J Autoimmun. 2025 May;153:103423. doi: 10.1016/j.jaut.2025.103423. Epub 2025 Apr 22.

ABSTRACT

BACKGROUND: Immune checkpoint inhibitors (ICIs) are among the most promising treatment options for cancer. However, frequent and sometimes life-threatening immune-related adverse events (irAEs) are associated with ICI treatment. Therefore, it is imperative to establish a model for predicting the risk of irAEs to identify high-risk groups, enable more accurate clinical risk‒benefit analysis for ICI treatment and decrease the incidence of irAEs. However, no ideal model for predicting irAEs has been applied in clinical practice. The aim of this study was to analyze the systemic immune characteristics of patients with irAEs and establish a model for predicting the risk of irAEs.

METHODS: We conducted a study to monitor irAEs in patients with advanced breast cancer undergoing immunotherapy during and following the treatment course. Peripheral blood mononuclear cells (PBMCs) were collected before and after two cycles of therapy. Mass cytometry time-of-flight (CyTOF) was employed to identify baseline and posttreatment immune cell subpopulations, and the relationships between the proportions of cells in these subpopulations and the occurrence of irAEs were explored. Additionally, we conducted subgroup analyses stratified by the anatomic location and time of onset of irAEs. Furthermore, we developed a logistic regression model to predict the risk of irAEs and validated this model using two independent validation cohorts from the Gene Expression Omnibus (GEO) database (accession numbers GSE189125 and GSE186143).

RESULTS: By analyzing 106 blood samples and samples from two independent validation cohorts (n = 16 and 60 patients), we found that high proportions of CXCR3+CCR6+CD4+ T cells and CD38+CD86+CXCR3+CCR6+CD8+ T cells and a low proportion of CXCR3lowCD56dim natural killer (NK) cells at baseline were significantly correlated with the incidence of irAEs (P = 0.0029, P < 0.001, and P = 0.0017, respectively). In the subgroup analysis, we observed consistent results in patients with immune-related pneumonitis (ir-pneumonitis) and immune-related thyroiditis (ir-thyroiditis). In the early irAE group, the baseline proportion of CXCR3+CCR6+CD4+ T cells was greater than that in the late irAE group (P = 0.011). An analysis of PBMCs before and after ICI treatment revealed thatthe dynamic changes in the proportions of naïve CD4+ T cells and CXCR3lowCD56dim NK cells were closely related to irAE occurrence. Finally, we ultimately developed a model for predicting the risk of irAEs, which yielded an area under the receiver operating characteristic curve (AUROC) of 0.79 in the training cohort and an AUROC of 0.75 in the single-cell validation cohort (GSE189125).

CONCLUSIONS: These findings indicate that different populations of immune cells are associated with different irAEs and that characterization of these cells may be used as biomarkers to predict the risk of specific toxicities. This will facilitate the management of irAEs and may lead to a reduction in the incidence of irAEs.

PMID:40267835 | DOI:10.1016/j.jaut.2025.103423

Immunotherapy. 2025 Apr;17(5):339-346. doi: 10.1080/1750743X.2025.2492541. Epub 2025 Apr 23.

ABSTRACT

AIMS: Immune-related adverse events (irAEs) are common side effects of immune checkpoint inhibitor (ICI) cancer therapy, affecting approximately half of ICI-treated patients. irAEs may be severe and result in hospitalization. This study examined the risk factors and outcomes of irAE-related hospitalization.

METHODS: We conducted a retrospective study including 202 metastatic cancer patients treated with ICIs at Kuopio University Hospital, Finland, in 2015-2022.

RESULTS: IrAEs occurred in 57.4% of the patients. About 26.0% of them required inpatient treatment. Hospitalization was associated with severe (grades III - IV) toxicities and need for systemic corticosteroids. Median overall survival (mOS) for hospitalized patients was 12.9 months and for outpatients with irAEs 26.9 months (p = 0.006). The duration of ICI therapy was 1.8 months in hospitalized patients and 5.0 months in outpatients (p < 0.001). The median maximum glucocorticoid doses were 52 mg and 100 mg, respectively (p < 0.001).

