Development and Validation of a MicroRNA Panel to Differentiate Between Patients with Rheumatoid Arthritis or Systemic Lupus Erythematosus and Controls

Advancing precision rheumatology: applications of machine learning for rheumatoid arthritis management

Rheumatoid arthritis (RA) is an autoimmune disease causing progressive joint damage. Early diagnosis and treatment is critical, but remains challenging due to RA complexity and heterogeneity. Machine learning (ML) techniques may enhance RA management by identifying patterns within multidimensional biomedical data to improve classification, diagnosis, and treatment predictions. In this review, we summarize the applications of ML for RA management. Emerging studies or applications have developed diagnostic and predictive models for RA that utilize a variety of data modalities, including electronic health records, imaging, and multi-omics data. High-performance supervised learning models have demonstrated an Area Under the Curve (AUC) exceeding 0.85, which is used for identifying RA patients and predicting treatment responses. Unsupervised learning has revealed potential RA subtypes. Ongoing research is integrating multimodal data with deep learning to further improve performance. However, key challenges remain regarding model overfitting, generalizability, validation in clinical settings, and interpretability. Small sample sizes and lack of diverse population testing risks overestimating model performance. Prospective studies evaluating real-world clinical utility are lacking. Enhancing model interpretability is critical for clinician acceptance. In summary, while ML shows promise for transforming RA management through earlier diagnosis and optimized treatment, larger scale multisite data, prospective clinical validation of interpretable models, and testing across diverse populations is still needed. As these gaps are addressed, ML may pave the way towards precision medicine in RA.

Anti-Inflammatory Effect of High-Density Lipoprotein Blunted by Delivery of Altered MicroRNA Cargo in Patients With Rheumatoid Arthritis

OBJECTIVE

High-density lipoprotein (HDL) has well-characterized anti-atherogenic cholesterol efflux and antioxidant functions. Another function of HDL uncharacterized in rheumatoid arthritis (RA) is its ability to transport microRNAs (miRNAs) between cells and thus alter cellular function. The study's purpose was to determine if HDL-miRNA cargo is altered and affects inflammation in RA.

METHODS

HDL-microRNAs were characterized in 30 RA and 30 control participants by next generation sequencing and quantitative polymerase chain reaction. The most abundant differentially expressed miRNA was evaluated further. The function of miR-1246 was assessed by miRNA mimics, antagomiRs, small interfering RNA knockdown, and luciferase assays. Monocyte-derived macrophages were treated with miR-1246-loaded HDL and unmodified HDL from RA and control participants to measure delivery of miR-1246 and its effect on interleukin-6 (IL-6).

RESULTS

The most abundant miRNA on HDL was miR-1246; it was significantly enriched two-fold on HDL from RA versus control participants. HDL-mediated miR-1246 delivery to macrophages significantly increased IL6 expression 43-fold. miR-1246 delivery significantly decreased DUSP3 1.5-fold and DUSP3 small interfering RNA knockdown increased macrophage IL6 expression. Luciferase assay indicated DUSP3 is a direct target of miR-1246. Unmodified HDL from RA delivered 1.6-fold more miR-1246 versus control participant HDL. Unmodified HDL from both RA and control participants attenuated activated macrophage IL6 expression, but this effect was significantly blunted in RA so that IL6 expression was 3.4-fold higher after RA versus control HDL treatment.

CONCLUSION

HDL-miR-1246 was increased in RA versus control participants and delivery of miR-1246 to macrophages increased IL-6 expression by targeting DUSP3. The altered HDL-miRNA cargo in RA blunted HDL's anti-inflammatory effect.

LncRNA expression profiling in exosomes derived from synovial fluid of patients with rheumatoid arthritis

OBJECTIVE

To identify the long non-coding RNA (lncRNA) expression profiling in exosomes derived from synovial fluid of rheumatoid arthritis (RA) patients, and carry out bioinformatics analysis on target genes of differentially expressed lncRNAs.

METHODS

Exosomes were isolated from synovial fluid via ultracentrifugation. RNAs were extracted from exosomes by using HiPure Liquid RNA/miRNA kits, followed by lncRNA sequencing. Differentially expressed lncRNAs in RA were screened, and bioinformatics analysis of their target genes was carried out. qRT-PCR was used to verify the lncRNA expression levels.

