OQuaRE quality metrics and can be extended in future extensions to be applied to other chemistry transport models. SCIENTIFIC CONTRIBUTIONS: Development of the first ontology for a chemistry transport model. FAIRification of physical models thanks to the generation of knowledge graphs from netCDF files. The ontology proposed is published in PURL ( https://purl.org/chimere-ontology ) and the knowledge graph generated for a 72-h simulation can be accessed in the following repository: https://doi.org/10.5281/zenodo.13981544 .

PMID:40450355 | DOI:10.1186/s13321-025-01025-0

Cardiovasc Drugs Ther. 2025 May 31. doi: 10.1007/s10557-025-07690-5. Online ahead of print.

ABSTRACT

PURPOSE: To evaluate the impact of statin therapy on warfarin dose requirements in diabetic patients and to assess the performance of various machine learning algorithms in predicting optimal warfarin dosing.

METHODS: The datasets available for total participants of 628 (216 diabetics and 412 non-diabetic patients) were analyzed. We categorized the patients according to height, weight, gender, race, and age, plasma international normalized ratio (INR) on reported therapeutic dose of warfarin, target INR, warfarin dose, statin therapy, and indications for warfarin. Various models were tested on data of patients from the International Warfarin Pharmacogenetics Consortium (IWPC). Data preprocessing involves structuring and handling missing values. Six predictive models, including least absolute shrinkage and selection operator (LASSO), k-nearest neighbors (KNN), support vector regression (SVR), linear regression (LR), decision tree, and random forest (RF), were employed in predicting optimal warfarin dosage. The best dose for each patient will be predicted using one of the six regression models.

RESULTS: This comparative study showed that the mean (and the standard deviation) of warfarin dose for diabetic and non-diabetic patients were 38.73 (15.37) and 34.50 (18.27) mg per week, respectively. Furthermore, the impact of various statin they use is considered and patient undergoing atorvastatin and rosuvastatin therapy against the necessity of high dose warfarin if the diabetic patients use lovastatin and fluvastatin.

CONCLUSION: Diabetic patients under statin therapy, considering the specific statin used, require different warfarin dose. Through the application of advanced machine learning, models as dosing predictors may attenuate the adverse effects of warfarin.

PMID:40448807 | DOI:10.1007/s10557-025-07690-5

Am J Transplant. 2025 May 29:S1600-6135(25)00286-2. doi: 10.1016/j.ajt.2025.05.029. Online ahead of print.

ABSTRACT

Acute cellular rejection (ACR) is a key contributor to chronic lung allograft dysfunction (CLAD) following transplantation; while treatable, traditional immunosuppressive (IS) therapies are associated with significant side effects. Gene therapy offers an approach to modulate recipient immune responses while minimizing the toxicity of conventional IS. In this study, we evaluated AAV-mediated PD-L1 overexpression, an inhibitory ligand of T cells, in a rat single lung transplant model. Allogeneic Brown Norway lungs were transplanted into Fischer F344 recipients and assigned to three groups: (1) AAV9-PD-L1 via the bronchus during static cold storage, (2) no virus control, or (3) AAV9-luciferase control. All animals received CTLA-4 immunoglobulin on postoperative day (POD) 1, and sacrificed on POD14. Rejection was evaluated by a blinded lung transplant pathologist, and PD-L1 expression and CD8+ T-cell infiltration assessed via immunohistochemistry. By POD14, the AAV9-PD-L1 group displayed significantly reduced rejection severity (mean score 1.40) compared to controls (mean 3.60; p=0.005). The AAV9-luciferase group exhibited comparable rejection scores to no-virus controls (mean 3.5). Immunohistochemistry confirmed exogenous PD-L1 expression, however no significant difference in CD8+ T-cell count was observed between groups. These findings demonstrate that AAV PD-L1 gene delivery can attenuate ACR in lung transplants, offering a potential strategy to improve outcomes.

PMID:40449603 | DOI:10.1016/j.ajt.2025.05.029