CONCLUSIONS: IrAE-related hospitalization deteriorated the survival of ICI-treated patients, likely due to decreased biological efficacy of ICIs resulting from short therapy periods and strong immunosuppression by glucocorticoids.

PMID:40264419 | DOI:10.1080/1750743X.2025.2492541

Paediatr Respir Rev. 2025 Apr 5:S1526-0542(25)00033-8. doi: 10.1016/j.prrv.2025.04.002. Online ahead of print.

ABSTRACT

It is a challenge to select the "best" recent publications in a field. This is especially so when faced with a feast of outstanding manuscripts across a broad range of topics. I therefore reached out to a Who's Who of friends and colleagues in pediatric pulmonary and sleep medicine for suggestions, and I was delighted and overwhelmed by the response - please see the Acknowledgements for those who contributed ideas. Overwhelmed, by having to read 77 publications suggested by one or more colleagues and having to winnow the list down to a somewhat reasonable number. I chose to include all papers mentioned by two or more of my colleagues and I then selected the remainder to cover the broad range of our field, based upon my belief that a manuscript represented an important contribution to our understanding and clinical care. What follows are the chosen papers organized by topic area. Given the number of papers that made the final cut, I have briefly summarized each of these manuscripts. I hope that you will find something new and exciting in these publications and that you will have as much fun in reading them as I did.

PMID:40268602 | DOI:10.1016/j.prrv.2025.04.002

Eur Respir J. 2025 Apr 23:2402490. doi: 10.1183/13993003.02490-2024. Online ahead of print.

NO ABSTRACT

PMID:40268505 | DOI:10.1183/13993003.02490-2024

Chest. 2025 Apr 22:S0012-3692(25)00518-5. doi: 10.1016/j.chest.2025.04.017. Online ahead of print.

ABSTRACT

BACKGROUND: Chronic obstructive pulmonary disease (COPD) is characterized by persistent inflammation that is responsible for remodeling the bronchovascular bundles, which may lead to poor quality of life. Quantitative computed tomography (QCT) textures of the lung can capture local disease patterns of inflammation and related respiratory morbidity.

RESEARCH QUESTION: Are bronchovascular bundle textures, obtained from the adaptive multiple feature method (AMFM), associated with systemic inflammation, morbidity, and mortality in COPD?

STUDY DESIGN AND METHODS: We analyzed data from the SPIROMICS (n = 2,981) and COPDGene (n = 10,305) studies. The predictors included two QCT biomarkers, the bronchovascular bundles (BVB) and CT density gradient (CTDG) textures, age, sex, BMI, race, smoking status, smoking pack-years, CT emphysema, and Pi10 (airway wall thickness). Outcomes included plasma biomarker concentrations from Meso Scale Discovery proteomics assays and complete blood counts, both as markers of inflammation, along with FEV1, FEV1/FVC ratio, SGRQ, 6MWD, and mMRC dyspnea scale. Associations of these QCT textures with FEV1 decline and all-cause mortality were also investigated.

RESULTS: Increased BVB texture was significantly associated with elevated neutrophil and monocyte counts, and the neutrophil-to-lymphocyte ratio (NLR), independent of clinical covariates, CT emphysema, and Pi10. Elevated CTDG was associated with increased neutrophil count, NLR, and tumor necrosis factor (TNF)-α. Increased CTDG and BVB textures were also associated with a lower FEV1 and six-minute walk distance. CTDG at baseline was also associated with decline in FEV1 at five-year follow-up in COPDGene. We observed a significant association of both BVB (HRSPIROMICS=1.084, 95% CI: 1.035, 1.135, P<0.001; HRCOPDGene=1.106, 95% CI: 1.080, 1.131, P<0.001) and CTDG (HRSPIROMICS=1.033, 95% CI: 1.003, 1.064, P=0.03; HRCOPDGene=1.079, 95% CI: 1.061, 1.096, P<0.001) textures with all-cause mortality independent of CT emphysema and Pi10.

INTERPRETATION: QCT textures may provide imaging evidence of the spatial heterogeneity of lung inflammation and overall disease burden in COPD.