RESULTS

Compared with osteoarthritis (OA), 347 lncRNAs were found differentially expressed in RA. Compared with gout, 805 lncRNAs were found differentially expressed in RA. Compared with both OA and gout, 85 lncRNAs were found specially expressed in RA (65 were upregulated (including ENST00000433825.1)). Functional analysis of target genes of the specially expressed lncRNAs revealed significant enrichment of "autophagy" and "mTOR signaling pathway". The qRT-PCR results indicated that ENST00000433825.1 was highly expressed in RA, compared with both OA and gout (P < 0.05), which matched the lncRNA sequencing results. Correlation analysis showed that the level of ENST00000433825.1 in RA patients was significantly and positively correlated with the level of C-reactive protein (CRP) (P < 0.001).

CONCLUSIONS

The lncRNA expression profiling in exosomes derived from synovial fluid of RA was significantly different from OA and gout. ENST00000433825.1 was highly and uniquely expressed in RA and significantly and positively correlated with CRP, which might provide a diagnostic and therapeutic biomarker for RA.

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.

Structural and Pharmacological Network Focused on MiRNAs Involved in Rheumatoid Arthritis: A Systematic Review

BACKGROUND

Rheumatoid Arthritis (RA) is a chronic autoimmune disease that has a prevalence of over one percent of the world population, causing substantial pain, joint deformity, and functional disability in patients. The identification and measurement of miRNAs are relatively easy to perform. Future studies will corroborate if miRNAs can fulfill their roles as biomarkers with either predictive or diagnostic evaluation of treatment potential and provide actual clinical utility.

METHODS

In the last decade, various advances have been made regarding the identification of the origin and exact functions of miRNAs, allowing us to have a potential use both in the research and clinical fields.

OBJECTIVE

This systematic review aimed to collect, analyze, and improve the current understanding of RA-related miRNAs and their applicability in therapeutics. A bibliographic search of the miRNAs involved in RA was carried out, and through the use of databases, their target genes and small molecules that had some relationship with their expression were searched. The analysis of these data was done through structural network analysis.

RESULTS

During the network analysis, miR-30a, miR-30c, let-7a, miR-144, miR-17-5p, miR-124, miR -23b, miR-23, miR-15a, miR-16 were the most connected, which could be used as possible biomarkers or be candidates for further analysis due to their interaction with other miRNAs and genes.

CONCLUSION

Additionally, this is the first systematic review, in which we proposed that small compounds like toxicants and drugs could have a potential role within RA because they regulate the expression of miRNAs involved in this pathology. Some of these compounds are commonly found as environmental contaminants, and others as drugs. These ideas open a new panorama of understanding RA, proposing possible causes or treatments against this pathology. Therefore, these small molecules would give us some indication of a relationship with RA, thereby helping in seeking causes, treatment, or prevention of this disease.

CONCLUSION

This is the first time it is intended to use structural network analysis to determine possible biomarkers of AR for diagnosis and prognosis through the expression of these miRNAs and their relationship with compounds of daily life.

Identification of a circRNA-miRNA-mRNA network to explore the effects of circRNAs on renal injury in systemic lupus erythematosus

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease. At present, the mechanism of non-coding RNA in renal injury in SLE patients is still unclear. A total of 64 DEcircRNAs, 75 DEmiRNAs, and 249 DEmRNAs were identified. We integrated 10 circRNAs, 10 miRNAs, and 88 target mRNAs into a circRNA-miRNA-mRNA network and obtained 9 hub genes (circ-0000006, miR-766-3p, miR-409-3p, miR-339-3p, miR-331-3p, miR-140-3p, miR-186-5p, miR-149-5p, PSME3). The ROC curve results showed that the diagnostic efficiency of 6 hub miRNA was higher than that of has_circ_0000006 and PSEME3. SsGSEA analysis revealed immune cell composition in SLE and control renal tissues, including 3 types of immune cells up-regulated (gamma delta T cell, effector memory CD4 T cell, central memory CD8 T cell) and 4 types down-regulated (memory B cell, mast cell, macrophage, immature dendritic cell, eosinophil) in SLE patients. In addition, PSME3 was negatively correlated with 3 up-regulated immune cells and positively correlated with 4 down-regulated immune cells in SLE patients. Our study provides a deeper understanding of the circRNA-related competing endogenous RNA regulatory mechanism in the renal injury of systemic lupus erythematosus.

Inhibition of miR-96-5p alleviates intervertebral disc degeneration by regulating the peroxisome proliferator-activated receptor γ/nuclear factor-kappaB pathway

BACKGROUND

Intervertebral disc degeneration (IDD) is the main pathogenesis of low back pain. MicroRNAs (miRNAs) have been found to exert regulatory function in IDD. This study aimed to investigate the effect and potential mechanism of miR-96-5p in IDD.

