Disease-associated and patient-specific immune cell signatures in juvenile-onset systemic lupus erythematosus: patient stratification using a machine-learning approach

Multilayered visual metabolomics analysis framework for enhanced exploration of functional components in wolfberry

Wolfberry, regarded as a nutritious fruit, has garnered significant attention in the food industry due to potential health benefits. However, the tissue-specific distribution and dynamic accumulation patterns of nutritional metabolites such as flavonoids are still unclear. In this study, a novel spatial metabolomics framework was developed, incorporating instrumental optimization, metabolite identification, molecular network analysis, metabolic pathway mapping, and machine learning-based imaging. Using DESI-MSI, this approach enabled rapid, non-destructive, in situ analysis of wolfberry metabolites with enhanced sensitivity and spatial resolution. Detailed insights into chemical and spatial changes during ripening were obtained, with a focus on flavonoids. The visualization of the flavonoid biosynthetic pathway highlighted the impact of C-3 hydroxylation on flavonoid redistribution. Furthermore, a classification model achieved a prediction accuracy exceeding 99 %, consistent with metabolic network analyses. This framework provides a powerful tool for plant metabolomics, facilitating the exploration of functional components and metabolic pathways.

Development of disease diagnosis technology based on coattention cross-fusion of multiomics data

BACKGROUND

Early diagnosis is vital for increasing the rates of curing diseases and patient survival in medicine. With the advancement of biotechnology, the types of bioomics data are increasing. The integration of multiomics data can provide more comprehensive biological information, thereby achieving more accurate diagnoses than single-omics data can. Nevertheless, current multiomics research is often limited to the intelligent diagnosis of a single disease or a few types of omics data and lacks a multiomics disease diagnosis model that can be widely applied to different diseases. Therefore, developing a model that can effectively utilize multiomics data and accurately diagnose diseases has become an important challenge in medical research.

RESULTS

On the basis of vibrational spectroscopy and metabolomics data, this study proposes an innovative coattention cross-fusion model for disease diagnosis on the basis of interactions of multiomics data. The model not only integrates the information of different omics data but also simulates the interactions between these data to achieve accurate diagnosis of diseases. Through comprehensive experiments, our method achieved accuracies of 95.00 %, 94.95 %, and 97.22 % and area under the curve (AUC) values of 95.00 %, 96.77 %, and 99.31 % on the cervical lymph node metastasis of the thyroid, systemic lupus erythematosus, and cancer datasets, respectively, indicating excellent performance in the diagnosis of multiple diseases.

SIGNIFICANCE

The proposed model outperforms existing multiomics models, enhancing medical diagnostic accuracy and offering new approaches for multiomics data use in disease diagnosis. The innovative coattention cross-fusion module enables more effective multiomics data processing and analysis, serving as a potent tool for early and precise disease diagnosis with substantial clinical and research implications.

Using machine learning to investigate the influence of the prenatal chemical exposome on neurodevelopment of young children

Research investigating the prenatal chemical exposome and child neurodevelopment has typically focused on a limited number of chemical exposures and controlled for sociodemographic factors and maternal mental health. Emerging machine learning approaches may facilitate more comprehensive examinations of the contributions of chemical exposures, sociodemographic factors, and maternal mental health to child neurodevelopment. A machine learning pipeline that utilized feature selection and ranking was applied to investigate which common prenatal chemical exposures and sociodemographic factors best predict neurodevelopmental outcomes in young children. Data from 406 maternal-child pairs enrolled in the APrON study were used. Maternal concentrations of 32 environmental chemical exposures (i.e., phthalates, bisphenols, per- and polyfluoroalkyl substances (PFAS), metals, trace elements) measured during pregnancy and 11 sociodemographic factors, as well as measures of maternal mental health and urinary creatinine were entered into the machine learning pipeline. The pipeline, which consisted of a RReliefF variable selection algorithm and support vector machine regression model, was used to identify and rank the best subset of variables predictive of cognitive, language, and motor development outcomes on the Bayley Scales of Infant Development-Third Edition (Bayley-III) at 2 years of age. Bayley-III cognitive scores were best predicted using 29 variables, resulting in a correlation coefficient of r = 0.27 (R2=0.07). For language outcomes, 45 variables led to the best result (r = 0.30; R2=0.09), whereas for motor outcomes 33 variables led to the best result (r = 0.28, R2=0.09). Environmental chemicals, sociodemographic factors, and maternal mental health were found to be highly ranked predictors of cognitive, language, and motor development in young children. Our findings demonstrate the potential of machine learning approaches to identify and determine the relative importance of different predictors of child neurodevelopmental outcomes. Future developmental neurotoxicology research should consider the prenatal chemical exposome as well as sample characteristics such as sociodemographic factors and maternal mental health as important predictors of child neurodevelopment.

Integration of transcriptome and immunophenotyping data highlights differences in the pathogenetic kinetics of B cells across immune-mediated disease

OBJECTIVE

To elucidate crucial immune cell subsets and associated immunological pathways by stratifying patients with immune-mediated diseases (IMDs) using immunophenotyping and transcriptomic approaches.

METHODS

We conducted flow cytometric and transcriptomic analyses in 23 immune cell subsets derived from 235 patients with six IMDs, using our database, utilizing our database, ImmuNexUT. Patients were stratified based on immunophenotyping data. Subsequently, we examined clinical and transcriptomic differences among these stratified clusters.

RESULTS

Patients with IMDs were stratified into two clusters based on their immunophenotypes. Cluster 1 was enriched with differentiated B cells, including unswitched memory B cells (USM B), switched memory B cells, double-negative B cells and plasmablasts, while cluster 2 was enriched with naïve B cells. Higher disease activity in rheumatoid arthritis and decreased respiratory functions in systemic sclerosis were observed in cluster 1, whereas the disease activity of systemic lupus erythematosus was higher in cluster 2. Numerous differentially expressed genes were detected in USM B. Cluster 1 was associated with glycosylation processes in USM B and elevated B cell-activating factor signalling from myeloid cells in B cells, while cluster 2 exhibited higher B-cell receptor signalling in USM B. Patients in cluster 2, which had an elevated age-associated B-cell signature, exhibited more frequent flares, suggesting that an increased proportion of naïve B cells with this signature is associated with poor prognosis.

CONCLUSION

Immunophenotyping-based clusters and transcriptome-based states revealed quantitative and qualitative differences in B cells. To predict IMD prognosis, assessing both the quantity and quality of naïve B cells may be crucial.

Systems Human Immunology and AI: Immune Setpoint and Immune Health

The immune system, critical for human health and implicated in many diseases, defends against pathogens, monitors physiological stress, and maintains tissue and organismal homeostasis. It exhibits substantial variability both within and across individuals and populations. Recent technological and conceptual progress in systems human immunology has provided predictive insights that link personal immune states to intervention responses and disease susceptibilities. Artificial intelligence (AI), particularly machine learning (ML), has emerged as a powerful tool for analyzing complex immune data sets, revealing hidden patterns across biological scales, and enabling predictive models for individualistic immune responses and potentially personalized interventions. This review highlights recent advances in deciphering human immune variation and predicting outcomes, particularly through the concepts of immune setpoint, immune health, and use of the immune system as a window for measuring health. We also provide a brief history of AI; review ML modeling approaches, including their applications in systems human immunology; and explore the potential of AI to develop predictive models and personal immune state embeddings to detect early signs of disease, forecast responses to interventions, and guide personalized health strategies.

Advances in Targeted Therapy for Systemic Lupus Erythematosus: Current Treatments and Novel Approaches

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with diverse clinical manifestations that can lead to severe organ damage. The complex pathophysiology of SLE makes treatment selection difficult. This review examines the current evidence for biological therapies in SLE, including the anti-B cell activating factor antibody belimumab; the type I interferon receptor antagonist anifrolumab; the novel calcineurin inhibitor voclosporin; and rituximab, which targets CD20 on B cells. We also describe emerging therapies, including novel agents in development and CD19-directed chimeric antigen receptor (CAR) T cell therapy, which has shown promise in early clinical experience. Recent advances in biomarker research, including interferon signatures and transcriptomic profiles, may facilitate patient stratification and treatment selection. This review offers insights into current and future treatment strategies for patients with SLE by analyzing clinical trial results and recent immunological findings.

Type I interferon and mitochondrial dysfunction are associated with dysregulated cytotoxic CD8+ T cell responses in juvenile systemic lupus erythematosus

Juvenile systemic lupus erythematosus (JSLE) is an autoimmune condition which causes significant morbidity in children and young adults and is more severe in its presentation than adult-onset SLE. While many aspects of immune dysfunction have been studied extensively in adult-onset SLE, there is limited and contradictory evidence of how cytotoxic CD8+ T cells contribute to disease pathogenesis and studies exploring cytotoxicity in JSLE are virtually non-existent. Here, we report that CD8+ T cell cytotoxic capacity is reduced in JSLE versus healthy controls, irrespective of treatment or disease activity. Transcriptomic and serum metabolomic analysis identified that this reduction in cytotoxic CD8+ T cells in JSLE was associated with upregulated type I interferon (IFN) signalling, mitochondrial dysfunction, and metabolic disturbances when compared to controls. Greater interrogation of the influence of these pathways on altered cytotoxic CD8+ T cell function demonstrated that JSLE CD8+ T cells had enlarged mitochondria and enhanced sensitivity to IFN-α leading to selective apoptosis of effector memory (EM) CD8+ T cells, which are enriched for cytotoxic mediator-expressing cells. This process ultimately contributes to the observed reduction in CD8+ T cell cytotoxicity in JSLE, reinforcing the growing evidence that mitochondrial dysfunction is a key pathogenic factor affecting multiple immune cell populations in type I IFN-driven rheumatic diseases.