PMID:40268239 | DOI:10.1016/j.chest.2025.04.017

Pharmacogenomics J. 2025 Apr 23;25(3):11. doi: 10.1038/s41397-025-00372-3.

ABSTRACT

In sub-Saharan Africa, sickle cell disease (SCD) remains a significant public health challenge. Despite the discovery of SCD over a century ago, progress in developing and accessing effective treatments has been limited. Hydroxyurea is the primary drug used for managing SCD and associated with improving clinical outcomes. However, up to 30% of patients do not respond to hydroxyurea, likely due to genetic factors. This study involved 148 individuals with SCD investigated the association of hydroxyurea response with genetic variants across 13 loci associated with HbF synthesis and drug metabolism, focusing on MYB, HBB, HBG1, HBG2, BCL11A, KLF10, HAO2, NOS1, ARG2, SAR1A, CYP2C9, and CYP2E1. Significant associations with hydroxyurea response were identified in CYP2C9, CYP2E1, KLF10, BCL11A, ARG2, HBG1, SAR1A, MYB, and NOS1 loci. Furthermore, pathway enrichment and gene-gene interaction analyses provide deeper insights into the genetic mechanisms underlying hydroxyurea treatment response, highlighting potential avenues for personalized therapy in SCD management.

PMID:40268903 | DOI:10.1038/s41397-025-00372-3

Rev Mal Respir. 2025 Apr 22:S0761-8425(25)00168-8. doi: 10.1016/j.rmr.2025.04.002. Online ahead of print.

ABSTRACT

INTRODUCTION: Fibrosing interstitial lung diseases (ILD) are severe respiratory conditions that can lead to respiratory failure and death. Over the past decade, antifibrotic therapies have represented a significant therapeutic advancement and are now widely used.

STATE OF THE ART: Pirfenidone and nintedanib have been approved for the treatment of idiopathic pulmonary fibrosis (IPF), while only nintedanib has been approved for systemic sclerosis-related ILD and progressive pulmonary fibrosis (PPF). Both drugs help to reduce the decline in forced vital capacity (FVC) characterizing these three indications and to decrease mortality, acute exacerbations, and quality of life impairment in patients with IPF and PPF.

PERSPECTIVES: Tolerance to these treatments remains a major challenge, prompting evaluation of alternative administration routes, such as inhalation. Numerous ongoing clinical trials and encouraging results from phase 3 studies are expected to lead to the approval of new antifibrotic molecules.

CONCLUSIONS: Antifibrotic therapies have proven to be crucial in the management of IPF and PPF. Prescription should be a shared decision with the patient and may be considered at an early stage, even in elderly individuals, provided that dedicated support is avaialble.

PMID:40268574 | DOI:10.1016/j.rmr.2025.04.002

Eur Respir J. 2025 Apr 23:2402317. doi: 10.1183/13993003.02317-2024. Online ahead of print.

NO ABSTRACT

PMID:40268504 | DOI:10.1183/13993003.02317-2024

Sci Rep. 2025 Apr 23;15(1):14177. doi: 10.1038/s41598-025-94023-z.

ABSTRACT

Machine learning (ML) and deep neural networks (DNN) have emerged as powerful tools for enhancing intrusion detection systems (IDS) in cybersecurity. However, recent studies have revealed their vulnerability to adversarial attacks, where maliciously perturbed traffic samples can deceive trained DNN-based detectors, leading to incorrect classifications and compromised system integrity. While numerous defense mechanisms have been proposed to mitigate these adversarial threats, many fail to achieve a balance between robustness against adversarial attacks, maintaining high detection accuracy on clean data, and preserving the functional integrity of traffic flow features. To address these limitations, this research investigates and integrates a comprehensive ensemble of adversarial defense strategies, implemented in two key phases. During the training phase, adversarial training, label smoothing, and Gaussian augmentation are employed to enhance the model's resilience against adversarial perturbations. Additionally, a proactive preprocessing defense strategy is deployed during the testing phase, utilizing a denoising sparse autoencoder to cleanse adversarial input samples before they are fed into the IDS classifier. Comparative evaluations demonstrate that the proposed ensemble defense framework significantly improves the adversarial robustness and classification performance of DNN-based IDS classifiers. Experimental results, validated on the CICIDS2017 and CICIDS2018 datasets, show that the proposed approach achieves aggregated prediction accuracies of 87.34% and 98.78% under majority voting and weighted average schemes, respectively. These findings underscore the effectiveness of the proposed framework in combating adversarial threats while maintaining robust detection capabilities, thereby advancing the state-of-the-art in adversarial defense for intrusion detection systems.