METHODS

In vitro cell model of IDD was established by treating human nucleus pulposus cells (HNPCs) with interleukin-1β (IL-1β). The level of peroxisome proliferator-activated receptor γ (PPARγ) was examined in the IDD cell model by Western blot and quantification real-time reverse transcription-polymerase chain reaction (qRT-PCR). The expression level of miR-96-5p was detected by RT-qPCR. Effects of PPARγ or/and PPARγ agonist on inflammatory factors, extracellular matrix (ECM), apoptosis, and nuclear factor-kappaB (NF-κB) nuclear translocation were examined through enzyme-linked immunosorbent assay (ELISA), Western blot, flow cytometry assay, and immunofluorescence staining. The Starbase database and dual luciferase reporter assay were used to predict and validate the targeting relationship between miR-96-5p and PPARγ, and rescue assay was performed to gain insight into the role of miR-96-5p on IDD through PPARγ/NF-κB signaling.

RESULTS

PPARγ expression reduced with concentration and time under IL-1β stimulation, while miR-96-5p expression showed the reverse trend (P < 0.05). Upregulation or/and activation of PPARγ inhibited IL-1β-induced the increase in inflammatory factor levels, apoptosis, degradation of the ECM, and the nuclear translocation of NF-κB (P < 0.05). MiR-96-5p was highly expressed but PPARγ was lowly expressed in IDD, while knockdown of PPARγ partially reversed remission of IDD induced by miR-96-5p downregulation (P < 0.05). MiR-96-5p promoted NF-κB entry into the nucleus but PPARγ inhibited this process.

CONCLUSION

Inhibition of miR-96-5p suppressed IDD progression by regulating the PPARγ/NF-κB pathway. MiR-96-5p may be a promising target for IDD treatment clinically.

MicroRNAs in Juvenile Idiopathic Arthritis: State of the Art and Future Perspectives

Juvenile Idiopathic Arthritis (JIA) represents the most common chronic pediatric arthritis in Western countries and a leading cause of disability in children. Despite recent clinical achievements, patient management is still hindered by a lack of diagnostic/prognostic biomarkers and targeted treatment protocols. MicroRNAs (miRNAs) are short non-coding RNAs playing a key role in gene regulation, and their involvement in many pathologies has been widely reported in the literature. In recent decades, miRNA's contribution to the regulation of the immune system and the pathogenesis of autoimmune diseases has been demonstrated. Furthermore, miRNAs isolated from patients' biological samples are currently under investigation for their potential as novel biomarkers. This review aims to provide an overview of the state of the art on miRNA investigation in JIA. The literature addressing the expression of miRNAs in different types of biological samples isolated from JIA patients was reviewed, focusing in particular on their potential application as diagnostic/prognostic biomarkers. The role of miRNAs in the regulation of immune responses in affected joints will also be discussed along with their potential utility as markers of patients' responses to therapeutic approaches. This information will be of value to investigators in the field of pediatric rheumatology, encouraging further research to increase our knowledge of miRNAs' potential for future clinical applications in JIA.

Role of miRNAs in Rheumatoid Arthritis Therapy

Rheumatoid arthritis (RA) is a chronic systemic inflammatory disease characterized by autoimmunity, synovial inflammation and joint destruction. Pannus formation in the synovial cavity can cause irreversible damage to the joint and cartilage and eventually permanent disability. Current conventional treatments for RA have limitations regarding efficacy, safety and cost. microRNA (miRNA) is a type of non-coding RNA (ncRNA) that regulates gene expression at the post-transcriptional level. The dysregulation of miRNA has been observed in RA patients and implicated in the pathogenesis of RA. miRNAs have emerged as potential biomarkers or therapeutic agents. In this review, we explore the role of miRNAs in various aspects of RA pathophysiology, including immune cell imbalance, the proliferation and invasion of fibroblast-like synovial (FLS) cell, the dysregulation of inflammatory signaling and disturbance in angiogenesis. We delve into the regulatory effects of miRNAs on Treg/Th17 and M1/M2 polarization, the activation of the NF-κB/NLRP3 signaling pathway, neovascular formation, energy metabolism induced by FLS-cell-induced energy metabolism, apoptosis, osteogenesis and mobility. These findings shed light on the potential applications of miRNAs as diagnostic or therapeutic biomarkers for RA management. Furthermore, there are some strategies to regulate miRNA expression levels by utilizing miRNA mimics or exosomes and to hinder miRNA activity via competitive endogenous RNA (ceRNA) network-based antagonists. We conclude that miRNAs offer a promising avenue for RA therapy with unlimited potential.