Effects of Nigella sativa on disease activity, T lymphocytes and inflammatory cytokine profiles in pediatric systemic lupus erythematosus: A randomized controlled trial

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease with diverse manifestations, requiring long-term treatment that can have side effects, particularly in pediatric patients. Nigella sativa has shown potential for improving SLE symptoms due to its anti-inflammatory and immunomodulatory effects. The aim of this study was to investigate the immunomodulatory effect of N. sativa oil (NSO) on disease activity, T lymphocyte activity and inflammatory cytokine profiles in pediatric SLE patients. A randomized, double-blinded, placebo-controlled clinical trial was conducted at Saiful Anwar Hospital in Malang, Indonesia. Pediatric patients with SLE were randomly assigned to receive either one gram of NSO or a placebo containing starch in capsule form as adjunct therapy alongside their SLE primary treatment. Blood samples were collected before treatment and after eight weeks of daily capsules. Disease activity was assessed using the SLE Disease Activity Index 2000 (SLEDAI-2 K); flow cytometry was used to identify T helper lymphocytes, and serum cytokine levels were measured using ELISA. The statistical analysis tests were performed to compare the outcomes between groups at baseline or after the treatment, and within-group comparisons before and after the study period, as appropriate. A total of 32 patients were included in the study. A significant decrease in the SLEDAI-2 K score was observed at post-treatment in both the NSO and placebo groups (p < 0.001 and p = 0.025, respectively). The percentage of T helper 17 (Th17) cells was significantly reduced in both the NSO and placebo groups post-treatment compared to pre-treatment (p = 0.026 and p = 0.034, respectively). Conversely, the post- treatment percentage of regulatory T (Treg) cells increased significantly in both groups. A significant reduction in interleukin (IL)-2 levels was observed in the NSO and placebo groups at post-treatment compared to pre-treatment (p = 0.006 and p = 0.046, respectively). Additionally, there were increases in IL-4 and IL-6 serum levels in both groups at post-treatment compared to pre-treatment (p < 0.05). This study highlights that although disease activity was not significantly different between NSO and placebo groups, NSO could affect the inflammatory cytokine profiles in pediatric SLE patients.

Diagnosis of systemic lupus erythematosus using cross-modal specific transfer fusion technology based on infrared spectra and metabolomics

BACKGROUND

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease. Currently, the medical diagnosis of SLE mainly relies on the clinical experience of physicians, and there is no universally accepted objective method for diagnosing SLE. Therefore, there is an urgent need to design an intelligent approach to accurately diagnose SLE to assist physicians in formulating appropriate treatment plans. With the rapid development of intelligent medical diagnostic technology, medical data is becoming increasingly multimodal. Multimodal data fusion can provide richer information than single-modal data, and the fusion of multiple modalities can effectively enhance the richness of data features to improve modeling performance.

RESULTS

In this paper, a cross-modal specific transfer fusion technique based on infrared spectra and metabolomics is proposed to effectively integrate infrared spectra and metabolomics by fully exploiting the intrinsic relationships between features across different modalities, thus achieving the diagnosis of SLE. In this research, a Decision Level Fusion module is also proposed to fuse the representations of two specific transfers further, obtaining the final prediction scores. Comprehensive experimental results demonstrate that the proposed method significantly improves the performance of SLE prediction, with accuracy and Area Under Curve (AUC) reaching 94.98 % and 97.13 %, respectively, outperforming existing methods.

SIGNIFICANCE

Our framework effectively integrates infrared spectra and metabolomics to achieve a more accurate prediction of SLE. Our research indicates that prediction methods based on different modalities outperform those using single-modality data. The Cross-modal Specific Transfer Fusion module effectively captures the complex relationships within each single modality and models the complex relationships between different modalities.

Identifying distinct phenotypes of patients with juvenile systemic lupus erythematosus: results from a cluster analysis by the Egyptian college of rheumatology (ECR) study group

PURPOSE

Juvenile systemic lupus erythematosus (J-SLE) is a complex, heterogeneous disease affecting multiple organs. However, the classification of its subgroups is still debated. Therefore, we investigated the aggregated clinical features in patients with J-SLE using cluster analysis.

METHODS

Patients (≤ 16 years) diagnosed using the Systemic Lupus International Collaborating Clinics (SLICC) classification criteria were identified from the clinical database of the Egyptian College of Rheumatology (ECR) SLE study group. Demographic data, clinical characteristics, laboratory features, and current therapies were selected. A cluster analysis was performed to identify different clinical phenotypes.

RESULTS

Overall, 404 patients, of whom 355 (87.9%) were female, had a mean age at diagnosis of 11.2 years and a mean disease duration of 2.3 years. We identified four distinct subsets of patients. Patients in cluster 1 (n = 103, 25.5%) were characterized predominantly by mucocutaneous and neurologic manifestations. Patients in cluster 2 (n = 101, 25%) were more likely to have arthritis and pulmonary manifestations. Cluster 3 (n = 71, 17.6%) had the lowest prevalence of arthritis and lupus nephritis (LN), indicative of mild disease intensity. Patients in cluster 4 (n = 129, 31.9%) have the highest frequency of arthritis, vasculitis, and LN. Cluster 1 and 4 patients had the highest disease activity index score and were less likely to use low-dose aspirin (LDA). The SLE damage index was comparable across clusters.

CONCLUSIONS

Four identified J-SLE clusters express distinct clinical phenotypes. Attention should be paid to including LDA in the therapeutic regimen for J-SLE. Further work is needed to replicate and clarify the phenotype patterns in J-SLE.

Development of a multivariate predictive nomogram among women with antepartum fetal death diagnosed at ≥ 34 weeks of gestation for outcome of TOLAC

OBJECTIVE

The present study was planned to develop a nomogram that will give a priori estimate on the probability of vaginal birth from maternal features in women with antepartum fetal death diagnosed at ≥ 34 week's gestation and previous one low transverse cesarean section (LTCS). This will help to reduce maternal complications and increase confidence when planning a trial of labor after cesarean section (TOLAC).

METHODS

A prospective observational study was planned where participants underwent induction of labor with Foley's catheter (unless already in spontaneous labor) within 24 h of enrolment. Participants with absent or inadequate contractions, oxytocin infusion as an additional agent was used. Data was collected on maternal predelivery features. Outcome of participants was categorized into two classes-vaginal and cesarean delivery. Classifiers were trained with data on maternal features and the accuracy of predicting outcome class determined. The classifier with maximum accuracy was used to develop a nomogram.

RESULT

Three hundred and one women underwent treatment as per protocol. Two hundred and twenty women attained successful vaginal delivery and eighty-one women underwent caesarean section. Factors having a significant impact on outcome were maternal body mass index (BMI), bishop score, duration of augmentation, estimated foetal weight, interval from previous LTCS, admission to active labor interval, vaginal delivery after LTCS and gestational age. The Naïve -Bayes model gave the highest prediction accuracy (0.88).

CONCLUSION

Non-linear classifiers by using selective features could predict the outcome of TOLAC among women with intra-uterine fetal death attempting vaginal birth at or beyond 34 weeks gestation with high accuracy.

Peripheral blood cells RNA-seq identifies differentially expressed gene network linked to lymphocyte subsets alterations and active lupus nephritis associated with declines in renal function

Background

The aim of this study was to investigate whether quantitative changes in lymphocyte subsets and gene expression in peripheral blood (PB) cells are related to the clinical manifestations and pathogenesis of lupus nephritis (LN).

Methods

We enrolled 95 pediatric-onset SLE patients with renal involvement who presented with 450 clinical episodes suspicious for LN flare. Percentages of lymphocyte subsets at each episode were determined. We stratified 55 of 95 patients as high or low subset group according to the median percentage of each lymphocyte subset and the association with changes in the eGFR (ΔeGFR) were analyzed. Peripheral blood bulk RNA-seq to identify differentially expressed genes (DEGs) in 9 active LN vs. 9 inactive LN patients and the DEG-derived network was constructed by Ingenuity Pathway Analysis (IPA).