PMID:40268978 | DOI:10.1038/s41598-025-94023-z

Sci Data. 2025 Apr 23;12(1):682. doi: 10.1038/s41597-025-04979-w.

ABSTRACT

Deep learning super-resolution microscopy has advanced rapidly in recent years. Super-resolution images acquired by single molecule localization microscopy (SMLM) are ideal sources for high-quality datasets. However, the scarcity of public datasets limits the development of deep learning methods. Here, we describe a biological image dataset, DL-SMLM, which provides paired low-resolution fluorescence images and super-resolution SMLM data for training super-resolution models. DL-SMLM consists of six different subcellular structures, including microtubules, lumen and membrane of endoplasmic reticulum (ER), Clathrin coated pits (CCPs), outer membrane of mitochondria (OMM) and inner membrane of mitochondria (IMM). There are 188 sets of raw SMLM data and 100 signal levels for each low-resolution image. This allows software developers to generate thousands of training pairs through data segmentation. The performance of the imaging system was further evaluated using DNA origami samples. Finally, we demonstrated examples of super-resolution models trained using data from DL-SMLM, highlighting the effectiveness of DL-SMLM for developing deep learning super-resolution microscopy.

PMID:40268962 | DOI:10.1038/s41597-025-04979-w

Interdiscip Sci. 2025 Apr 23. doi: 10.1007/s12539-025-00691-w. Online ahead of print.

ABSTRACT

Accurate tumor grading and regional identification of cervical tumors are important for diagnosis and prognosis. Traditional manual microscopy methods suffer from time-consuming, labor-intensive, and subjective bias problems, so tumor segmentation methods based on deep learning are gradually becoming a hotspot in current research. Cervical tumors have diverse morphologies, which leads to low similarity between the mask edge and ground-truth edge of existing semantic segmentation models. Moreover, the texture and geometric arrangement features of normal tissues and tumors are highly similar, which causes poor pixel connectivity in the mask of the segmentation model. To this end, we propose an end-to-end semantic consistency network with the edge learner and the connectivity enhancer, i.e., ERNet. First, the edge learner consists of a stacked shallow convolutional neural network, so it can effectively enhance the ability of ERNet to learn and represent polymorphic tumor edges. Second, the connectivity enhancer learns detailed information and contextual information of tumor images, so it can enhance the pixel connectivity of the masks. Finally, edge features and pixel-level features are adaptively coupled, and the segmentation results are additionally optimized by the tumor classification task as a whole. The results show that, compared with those of other state-of-the-art segmentation models, the structural similarity and the mean intersection over union of ERNet are improved to 88.17% and 83.22%, respectively, which reflects the excellent edge similarity and pixel connectivity of the proposed model. Finally, we conduct a generalization experiment on laryngeal tumor images. Therefore, the ERNet network has good clinical popularization and practical value.

PMID:40268829 | DOI:10.1007/s12539-025-00691-w

Neuroimage. 2025 Apr 21:121202. doi: 10.1016/j.neuroimage.2025.121202. Online ahead of print.

ABSTRACT

Chemical exchange saturation transfer (CEST) magnetic resonance imaging (MRI) has emerged as a powerful tool to image endogenous or exogeneous macromolecules. CEST contrast highly depends on radiofrequency irradiation B1 level. Spatial inhomogeneity of B1 field would bias CEST measurement. Conventional interpolation-based B1 correction method required CEST dataset acquisition under multiple B1 levels, substantially prolonging scan time. The recently proposed supervised deep learning approach reconstructed B1 inhomogeneity corrected CEST effect at the identical B1 as of the training data, hindering its generalization to other B1 levels. In this study, we proposed a Conditional Variational Autoencoder (CVAE)-based generative model to generate B1 inhomogeneity corrected Z spectra from single CEST acquisition. The model was trained from pixel-wise source-target paired Z spectra under multiple B1 with target B1 as a conditional variable. Numerical simulation and healthy human brain imaging at 5 T were respectively performed to evaluate the performance of proposed model in B1 inhomogeneity corrected CEST MRI. Results showed that the generated B1-corrected Z spectra agreed well with the reference averaged from regions with subtle B1 inhomogeneity. Moreover, the performance of the proposed model in correcting B1 inhomogeneity in APT CEST effect, as measured by both MTRasym and [Formula: see text] at 3.5 ppm, were superior over conventional Z/contrast-B1-interplation and other deep learning methods, especially when target B1 were not included in sampling or training dataset. In summary, the proposed model allows generalized B1 inhomogeneity correction, benefiting quantitative CEST MRI in clinical routines.

PMID:40268259 | DOI:10.1016/j.neuroimage.2025.121202

Magn Reson Imaging. 2025 Apr 21:110396. doi: 10.1016/j.mri.2025.110396. Online ahead of print.

ABSTRACT

PURPOSE: To accelerate 3D whole-heart joint T1/T2 mapping for myocardial tissue characterization using an end-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction of multi-contrast undersampled data.

METHODS: A free-breathing high-resolution motion-compensated 3D joint T1/T2 water/fat sequence is employed. The sequence consists of the acquisition of four interleaved volumes with 2-echo encoding, resulting in eight volumes with different contrasts. An end-to-end non-rigid motion-corrected reconstruction network is used to estimate high quality motion-corrected reconstructions from the eight multi-contrast undersampled data for subsequent joint T1/T2 mapping. Reconstruction with the proposed approach was compared against state-of-the-art motion-corrected HD-PROST reconstruction.

RESULTS: The proposed approach yields images with good visual agreement compared to the reference reconstructions. The comparison of the quantitative values in the T1 and T2 maps showed the absence of systematic errors, and a small bias of -6.35 ms and -1.8 ms, respectively. The proposed reconstruction time was 24 seconds in comparison to 2.5 hours with motion-corrected HD-PROST, resulting in a reconstruction speed-up of over 370 times.

CONCLUSION: In conclusion, this study presents a promising method for efficient whole-heart myocardial tissue characterization. Specifically, the research highlights the potential of the multi-contrast end-to-end deep learning algorithm for joint motion estimation and model-based motion-corrected reconstruction of multi-contrast undersampled data. The findings underscore its ability to compute T1 and T2 values with good agreement when compared to the reference motion-corrected HD-PROST method, while substantially reducing reconstruction time.

PMID:40268172 | DOI:10.1016/j.mri.2025.110396

Methods. 2025 Apr 21:S1046-2023(25)00108-2. doi: 10.1016/j.ymeth.2025.04.011. Online ahead of print.

ABSTRACT

RNA modifications play a crucial role in enhancing the structural and functional diversity of RNA molecules and regulating various stages of the RNA life cycle. Among these modifications, N6-Methyladenosine (m6A) is the most common internal modification in eukaryotic mRNAs and has been extensively studied over the past decade. Accurate identification of m6A modification sites is essential for understanding their function and underlying mechanisms. Traditional methods predominantly rely on machine learning techniques to recognize m6A sites, which often fail to capture the contextual features of these sites comprehensively. In this study, we comprehensively summarize previously published methods based on machine learning and deep learning. We also validate multiple deep learning approaches on benchmark dataset, including previously underutilized methods in m6A site prediction, pre-trained models specifically designed for biological sequence and other basic deep learning methods. Additionally, we further analyze the dataset features and interpret the model's predictions to enhance understanding. Our experimental results clearly demonstrate the effectiveness of the deep learning models, elucidating their strong potential in accurately recognizing m6A modification sites.

PMID:40268153 | DOI:10.1016/j.ymeth.2025.04.011

Biochim Biophys Acta Mol Cell Res. 2025 Apr 21:119964. doi: 10.1016/j.bbamcr.2025.119964. Online ahead of print.