Results

The mean ΔeGFR of low NK-low memory CD4+ T-high naive CD4+ T group (31.01 mL/min/1.73 m2) was significantly greater than that of high NK-high memory CD4+ T-low naive CD4+ T group (11.83 mL/min/1.73 m2; P = 0.0175). Kaplan-Meier analysis showed that the median time for ΔeGFR decline to mean ΔeGFR is approximately 10 years for high NK-high memory CD4+ T-low naive CD4+ T group and approximately 5 years for low NK-low memory CD4+ T-high naive CD4+ T group (log-rank test P = 0.0294).

Conclusions

Our study highlighted important connections between DEG-derived network, lymphocyte subset composition, and disease status of LN and GN. A novel scoring system based on lymphocyte subset proportions effectively stratified patients into groups with differential risks for declining renal function.

Predicting malaria outbreak in The Gambia using machine learning techniques

Malaria is the most common cause of death among the parasitic diseases. Malaria continues to pose a growing threat to the public health and economic growth of nations in the tropical and subtropical parts of the world. This study aims to address this challenge by developing a predictive model for malaria outbreaks in each district of The Gambia, leveraging historical meteorological data. To achieve this objective, we employ and compare the performance of eight machine learning algorithms, including C5.0 decision trees, artificial neural networks, k-nearest neighbors, support vector machines with linear and radial kernels, logistic regression, extreme gradient boosting, and random forests. The models are evaluated using 10-fold cross-validation during the training phase, repeated five times to ensure robust validation. Our findings reveal that extreme gradient boosting and decision trees exhibit the highest prediction accuracy on the testing set, achieving 93.3% accuracy, followed closely by random forests with 91.5% accuracy. In contrast, the support vector machine with a linear kernel performs less favorably, showing a prediction accuracy of 84.8% and underperforming in specificity analysis. Notably, the integration of both climatic and non-climatic features proves to be a crucial factor in accurately predicting malaria outbreaks in The Gambia.

Blood immunophenotyping identifies distinct kidney histopathology and outcomes in patients with lupus nephritis

Lupus nephritis (LN) is a frequent manifestation of systemic lupus erythematosus, and fewer than half of patients achieve complete renal response with standard immunosuppressants. Identifying non-invasive, blood-based pathologic immune alterations associated with renal injury could aid therapeutic decisions. Here, we used mass cytometry immunophenotyping of peripheral blood mononuclear cells in 145 patients with biopsy-proven LN and 40 healthy controls to evaluate the heterogeneity of immune activation in patients with LN and to identify correlates of renal parameters and treatment response. Unbiased analysis identified 3 immunologically distinct groups of patients with LN that were associated with different patterns of histopathology, renal cell infiltrates, urine proteomic profiles, and treatment response at one year. Patients with enriched circulating granzyme B+ T cells at baseline showed more severe disease and increased numbers of activated CD8 T cells in the kidney, yet they had the highest likelihood of treatment response. A second group characterized primarily by a high type I interferon signature had a lower likelihood of response to therapy, while a third group appeared immunologically inactive by immunophenotyping at enrollment but with chronic renal injuries. Main immune profiles could be distilled down to 5 simple cytometric parameters that recapitulate several of the associations, highlighting the potential for blood immune profiling to translate to clinically useful non-invasive metrics to assess immune-mediated disease in LN.

Systemic lupus in the era of machine learning medicine

Artificial intelligence and machine learning applications are emerging as transformative technologies in medicine. With greater access to a diverse range of big datasets, researchers are turning to these powerful techniques for data analysis. Machine learning can reveal patterns and interactions between variables in large and complex datasets more accurately and efficiently than traditional statistical methods. Machine learning approaches open new possibilities for studying SLE, a multifactorial, highly heterogeneous and complex disease. Here, we discuss how machine learning methods are rapidly being integrated into the field of SLE research. Recent reports have focused on building prediction models and/or identifying novel biomarkers using both supervised and unsupervised techniques for understanding disease pathogenesis, early diagnosis and prognosis of disease. In this review, we will provide an overview of machine learning techniques to discuss current gaps, challenges and opportunities for SLE studies. External validation of most prediction models is still needed before clinical adoption. Utilisation of deep learning models, access to alternative sources of health data and increased awareness of the ethics, governance and regulations surrounding the use of artificial intelligence in medicine will help propel this exciting field forward.

Prediction of Treatment Effect of SLE-ITP Patients Based on Cost-Sensitive Neural Network and Variational Autoencoder

OBJECTIVE

The aim of the study was to examine the factors influencing the therapeutic effect of patients with systemic lupus erythematosus combined with immune thrombocytopenia (SLE-ITP) and develop a prediction model to predict the therapeutic effect of SLE-ITP.

METHODS

Three hundred twenty-four SLE-ITP patients were retrieved from the electronic health record database of SLE patients in Jiangsu Province according to the latest treatment response criteria for ITP. We adopted the Cox model based on the least absolute shrinkage and selection operator to explore the impact factors affecting patient therapeutic effect, and we developed neural network model to predict therapeutic effect, and in prediction model, cost-sensitivity was introduced to address data category imbalance, and variational autoencoder was used to achieve data augmentation. The performance of each model was evaluated by accuracy and the area under the receiver operator curve.

RESULTS

The results showed that B-lymphocyte count, H-cholesterol level, complement-3 level, anticardiolipin antibody, and so on could be used as predictors of SLE-ITP curative effect, and abnormal levels of alanine transaminase, immunoglobulin A, and apolipoprotein B predicted adverse treatment response. The neural network treatment effect prediction model based on cost-sensitivity and variational autoencoder was better than the traditional classifiers, with an overall accuracy rate closed to 0.9 and a specificity of more than 0.9, which was useful for clinical practice to identify patients at risk of ineffective treatment response and to achieve better individualized management.

CONCLUSIONS

By predicting the curative effect of SLE-ITP, the severity of patients can be determined, and then the best treatment strategy can be planned to avoid ineffective treatment.

Maladaptive cognitive regulation moderates the mediating role of emotion dysregulation on the association between psychosocial factors and non-suicidal self-injury in depression

Introduction

Non-suicidal self-injury (NSSI) is highly prevalent in depression, and is associated with psychosocial factors, emotion dysregulation, and strategies of cognitive emotion regulation. However, the internal combination and interactions of these risk factors in depression remain unclear.

Methods

Data from 122 patients with depression, including 56 with NSSI and 66 without NSSI, were analyzed. Self-rating scales were used to assess psychosocial factors, emotion dysregulation, and cognitive regulation strategies. Sparse partial least squares discriminant analysis (sPLS-DA) was employed to explore internal combinations in each profile. A moderated mediation model was applied to examine their interactional relationship.

Results

The results identified an NSSI-related psychosocial profile characterized by high neuroticism, childhood trauma, poor family functioning, and low psychological resilience. Emotion dysregulation, including high levels of alexithymia, anhedonia, and emotion regulation difficulties, mediated the association between this psychosocial profile and NSSI. The mediated effect was further moderated by maladaptive cognitive regulation strategies.

Limitations

Lack of sufficient information on NSSI frequency and severity. Relatively small sample size for discussing the impact of gender and age of depressive patients with NSSI.

Conclusion

These findings hold important implications for the prevention, treatment, and rehabilitation of NSSI.

Phenotypic Clustering of Beta-Thalassemia Intermedia Patients Using Cardiovascular Magnetic Resonance

We employed an unsupervised clustering method that integrated demographic, clinical, and cardiac magnetic resonance (CMR) data to identify distinct phenogroups (PGs) of patients with beta-thalassemia intermedia (β-TI). We considered 138 β-TI patients consecutively enrolled in the Myocardial Iron Overload in Thalassemia (MIOT) Network who underwent MR for the quantification of hepatic and cardiac iron overload (T2* technique), the assessment of biventricular size and function and atrial dimensions (cine images), and the detection of replacement myocardial fibrosis (late gadolinium enhancement technique). Three mutually exclusive phenogroups were identified based on unsupervised hierarchical clustering of principal components: PG1, women; PG2, patients with replacement myocardial fibrosis, increased biventricular volumes and masses, and lower left ventricular ejection fraction; and PG3, men without replacement myocardial fibrosis, but with increased biventricular volumes and masses and lower left ventricular ejection fraction. The hematochemical parameters and the hepatic and cardiac iron levels did not contribute to the PG definition. PG2 exhibited a significantly higher risk of future cardiovascular events (heart failure, arrhythmias, and pulmonary hypertension) than PG1 (hazard ratio-HR = 10.5; p = 0.027) and PG3 (HR = 9.0; p = 0.038). Clustering emerged as a useful tool for risk stratification in TI, enabling the identification of three phenogroups with distinct clinical and prognostic characteristics.