ABSTRACT

Although image analysis has emerged as a key technology in the study of organelle dynamics, the commonly used image-processing methods, such as threshold-based segmentation and manual setting of regions of interests (ROIs), are error-prone and laborious. Here, we present a highly accurate high-throughput image analysis pipeline called OrgaMeas for measuring the morphology and dynamics of organelles. This pipeline mainly consists of two deep learning-based tools: OrgaSegNet and DIC2Cells. OrgaSegNet quantifies many aspects of different organelles by precisely segmenting them. To further process the segmented data at a single-cell level, DIC2Cells automates ROI settings through accurate segmentation of individual cells in differential interference contrast (DIC) images. This pipeline was designed to be low cost and require less coding, to provide an easy-to-use platform. Thus, we believe that OrgaMeas has potential to be readily applied to basic biomedical research, and hopefully to other practical uses such as drug discovery.

PMID:40268058 | DOI:10.1016/j.bbamcr.2025.119964

Int J Biol Macromol. 2025 Apr 21:143308. doi: 10.1016/j.ijbiomac.2025.143308. Online ahead of print.

ABSTRACT

Biological interactions between RNA and small-molecule ligands play a crucial role in determining the specific functions of RNA, such as catalysis and folding, and are essential for guiding drug design in the medical field. Accurately predicting the binding sites of ligands within RNA structures is therefore of significant importance. To address this challenge, we introduced a computational approach named RLBSIF (RNA-Ligand Binding Surface Interaction Fingerprints) based on geometric deep learning. This model utilizes surface geometric features, including shape index and distance-dependent curvature, combined with chemical features represented by atomic charge, to comprehensively characterize RNA-ligand interactions through MaSIF-based surface interaction fingerprints. Additionally, we employ the ResNet18 network to analyze these fingerprints for identifying ligand binding pockets. Trained on 440 binding pockets, RLBSIF achieves an overall pocket-level classification accuracy of 90 %. Through a full-space enumeration method, it can predict binding sites at nucleotide resolution. In two independent tests, RLBSIF outperformed competing models, demonstrating its efficacy in accurately identifying binding sites within complex molecular structures. This method shows promise for drug design and biological product development, providing valuable insights into RNA-ligand interactions and facilitating the design of novel therapeutic interventions. For access to the related source code, please visit RLBSIF on GitHub (https://github.com/ZUSTSTTLAB/RLBSIF).

PMID:40268011 | DOI:10.1016/j.ijbiomac.2025.143308

Phys Med Biol. 2025 Apr 23. doi: 10.1088/1361-6560/adcfeb. Online ahead of print.

ABSTRACT

\textit{Objective.} This study investigates key factors influencing deep learning-based dose prediction models for head and neck cancer radiation therapy (RT). The goal is to evaluate model accuracy, robustness, and computational efficiency, and to identify key components necessary for optimal performance.&#xD;\\&#xD;\textit{Approach.} We systematically analyze the impact of input and dose grid resolution, input type, loss function, model architecture, and noise on model performance. Two datasets are used: a public dataset (OpenKBP) and an in-house clinical dataset (LUMC). Model performance is primarily evaluated using two metrics: dose score and dose-volume histogram (DVH) score.&#xD;\\&#xD;\textit{Main results.} &#xD;High-resolution inputs improve prediction accuracy (dose score and DVH score) by 8.6--13.5\% compared to low resolution. Using a combination of CT, planning target volumes (PTVs), and organs-at-risk (OARs) as input significantly enhances accuracy, with improvements of 57.4--86.8\% over using CT alone. Integrating mean absolute error (MAE) loss with value-based and criteria-based DVH loss functions further boosts DVH score by 7.2--7.5\% compared to MAE loss alone. In the robustness analysis, most models show minimal degradation under Poisson noise (0--0.3 Gy) but are more susceptible to adversarial noise (0.2--7.8 Gy). Notably, certain models, such as SwinUNETR, demonstrate superior robustness against adversarial perturbations.&#xD;\\&#xD;\textit{Significance.}&#xD;These findings highlight the importance of optimizing deep learning models and provide valuable guidance for achieving more accurate and reliable radiotherapy dose prediction.

PMID:40267938 | DOI:10.1088/1361-6560/adcfeb