Lupus Nephritis Risk Factors and Biomarkers: An Update

Lupus nephritis (LN) represents the most severe organ manifestation of systemic lupus erythematosus (SLE) in terms of morbidity and mortality. To reduce these risks, tremendous efforts have been made in the last decade to characterize the different steps of the disease and to develop biomarkers in order to better (i) unravel the pre-SLE stage (e.g., anti-nuclear antibodies and interferon signature); (ii) more timely initiation of therapy by improving early and accurate LN diagnosis (e.g., pathologic classification was revised); (iii) monitor disease activity and therapeutic response (e.g., recommendation to re-biopsy, new urinary biomarkers); (iv) prevent disease flares (e.g., serologic and urinary biomarkers); (v) mitigate the deterioration in the renal function; and (vi) reduce side effects with new therapeutic guidelines and novel therapies. However, progress is poor in terms of improvement with early death attributed to active SLE or infections, while later deaths are related to the chronicity of the disease and the use of toxic therapies. Consequently, an individualized treat-to-target strategy is mandatory, and for that, there is an unmet need to develop a set of accurate biomarkers to be used as the standard of care and adapted to each stage of the disease.

Systemic lupus erythematosus with high disease activity identification based on machine learning

OBJECTIVE

Clinical evaluation of systemic lupus erythematosus (SLE) disease activity is limited and inconsistent, and high disease activity significantly, seriously impacts on SLE patients. This study aims to generate a machine learning model to identify SLE patients with high disease activity.

METHOD

A total of 1014 SLE patients with low disease activity and 453 SLE patients with high disease activity were included. A total of 94 clinical, laboratory data and 17 meteorological indicators were collected. After data preprocessing, we use mutual information and multisurf to evaluate and select the importance of features. The selected features are used for machine learning modeling. Performance of the model is evaluated and verified by a series of binary classification indicators.

RESULTS

We screened out hematuria, proteinuria, pyuria, low complement, precipitation, sunlight and other features for model construction by integrated feature selection. After hyperparameter optimization, the LGB has the best performance (ROC: AUC = 0.930; PRC: AUC = 0.911, APS = 0.913; balance accuracy: 0.856), and the worst is the naive bayes (ROC: AUC = 0.849; PRC: AUC = 0.719, APS = 0.714; balance accuracy: 0.705). Finally, the selection of features has good consistency in the composite feature importance bar plot.

CONCLUSION

We identify SLE patients with high disease activity by a simple machine learning pipeline, especially the LGB model based on the characteristics of proteinuria, hematuria, pyuria and other feathers screened out by collective feature selection.

Deep stratification by transcriptome molecular characters for precision treatment of patients with systemic lupus erythematosus

OBJECTIVES

To leverage the high clinical heterogeneity of systemic lupus erythematosus (SLE), we developed and validated a new stratification scheme by integrating genome-scale transcriptomic profiles to identify patient subtypes sharing similar transcriptomic markers and drug targets.

METHODS

A normalized compendium of transcription profiles was created from peripheral blood mononuclear cells (PBMCs) of 1046 SLE patients and 86 healthy controls (HCs), covering an intersection of 13 689 genes from six microarray datasets. Upregulated differentially expressed genes were subjected to functional and network analysis in which samples were grouped using unsupervised clustering to identify patient subtypes. Then, clustering stability was evaluated by the stratification of six integrated RNA-sequencing datasets using the same method. Finally, the Xgboost classifier was applied to the independent datasets to identify factors associated with treatment outcomes.

RESULTS

Based on 278 upregulated DEGs of the transcript profiles, SLE patients were classified into three subtypes (subtype A-C) each with distinct molecular and cellular signatures. Neutrophil activation-related pathways were markedly activated in subtype A (named NE-driving), whereas lymphocyte and IFN-related pathways were more enriched in subtype B (IFN-driving). As the most severe subtype, subtype C [NE-IFN-dual-driving (Dual-driving)] shared functional mechanisms with both NE-driving and IFN-driving, which was closely associated with clinical features and could be used to predict the responses of treatment.

CONCLUSION

We developed the largest cohesive SLE transcriptomic compendium for deep stratification using the most comprehensive microarray and RNA sequencing datasets to date. This result could guide future design of molecular diagnosis and the development of stratified therapy for SLE patients.

Deconvolution of whole blood transcriptomics identifies changes in immune cell composition in patients with systemic lupus erythematosus (SLE) treated with mycophenolate mofetil

BACKGROUND

Systemic lupus erythematosus (SLE) is a clinically and biologically heterogeneous autoimmune disease. We explored whether the deconvolution of whole blood transcriptomic data could identify differences in predicted immune cell frequency between active SLE patients, and whether these differences are associated with clinical features and/or medication use.

METHODS

Patients with active SLE (BILAG-2004 Index) enrolled in the BILAG-Biologics Registry (BILAG-BR), prior to change in therapy, were studied as part of the MASTERPLANS Stratified Medicine consortium. Whole blood RNA-sequencing (RNA-seq) was conducted at enrolment into the registry. Data were deconvoluted using CIBERSORTx. Predicted immune cell frequencies were compared between active and inactive disease in the nine BILAG-2004 domains and according to immunosuppressant use (current and past).

RESULTS

Predicted cell frequency varied between 109 patients. Patients currently, or previously, exposed to mycophenolate mofetil (MMF) had fewer inactivated macrophages (0.435% vs 1.391%, p = 0.001), naïve CD4 T cells (0.961% vs 2.251%, p = 0.002), and regulatory T cells (1.858% vs 3.574%, p = 0.007), as well as a higher proportion of memory activated CD4 T cells (1.826% vs 1.113%, p = 0.015), compared to patients never exposed to MMF. These differences remained statistically significant after adjusting for age, gender, ethnicity, disease duration, renal disease, and corticosteroid use. There were 2607 differentially expressed genes (DEGs) in patients exposed to MMF with over-representation of pathways relating to eosinophil function and erythrocyte development and function. Within CD4 + T cells, there were fewer predicted DEGs related to MMF exposure. No significant differences were observed for the other conventional immunosuppressants nor between patients according disease activity in any of the nine organ domains.

CONCLUSION

MMF has a significant and persisting effect on the whole blood transcriptomic signature in patients with SLE. This highlights the need to adequately adjust for background medication use in future studies using whole blood transcriptomics.

Precision medicine in systemic lupus erythematosus

Systemic lupus erythematosus (SLE) is an autoimmune disease that has diverse clinical manifestations, ranging from restricted cutaneous involvement to life-threatening systemic organ involvement. The heterogeneity of pathomechanisms that lead to SLE contributes to between-patient variation in clinical phenotype and treatment response. Ongoing efforts to dissect cellular and molecular heterogeneity in SLE could facilitate the future development of stratified treatment recommendations and precision medicine, which is a considerable challenge for SLE. In particular, some genes involved in the clinical heterogeneity of SLE and some phenotype-related loci (STAT4, IRF5, PDGF genes, HAS2, ITGAM and SLC5A11) have an association with clinical features of the disease. An important part is also played by epigenetic varation (in DNA methylation, histone modifications and microRNAs) that influences gene expression and affects cell function without modifying the genome sequence. Immune profiling can help to identify an individual's specific response to a therapy and can potentially predict outcomes, using techniques such as flow cytometry, mass cytometry, transcriptomics, microarray analysis and single-cell RNA sequencing. Furthermore, the identification of novel serum and urinary biomarkers would enable the stratification of patients according to predictions of long-term outcomes and assessments of potential response to therapy.

Targeted multiomics in childhood-onset SLE reveal distinct biological phenotypes associated with disease activity: results from an explorative study

OBJECTIVE

To combine targeted transcriptomic and proteomic data in an unsupervised hierarchical clustering method to stratify patients with childhood-onset SLE (cSLE) into similar biological phenotypes, and study the immunological cellular landscape that characterises the clusters.

METHODS

Targeted whole blood gene expression and serum cytokines were determined in patients with cSLE, preselected on disease activity state (at diagnosis, Low Lupus Disease Activity State (LLDAS), flare). Unsupervised hierarchical clustering, agnostic to disease characteristics, was used to identify clusters with distinct biological phenotypes. Disease activity was scored by clinical SELENA-SLEDAI (Safety of Estrogens in Systemic Lupus Erythematosus National Assessment-Systemic Lupus Erythematosus Disease Activity Index). High-dimensional 40-colour flow cytometry was used to identify immune cell subsets.

RESULTS

Three unique clusters were identified, each characterised by a set of differentially expressed genes and cytokines, and by disease activity state: cluster 1 contained primarily patients in LLDAS, cluster 2 contained mainly treatment-naïve patients at diagnosis and cluster 3 contained a mixed group of patients, namely in LLDAS, at diagnosis and disease flare. The biological phenotypes did not reflect previous organ system involvement and over time, patients could move from one cluster to another. Healthy controls clustered together in cluster 1. Specific immune cell subsets, including CD11c+ B cells, conventional dendritic cells, plasmablasts and early effector CD4+ T cells, differed between the clusters.

CONCLUSION

Using a targeted multiomic approach, we clustered patients into distinct biological phenotypes that are related to disease activity state but not to organ system involvement. This supports a new concept where choice of treatment and tapering strategies are not solely based on clinical phenotype but includes measuring novel biological parameters.

The Impact of Artificial Intelligence in the Odyssey of Rare Diseases

Emerging machine learning (ML) technologies have the potential to significantly improve the research and treatment of rare diseases, which constitute a vast set of diseases that affect a small proportion of the total population. Artificial Intelligence (AI) algorithms can help to quickly identify patterns and associations that would be difficult or impossible for human analysts to detect. Predictive modeling techniques, such as deep learning, have been used to forecast the progression of rare diseases, enabling the development of more targeted treatments. Moreover, AI has also shown promise in the field of drug development for rare diseases with the identification of subpopulations of patients who may be most likely to respond to a particular drug. This review aims to highlight the achievements of AI algorithms in the study of rare diseases in the past decade and advise researchers on which methods have proven to be most effective. The review will focus on specific rare diseases, as defined by a prevalence rate that does not exceed 1-9/100,000 on Orphanet, and will examine which AI methods have been most successful in their study. We believe this review can guide clinicians and researchers in the successful application of ML in rare diseases.

Artificial intelligence and high-dimensional technologies in the theragnosis of systemic lupus erythematosus

Systemic lupus erythematosus is a complex, systemic autoimmune disease characterised by immune dysregulation. Pathogenesis is multifactorial, contributing to clinical heterogeneity and posing challenges for diagnosis and treatment. Although strides in treatment options have been made in the past 15 years, with the US Food and Drug Administration approval of belimumab in 2011, there are still many patients who have inadequate responses to therapy. A better understanding of underlying disease mechanisms with a holistic and multiparametric approach is required to improve clinical assessment and treatment. This Review discusses the evolution of genomics, epigenomics, transcriptomics, and proteomics in the study of systemic lupus erythematosus and ways to amalgamate these silos of data with a systems-based approach while also discussing ways to strengthen the overall process. These mechanistic insights will facilitate the discovery of functionally relevant biomarkers to guide rational therapeutic selection to improve patient outcomes.

Current state and completeness of reporting clinical prediction models using machine learning in systemic lupus erythematosus: A systematic review

OBJECTIVE

We carried out a systematic review (SR) of adherence in diagnostic and prognostic applications of ML in SLE using the Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis (TRIPOD) Statement.

METHODS

A SR employing five databases was conducted from its inception until December 2021. We identified articles that evaluated the utilization of ML for prognostic and/or diagnostic purposes. This SR was reported based on the PRISMA guidelines. The TRIPOD statement assessed adherence to reporting standards. Assessment for risk of bias was done using PROBAST tool.

RESULTS

We included 45 studies: 29 (64.4%) diagnostic and 16 (35.5%) prognostic prediction- model studies. Overall, articles adhered by between 17% and 67% (median 43%, IQR 37-49%) to TRIPOD items. Only few articles reported the model's predictive performance (2.3%, 95% CI 0.06-12.0), testing of interaction terms (2.3%, 95% CI 0.06-12.0), flow of participants (50%, 95% CI; 34.6-65.4), blinding of predictors (2.3%, 95% CI 0.06-12.0), handling of missing data (36.4%, 95% CI 22.4-52.2), and appropriate title (20.5%, 95% CI 9.8-35.3). Some items were almost completely reported: the source of data (88.6%, 95% CI 75.4-96.2), eligibility criteria (86.4%, 95% CI 76.2-96.5), and interpretation of findings (88.6%, 95% CI 75.4-96.2). In addition, most of model studies had high risk of bias.

CONCLUSIONS

The reporting adherence of ML-based model developed for SLE, is currently inadequate. Several items deemed crucial for transparent reporting were not fully reported in studies on ML-based prediction models.

REVIEW REGISTRATION

PROSPERO ID# CRD42021284881. (Amended to limit the scope).

Narrative Review of Machine Learning in Rheumatic and Musculoskeletal Diseases for Clinicians and Researchers: Biases, Goals, and Future Directions

There has been rapid growth in the use of artificial intelligence (AI) analytics in medicine in recent years, including in rheumatic and musculoskeletal diseases (RMDs). Such methods represent a challenge to clinicians, patients, and researchers, given the "black box" nature of most algorithms, the unfamiliarity of the terms, and the lack of awareness of potential issues around these analyses. Therefore, this review aims to introduce this subject area in a way that is relevant and meaningful to clinicians and researchers. We hope to provide some insights into relevant strengths and limitations, reporting guidelines, as well as recent examples of such analyses in key areas, with a focus on lessons learned and future directions in diagnosis, phenotyping, prognosis, and precision medicine in RMDs.

Investigating sex differences in T regulatory cells from cisgender and transgender healthy individuals and patients with autoimmune inflammatory disease: a cross-sectional study

Background

Sexual dimorphisms, which vary depending on age group and pubertal status, have been described across both the innate and adaptive immune system. We explored the influence of sex hormones on immune phenotype in the context of adolescent health and autoimmunity.

Methods

In this cross-sectional study, healthy, post-pubertal cisgender individuals (aged 16-25 years); healthy, pre-pubertal cisgender individuals (aged 6-11 years); transgender individuals (aged 18-19 years) undergoing gender-affirming treatment (testosterone in individuals assigned female sex at birth and oestradiol in individuals assigned male sex at birth); and post-pubertal cisgender individuals (aged 14-25 years) with juvenile-onset systemic lupus erythematosus (SLE) age-matched to cisgender individuals without juvenile-onset SLE were eligible for inclusion. Frequencies of 28 immune-cell subsets (including different T cell, B cell, and monocyte subsets) from each participant were measured in peripheral blood mononuclear cells by flow cytometry and analysed by balanced random forest machine learning. RNA-sequencing was used to compare sex and gender differences in regulatory T (Treg) cell phenotype between participants with juvenile-onset SLE, age-matched cis-gender participants without the disease, and age matched transgender individuals on gender-affirming sex hormone treatment. Differentially expressed genes were analysed by cluster and pathway analysis. Suppression assays assessed the anti-inflammatory function of Treg cells in vitro.

Findings

Between Sept 5, 2012, and Nov 6, 2019, peripheral blood was collected from 39 individuals in the post-pubertal group (17 [44%] cisgender men, mean age 18·76 years [SD 2·66]; 22 [56%] cisgender women, mean age 18·59 years [2·81]), 14 children in the cisgender pre-pubertal group (seven [50%] cisgender boys, mean age 8·90 [1·66]; seven [50%] cisgender girls, mean age 8·40 [1·58]), ten people in the transgender group (five [50%] transgender men, mean age 18·20 years [0·47]; five [50%] transgender women, mean age 18·70 years [0·55]), and 35 people in the juvenile-onset SLE group (12 [34%] cisgender men, mean age 18·58 years [2·35]; 23 [66%] cisgender women, mean age 19·48 [3·08]). Statistically significantly elevated frequencies of Treg cells were one of the top immune-cell features differentiating young post-pubertal cisgender men from similarly aged cisgender women (p=0·0097). Treg cells from young cisgender men had a statistically significantly increased suppressive capacity in vitro compared with those from cisgender women and a distinct transcriptomic signature significantly enriched for genes in the PI3K-AKT signalling pathway. Gender-affirming sex hormones in transgender men and transgender women induced multiple statistically significant changes in the Treg-cell transcriptome, many of which enriched functional pathways that overlapped with those altered between cisgender men and cisgender women, highlighting a hormonal influence on Treg-cell function by gender. Finally, sex differences in Treg-cell frequency were absent and suppressive capacity was reversed in patients with juvenile-onset SLE, but sex differences in Treg-cell transcriptional signatures were significantly more pronounced in patients with juvenile-onset SLE compared with individuals without juvenile-onset SLE, suggesting that sex hormone signalling could be dysregulated in autoimmunity.

Interpretation

Sex-chromosomes and hormones might drive changes in Treg-cell frequency and function. Young post-pubertal men have a more anti-inflammatory Treg-cell profile, which could explain inflammatory disease susceptibilities, and inform sex-tailored therapeutic strategies.

Funding

Versus Arthritis, UK National Institute for Health Research University College London Hospital Biomedical Research Centre, Lupus UK, and The Rosetrees Trust.

CD8+ T Cell Phenotype and Function in Childhood and Adult-Onset Connective Tissue Disease

CD8+ T cells are cytotoxic lymphocytes that destroy pathogen infected and malignant cells through release of cytolytic molecules and proinflammatory cytokines. Although the role of CD8+ T cells in connective tissue diseases (CTDs) has not been explored as thoroughly as that of other immune cells, research focusing on this key component of the immune system has recently gained momentum. Aberrations in cytotoxic cell function may have implications in triggering autoimmunity and may promote tissue damage leading to exacerbation of disease. In this comprehensive review of current literature, we examine the role of CD8+ T cells in systemic lupus erythematosus, Sjögren's syndrome, systemic sclerosis, polymyositis, and dermatomyositis with specific focus on comparing what is known about CD8+ T cell peripheral blood phenotypes, CD8+ T cell function, and CD8+ T cell organ-specific profiles in adult and juvenile forms of these disorders. Although, the precise role of CD8+ T cells in the initiation of autoimmunity and disease progression remains to be elucidated, increasing evidence indicates that CD8+ T cells are emerging as an attractive target for therapy in CTDs.

Ethnicity influences phenotype and clinical outcomes: Comparing a South American with a North American inflammatory bowel disease cohort

Inflammatory bowel disease (IBD), including ulcerative colitis (UC) and Crohn disease (CD), has emerged as a global disease with an increasing incidence in developing and newly industrialized regions such as South America. This global rise offers the opportunity to explore the differences and similarities in disease presentation and outcomes across different genetic backgrounds and geographic locations. Our study includes 265 IBD patients. We performed an exploratory analysis of the databases of Chilean and North American IBD patients to compare the clinical phenotypes between the cohorts. We employed an unsupervised machine-learning approach using principal component analysis, uniform manifold approximation, and projection, among others, for each disease. Finally, we predicted the cohort (North American vs Chilean) using a random forest. Several unsupervised machine learning methods have separated the 2 main groups, supporting the differences between North American and Chilean patients with each disease. The variables that explained the loadings of the clinical metadata on the principal components were related to the therapies and disease extension/location at diagnosis. Our random forest models were trained for cohort classification based on clinical characteristics, obtaining high accuracy (0.86 = UC; 0.79 = CD). Similarly, variables related to therapy and disease extension/location had a high Gini index. Similarly, univariate analysis showed a later CD age at diagnosis in Chilean IBD patients (37 vs 24; P = .005). Our study suggests a clinical difference between North American and Chilean IBD patients: later CD age at diagnosis with a predominantly less aggressive phenotype (39% vs 54% B1) and more limited disease, despite fewer biological therapies being used in Chile for both diseases.

Recent advances in the use of machine learning and artificial intelligence to improve diagnosis, predict flares, and enrich clinical trials in lupus

PURPOSE OF REVIEW

Machine learning is a computational tool that is increasingly used for the analysis of medical data and has provided the promise of more personalized care.

RECENT FINDINGS

The frequency with which machine learning analytics are reported in lupus research is comparable with that of rheumatoid arthritis and cancer, yet the clinical application of these computational tools has yet to be translated into better care. Considerable work has been applied to the development of machine learning models for lupus diagnosis, flare prediction, and classification of disease using histology or other medical images, yet few models have been tested in external datasets and independent centers. Application of machine learning has yet to be reported for lupus clinical trial enrichment and automated identification of eligible patients. Integration of machine learning into lupus clinical care and clinical trials would benefit from collaborative development between clinicians and data scientists.

SUMMARY

Although the application of machine learning to lupus data is at a nascent stage, initial results suggest a promising future.

Correlation between peripheral blood lymphocyte subpopulations and primary systemic lupus erythematosus

This study explored the correlation between peripheral blood CD3＋, CD3＋/CD4＋, CD3+/CD8+, CD4＋/CD8＋, CD3-/CD16+ CD56+, and CD3-CD19+ and disease activity of different subtypes of systemic lupus erythematosus (SLE). The percentages of CD3+, CD3+/CD4+, CD3+/CD8+, CD4+/CD8+, CD3-/CD16+ CD56+, and CD3-CD19+ in the peripheral blood of patients (n = 80) classified into lupus nephritis, blood involvement, and joint involvement and SLE in different active stages were detected by flow cytometry. Their correlations with baseline clinical experimental indicators of SLE patients' SLE disease activity index score (SLEDAI) and complement C3 were analyzed. The results showed that CD3+, CD3+/CD4+, and CD3+/CD8+ at baseline level were negatively correlated with SLEDAI scores. These were positively correlated with C3. In conclusion, T-lymphocyte subpopulations are closely related to SLE activity and can be used as reference indicators to evaluate the SLE activity.

Peripheral blood T-cell subset and its clinical significance in lupus nephritis patients

Objectives

Lupus nephritis (LN) is a common and severe manifestation of SLE. Memory T (TM) cells have been implicated in the pathogenesis of SLE. This study aimed to investigate the clinical significance of T-cell subsets in a cohort of patients with LN.

Method

The peripheral blood T cells of 24 patients with LN and 13 patients with idiopathic membranous nephropathy (iMN) were analysed by flow cytometry. SLE disease activity was evaluated by SLE Disease Activity Index-2000 (SLEDAI-2K). Patients with LN were followed up for >6 months.

Results

Patients with LN presented lower frequency of CD4⁺ cells and higher percentage of CD8⁺ cells than patients with iMN, which was primarily due to lower CD4⁺ cell count. Interestingly, patients with LN under immunosuppressants had lower CD8⁺CD45RO⁺ TM frequency (p=0.007), fewer regulatory CD4⁺ T cells (p=0.04) than those without immunosuppressants. Most CD4⁺ and CD8⁺ TM cells in patients with LN showed an effector memory (CD45RO⁺CCR7⁺) phenotype. The frequency of CD8⁺CD45RO⁺ TM cells among the CD8⁺ T cells was negatively correlated with white blood cell count, haemoglobin, platelet and serum levels of complements C3 and C4, but was positively correlated with serum IgG, erythrocyte sedimentation rate and SLEDAI score (p<0.05 each). Consistently, the frequency of CD8⁺CD45RO⁺ TM cells was higher in patients with LN with positive antidouble-stranded DNA antibody, active renal disease, extrarenal manifestations and with sclerotic glomeruli or moderate-to-severe mesangial hypercellularity in renal pathology (p<0.05). Additionally, CD8⁺CD45RO⁺ TM cell frequency was significantly lower in patients with LN with renal complete remission than that in non-remission LN (18.7% vs 31.2%, p<0.05). None of these significant correlations was observed in CD4⁺ TM cells.

Conclusion

The frequency of CD8⁺ TM cells correlates with disease activity and treatment response to immunosuppressant in patients with LN. CD8⁺ TM monitoring in patients with LN could provide more helpful indices for the monitoring and management of this disease.

Cell activation, PD-1 expression and in vitro cytokine production in patients with juvenile systemic lupus erythematosus

BACKGROUND

Juvenile systemic lupus erythematosus (jSLE) is known to be more severe and with a higher frequency of renal and central nervous system impairment when compared to systemic lupus erythematosus in adults. The study of immunological characteristics of jSLE patients might help to envisage better treatment strategies to reduce the burden of the disease.

OBJECTIVE

To characterize peripheral lymphocytes, assessing activation markers, and PD-1 expression on T cells; to evaluate in vitro cytokine expression upon stimulation in jSLE patients and age-matched controls.

METHODOLOGY

Eighteen jSLE patients on low disease activity and 25 matched healthy adolescents were evaluated for immune activation and PD-1 expression on peripheral blood lymphocytes by flow cytometry. Twenty-one cytokines were assessed by X-MAP technology after in vitro stimulation of peripheral blood with phytohemagglutinin.

RESULTS

jSLE patients had lower numbers of CD4 T, CD8 T, B, and NK cells; higher central memory CD8 T cell percentages were noted in jSLE adolescents in comparison with controls (p = 0.014). B cells subsets showed a higher percentage of exhausted memory subset than controls (p = 0.014). The expression of PD-1 on CD4 T and CD8 T cells did not show relevant changes in jSLE adolescents. After stimulation of peripheral blood, cell supernatant of jSLE patients showed a trend to lower concentrations of IL-10 (p=0.080) and higher concentrations of IL-23 (p = 0.063) than controls.

CONCLUSIONS

jSLE patients on low disease activity maintain lymphopenia of all subsets, with a B cell profile of exhaustion. Upon in vitro stimulation, peripheral blood cell supernatant showed a shift to IL-23, suggesting a role of inhibitors of this cytokine as another potential therapeutic target for those patients.

The promise of precision medicine in rheumatology

Systemic autoimmune rheumatic diseases (SARDs) exhibit extensive heterogeneity in clinical presentation, disease course, and treatment response. Therefore, precision medicine - whereby treatment is tailored according to the underlying pathogenic mechanisms of an individual patient at a specific time - represents the 'holy grail' in SARD clinical care. Current strategies include treat-to-target therapies and autoantibody testing for patient stratification; however, these are far from optimal. Recent innovations in high-throughput 'omic' technologies are now enabling comprehensive profiling at multiple levels, helping to identify subgroups of patients who may taper off potentially toxic medications or better respond to current molecular targeted therapies. Such advances may help to optimize outcomes and identify new pathways for treatment, but there are many challenges along the path towards clinical translation. In this Review, we discuss recent efforts to dissect cellular and molecular heterogeneity across multiple SARDs and future directions for implementing stratification approaches for SARD treatment in the clinic.

CD8+ T-Cells in Juvenile-Onset SLE: From Pathogenesis to Comorbidities

Diagnosis of systemic lupus erythematosus (SLE) in childhood [juvenile-onset (J) SLE], results in a more severe disease phenotype including major organ involvement, increased organ damage, cardiovascular disease risk and mortality compared to adult-onset SLE. Investigating early disease course in these younger JSLE patients could allow for timely intervention to improve long-term prognosis. However, precise mechanisms of pathogenesis are yet to be elucidated. Recently, CD8+ T-cells have emerged as a key pathogenic immune subset in JSLE, which are increased in patients compared to healthy individuals and associated with more active disease and organ involvement over time. CD8+ T-cell subsets have also been used to predict disease prognosis in adult-onset SLE, supporting the importance of studying this cell population in SLE across age. Recently, single-cell approaches have allowed for more detailed analysis of immune subsets in JSLE, where type-I IFN-signatures have been identified in CD8+ T-cells expressing high levels of granzyme K. In addition, JSLE patients with an increased cardiometabolic risk have increased CD8+ T-cells with elevated type-I IFN-signaling, activation and apoptotic pathways associated with atherosclerosis. Here we review the current evidence surrounding CD8+ T-cell dysregulation in JSLE and therapeutic strategies that could be used to reduce CD8+ T-cell inflammation to improve disease prognosis.

Identification of biomarkers by machine learning classifiers to assist diagnose rheumatoid arthritis-associated interstitial lung disease

Background

This study aimed to search for blood biomarkers among the profiles of patients with RA-ILD by using machine learning classifiers and probe correlations between the markers and the characteristics of RA-ILD.

Methods

A total of 153 RA patients were enrolled, including 75 RA-ILD and 78 RA-non-ILD. Routine laboratory data, the levels of tumor markers and autoantibodies, and clinical manifestations were recorded. Univariate analysis, least absolute shrinkage and selection operator (LASSO), random forest (RF), and partial least square (PLS) were performed, and the receiver operating characteristic (ROC) curves were plotted.

Results

Univariate analysis showed that, compared to RA-non-ILD, patients with RA-ILD were older (p < 0.001), had higher white blood cell (p = 0.003) and neutrophil counts (p = 0.017), had higher erythrocyte sedimentation rate (p = 0.003) and C-reactive protein (p = 0.003), had higher levels of KL-6 (p < 0.001), D-dimer (p < 0.001), fibrinogen (p < 0.001), fibrinogen degradation products (p < 0.001), lactate dehydrogenase (p < 0.001), hydroxybutyrate dehydrogenase (p < 0.001), carbohydrate antigen (CA) 19–9 (p < 0.001), carcinoembryonic antigen (p = 0.001), and CA242 (p < 0.001), but a significantly lower albumin level (p = 0.003). The areas under the curves (AUCs) of the LASSO, RF, and PLS models attained 0.95 in terms of differentiating patients with RA-ILD from those without. When data from the univariate analysis and the top 10 indicators of the three machine learning models were combined, the most discriminatory markers were age and the KL-6, D-dimer, and CA19-9, with AUCs of 0.814 [95% confidence interval (CI) 0.731–0.880], 0.749 (95% CI 0.660–0.824), 0.749 (95% CI 0.660–0.824), and 0.727 (95% CI 0.637–0.805), respectively. When all four markers were combined, the AUC reached 0.928 (95% CI 0.865–0.968). Notably, neither the KL-6 nor the CA19-9 level correlated with disease activity in RA-ILD group.

Conclusions

The levels of KL-6, D-dimer, and tumor markers greatly aided RA-ILD identification. Machine learning algorithms combined with traditional biostatistical analysis can diagnose patients with RA-ILD and identify biomarkers potentially associated with the disease.

Supplementary Information

The online version contains supplementary material available at 10.1186/s13075-022-02800-2.

Associations of lymphocyte subpopulations with clinical phenotypes and long-term outcomes in juvenile-onset systemic lupus erythematosus

OBJECTIVE

Juvenile-onset systemic lupus erythematosus (JSLE) is a complex and heterogeneous immune-mediated disease. Cellular components have crucial roles in disease phenotypes and outcomes. We aimed to determine the associations of lymphocyte subsets with clinical manifestations and long-term outcomes in JSLE patients.

METHODS

A cohort of 60 JSLE patients provided blood samples during active disease, of whom 34 provided further samples during inactive disease. In a longitudinal study, blood samples were obtained from 49 of the JSLE patients at 0, 3, and 6 months. The healthy control (HC) group consisted of 42 age-matched children. Lymphocyte subsets were analyzed by flow cytometry.

RESULTS

The percentages of CD4+ T, γδ T, and NK cells were significantly decreased in JSLE patients compared with HC, while the percentages of CD8+ T, NKT, and CD19+ B cells were significantly increased. The percentage of regulatory T cells (Tregs) was significantly lower in JSLE patients with lupus nephritis (LN) than in non-LN JSLE patients and HC. The patients were stratified into high and low groups by the median frequency of each lymphocyte subset. The γδ T cells high group and NK cells high group were significantly related to mucosal ulcer. The CD4+ T cells high group was significantly associated with arthritis, and the NKT cells high group was substantially linked with autoimmune hemolytic anemia. The CD8+ T cells low group was mainly related to vasculitis, and the Tregs low group was significantly associated with LN. The percentage of Tregs was significantly increased at 6 months of follow-up, and the LN JSLE group had a lower Treg percentage than the non-LN JSLE group. Predictors of remission on therapy were high Tregs, high absolute lymphocyte count, direct Coombs test positivity, and LN absence at enrollment.

CONCLUSION

JSLE patients exhibited altered lymphocyte subsets, which were strongly associated with clinical phenotypes and long-term outcomes.

Signatures of GVHD and relapse after posttransplant cyclophosphamide revealed by immune profiling and machine learning

The key immunologic signatures associated with clinical outcomes after posttransplant cyclophosphamide (PTCy)-based HLA-haploidentical (haplo) and HLA-matched bone marrow transplantation (BMT) are largely unknown. To address this gap in knowledge, we used machine learning to decipher clinically relevant signatures from immunophenotypic, proteomic, and clinical data and then examined transcriptome changes in the lymphocyte subsets that predicted major posttransplant outcomes. Kinetics of immune subset reconstitution after day 28 were similar for 70 patients undergoing haplo and 75 patients undergoing HLA-matched BMT. Machine learning based on 35 candidate factors (10 clinical, 18 cellular, and 7 proteomic) revealed that combined elevations in effector CD4+ conventional T cells (Tconv) and CXCL9 at day 28 predicted acute graft-versus-host disease (aGVHD). Furthermore, higher NK cell counts predicted improved overall survival (OS) due to a reduction in both nonrelapse mortality and relapse. Transcriptional and flow-cytometric analyses of recovering lymphocytes in patients with aGVHD identified preserved hallmarks of functional CD4+ regulatory T cells (Tregs) while highlighting a Tconv-driven inflammatory and metabolic axis distinct from that seen with conventional GVHD prophylaxis. Patients developing early relapse displayed a loss of inflammatory gene signatures in NK cells and a transcriptional exhaustion phenotype in CD8+ T cells. Using a multimodality approach, we highlight the utility of systems biology in BMT biomarker discovery and offer a novel understanding of how PTCy influences alloimmune responses. Our work charts future directions for novel therapeutic interventions after these increasingly used GVHD prophylaxis platforms. Specimens collected on NCT0079656226 and NCT0080927627 https://clinicaltrials.gov/.

A Study of the Recent Trends of Immunology: Key Challenges, Domains, Applications, Datasets, and Future Directions

The human immune system is very complex. Understanding it traditionally required specialized knowledge and expertise along with years of study. However, in recent times, the introduction of technologies such as AIoMT (Artificial Intelligence of Medical Things), genetic intelligence algorithms, smart immunological methodologies, etc., has made this process easier. These technologies can observe relations and patterns that humans do and recognize patterns that are unobservable by humans. Furthermore, these technologies have also enabled us to understand better the different types of cells in the immune system, their structures, their importance, and their impact on our immunity, particularly in the case of debilitating diseases such as cancer. The undertaken study explores the AI methodologies currently in the field of immunology. The initial part of this study explains the integration of AI in healthcare and how it has changed the face of the medical industry. It also details the current applications of AI in the different healthcare domains and the key challenges faced when trying to integrate AI with healthcare, along with the recent developments and contributions in this field by other researchers. The core part of this study is focused on exploring the most common classifications of health diseases, immunology, and its key subdomains. The later part of the study presents a statistical analysis of the contributions in AI in the different domains of immunology and an in-depth review of the machine learning and deep learning methodologies and algorithms that can and have been applied in the field of immunology. We have also analyzed a list of machine learning and deep learning datasets about the different subdomains of immunology. Finally, in the end, the presented study discusses the future research directions in the field of AI in immunology and provides some possible solutions for the same.

An introduction to machine learning and analysis of its use in rheumatic diseases

Machine learning (ML) is a computerized analytical technique that is being increasingly employed in biomedicine. ML often provides an advantage over explicitly programmed strategies in the analysis of multidimensional information by recognizing relationships in the data that were not previously appreciated. As such, the use of ML in rheumatology is increasing, and numerous studies have employed ML to classify patients with rheumatic autoimmune inflammatory diseases (RAIDs) from medical records and imaging, biometric or gene expression data. However, these studies are limited by sample size, the accuracy of sample labelling, and absence of datasets for external validation. In addition, there is potential for ML models to overfit or underfit the data and, thereby, these models might produce results that cannot be replicated in an unrelated dataset. In this Review, we introduce the basic principles of ML and discuss its current strengths and weaknesses in the classification of patients with RAIDs. Moreover, we highlight the successful analysis of the same type of input data (for example, medical records) with different algorithms, illustrating the potential plasticity of this analytical approach. Altogether, a better understanding of ML and the future application of advanced analytical techniques based on this approach, coupled with the increasing availability of biomedical data, may facilitate the development of meaningful precision medicine for patients with RAIDs.

Machine Learning Techniques for Personalised Medicine Approaches in Immune-Mediated Chronic Inflammatory Diseases: Applications and Challenges

In the past decade, the emergence of machine learning (ML) applications has led to significant advances towards implementation of personalised medicine approaches for improved health care, due to the exceptional performance of ML models when utilising complex big data. The immune-mediated chronic inflammatory diseases are a group of complex disorders associated with dysregulated immune responses resulting in inflammation affecting various organs and systems. The heterogeneous nature of these diseases poses great challenges for tailored disease management and addressing unmet patient needs. Applying novel ML techniques to the clinical study of chronic inflammatory diseases shows promising results and great potential for precision medicine applications in clinical research and practice. In this review, we highlight the clinical applications of various ML techniques for prediction, diagnosis and prognosis of autoimmune rheumatic diseases, inflammatory bowel disease, autoimmune chronic kidney disease, and multiple sclerosis, as well as ML applications for patient stratification and treatment selection. We highlight the use of ML in drug development, including target identification, validation and drug repurposing, as well as challenges related to data interpretation and validation, and ethical concerns related to the use of artificial intelligence in clinical research.

Predictive modeling of hypophosphatasia based on a case series of adult patients with persistent hypophosphatasemia

Approximately half of individuals with hypophosphatasemia (low levels of serum alkaline phosphatase) have hypophosphatasia, a rare genetic disease in which patients may have stress fractures, bone and joint pain, or premature tooth loss. We developed a predictive model based on specific biomarkers of this disease to better diagnose this condition.

INTRODUCTION

Hypophosphatasemia is a condition in which low levels of alkaline phosphatase (ALP) are detected in the serum. Some individuals presenting with this condition may have a rare genetic disease called hypophosphatasia (HPP), which involves mineralization of the bone and teeth. Lack of awareness of HPP and its nonspecific symptoms make this genetic disease difficult to diagnose. We developed a predictive model based on biomarkers of HPP such as ALP and pyridoxal 5'-phosphate (PLP), because clinical manifestations sometimes are not recognized as symptoms of HPP.

METHODS

We assessed 325,000 ALP results between 2010 and 2015 to identify individuals suspected of having HPP. We performed univariate and multivariate analyses to characterize the relationship between hypophosphatasemia and HPP. Using several machine learning algorithms, we developed several models based on biomarkers and compared their performance to determine the best model.

RESULTS

The final cohort included 45 patients who underwent a genetic test. Half (23 patients) showed a mutation of the ALPL gene that encodes the tissue-nonspecific ALP enzyme. ALP (odds ratio [OR] 0.61, 95% confidence interval [CI] 0.3-0.8, p = 0.01) and PLP (OR 1.06, 95% CI 1.01-1.15, p = 0.04) were the only variables significantly associated with the presence of HPP. Support vector machines and logistic regression were the machine learning algorithms that provided the best predictive models in terms of classification (area under the curve 0.936 and 0.844, respectively).

CONCLUSIONS

Given the high probability of a misdiagnosis, its nonspecific symptoms, and a lack of awareness of serum ALP levels, it is difficult to make a clinical diagnosis of HPP. Predictive models based on biomarkers are necessary to achieve a proper diagnosis. Our proposed machine learning approaches achieved reasonable performance compared to traditional statistical methods used in biomedicine, increasing the likelihood of properly diagnosing such a rare disease as HPP.

Congenital Heart Disease: An Immunological Perspective

Congenital heart disease (CHD) poses a significant global health and economic burden-despite advances in treating CHD reducing the mortality risk, globally CHD accounts for approximately 300,000 deaths yearly. Children with CHD experience both acute and chronic cardiac complications, and though treatment options have improved, some remain extremely invasive. A challenge in addressing these morbidity and mortality risks is that little is known regarding the cause of many CHDs and current evidence suggests a multifactorial etiology. Some studies implicate an immune contribution to CHD development; however, the role of the immune system is not well-understood. Defining the role of the immune and inflammatory responses in CHD therefore holds promise in elucidating mechanisms underlying these disorders and improving upon current diagnostic and treatment options. In this review, we address the current knowledge coinciding CHDs with immune and inflammatory associations, emphasizing conditions where this understanding would provide clinical benefit, and challenges in studying these mechanisms.

Stratification of Patients With Sjögren's Syndrome and Patients With Systemic Lupus Erythematosus According to Two Shared Immune Cell Signatures, With Potential Therapeutic Implications

OBJECTIVE

Similarities in the clinical and laboratory features of primary Sjögren's syndrome (SS) and systemic lupus erythematosus (SLE) have led to attempts to treat patients with primary SS or SLE with similar biologic therapeutics. However, the results of many clinical trials are disappointing, and no biologic treatments are licensed for use in primary SS, while only a few biologic agents are available to treat SLE patients whose disease has remained refractory to other treatments. With the aim of improving treatment selections, this study was undertaken to identify distinct immunologic signatures in patients with primary SS and patients with SLE, using a stratification approach based on immune cell endotypes.

METHODS

Immunophentyping of 29 immune cell subsets was performed using flow cytometry in peripheral blood from patients with primary SS (n = 45), patients with SLE (n = 29), and patients with secondary SS associated with SLE (SLE/SS) (n = 14), all of whom were considered to have low disease activity or be in clinical remission, and sex-matched healthy controls (n = 31). Data were analyzed using supervised machine learning (balanced random forest, sparse partial least squares discriminant analysis), logistic regression, and multiple t-tests. Patients were stratified by K-means clustering and clinical trajectory analysis.

RESULTS

Patients with primary SS and patients with SLE had a similar immunologic architecture despite having different clinical presentations and prognoses. Stratification of the combined primary SS, SLE, and SLE/SS patient cohorts by K-means cluster analysis revealed 2 endotypes, characterized by distinct immune cell profiles spanning the diagnoses. A signature of 8 T cell subsets that distinctly differentiated the 2 endotypes with high accuracy (area under the curve 0.9979) was identified in logistic regression and machine learning models. In clinical trajectory analyses, the change in damage scores and disease activity levels from baseline to 5 years differed between the 2 endotypes.

CONCLUSION

These findings identify an immune cell toolkit that may be useful for differentiating, with high accuracy, the immunologic profiles of patients with primary SS and patients with SLE as a way to achieve targeted therapeutic approaches.

Patients with Common Variable Immunodeficiency Complicated by Autoimmune Phenomena Have Lymphopenia and Reduced Treg, Th17, and NK Cells

Most patients with primary immune deficiency suffer from recurrent infections; however, paradoxical autoimmune phenomena can also manifest. The aim of this study was to identify immunological markers of autoimmune phenomena associated with common variable immunodeficiency (CVID). The study included 33 adults with CVID divided into two groups: (1) those with noninfectious autoimmune complications (CVID-C (n = 24)) and (2) those with only infectious symptoms (CVID-OI (n = 9)). Flow cytometry of peripheral blood was performed and compared with systemic lupus erythematosus (SLE) patients (n = 17) and healthy controls (n = 20). We found that all lymphocytes were lower in CVID-C and SLE. NK cells were lowest in CVID-C. Th17 cells were significantly reduced in CVID-C and SLE. Tregs were significantly lower in CVID-C and SLE. Bregs did not significantly differ between any groups. Class-switched memory B cells were significantly lower in CVID-C and CVID-OI. Lastly, plasmablasts were significantly higher in SLE. Among the T cell subsets, CVID-C patients had lower naive and recent thymic emigrant CD4+ T cells. In conclusion, reduced Treg, Th17, and NK cells are features of CVID with autoimmune complications, and class-switched memory B cells can help distinguish patients with different causes of autoimmunity. Future studies are needed to confirm whether reductions of Treg, Th17, and NK cells might be a biomarker of more complicated CVID cases.

Biomarkers Associated with Organ-Specific Involvement in Juvenile Systemic Lupus Erythematosus

Juvenile systemic lupus erythematosus (JSLE) is characterised by onset before 18 years of age and more severe disease phenotype, increased morbidity and mortality compared to adult-onset SLE. Management strategies in JSLE rely heavily on evidence derived from adult-onset SLE studies; therefore, identifying biomarkers associated with the disease pathogenesis and reflecting particularities of JSLE clinical phenotype holds promise for better patient management and improved outcomes. This narrative review summarises the evidence related to various traditional and novel biomarkers that have shown a promising role in identifying and predicting specific organ involvement in JSLE and appraises the evidence regarding their clinical utility, focusing in particular on renal biomarkers, while also emphasising the research into cardiovascular, haematological, neurological, skin and joint disease-related JSLE biomarkers, as well as genetic biomarkers with potential clinical applications.