Artificial Intelligence and laboratory data in rheumatic diseases

Artificial intelligence (AI)-based medical technologies are rapidly evolving into actionable solutions for clinical practice. Machine learning (ML) algorithms can process increasing amounts of laboratory data such as gene expression immunophenotyping data and biomarkers. In recent years, the analysis of ML has become particularly useful for the study of complex chronic diseases, such as rheumatic diseases, heterogenous conditions with multiple triggers. Numerous studies have used ML to classify patients and improve diagnosis, to stratify the risk and determine disease subtypes, as well as to discover biomarkers and gene signatures. This review aims to provide examples of ML models for specific rheumatic diseases using laboratory data and some insights into relevant strengths and limitations. A better understanding and future application of these analytical strategies could facilitate the development of precision medicine for rheumatic patients.

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).

miRNAs insights into rheumatoid arthritis: Favorable and detrimental aspects of key performers

Rheumatoid arthritis (RA) is a severe autoimmune inflammation that mostly affects the joints. It's a multifactorial disease. Its clinical picture depends on genetic and epigenetic factors such as miRNAs. The miRNAs are small noncoding molecules that are able to negatively or positively modulate their target gene expression. In RA, miRNAs are linked to its pathogenesis. They disrupt immunity balance by controlling granulocytes, triggering the release of several proinflammatory cytokines such as interleukin-6 and tumor necrosis factor-α, finally leading to synovium hyperplasia and inflammation. Besides, they also may trigger activation of some pathways as nuclear factor kappa-β disrupts the balance between osteoclast and osteoblast activity, leading to increased bone destruction. Moreover, miRNAs are also applied with efficiency in RA diagnosis and prognosis. Besides the significant association between miRNAs and RA response to treatment, they are also applied as a choice for treatment based on their effects on the immune system and inflammatory cytokines. Hence, the review aims to present an updated overview of miRNAs, their biogenesis, implications in RA pathogenesis, and finally, the role of miRNAs in RA treatment.

Comprehensive overview of microRNA function in rheumatoid arthritis

MicroRNAs (miRNAs), a class of endogenous single-stranded short noncoding RNAs, have emerged as vital epigenetic regulators of both pathological and physiological processes in animals. They direct fundamental cellular pathways and processes by fine-tuning the expression of multiple genes at the posttranscriptional level. Growing evidence suggests that miRNAs are implicated in the onset and development of rheumatoid arthritis (RA). RA is a chronic inflammatory disease that mainly affects synovial joints. This common autoimmune disorder is characterized by a complex and multifaceted pathogenesis, and its morbidity, disability and mortality rates remain consistently high. More in-depth insights into the underlying mechanisms of RA are required to address unmet clinical needs and optimize treatment. Herein, we comprehensively review the deregulated miRNAs and impaired cellular functions in RA to shed light on several aspects of RA pathogenesis, with a focus on excessive inflammation, synovial hyperplasia and progressive joint damage. This review also provides promising targets for innovative therapies of RA. In addition, we discuss the regulatory roles and clinical potential of extracellular miRNAs in RA, highlighting their prospective applications as diagnostic and predictive biomarkers.

Identification and comprehensive analysis of circRNA-miRNA-mRNA regulatory networks in osteoarthritis

Osteoarthritis (OA) is a common orthopedic degenerative disease, leading to high disability in activities of daily living. There remains an urgent need to identify the underlying mechanisms and identify new therapeutic targets in OA diagnosis and treatment. Circular RNAs (circRNAs) play a role in the development of multiple diseases. Many studies have reported that circRNAs regulate microRNAs (miRNAs) through an endogenous competitive mechanism. However, it remains unclear if an interplay between circRNAs, miRNAs, and target genes plays a deeper regulatory role in OA. Four datasets were downloaded from the GEO database, and differentially expressed circRNAs (DECs), differentially expressed miRNAs (DEMs), and differentially expressed genes (DEGs) were identified. Functional annotation and pathway enrichment analysis of DEGs and DECs were carried out to determine the main associated mechanism in OA. A protein-protein network (PPI) was constructed to analyze the function of, and to screen out, hub DEGs in OA. Based on the artificial intelligence prediction of protein crystal structures of two hub DEGs, TOP2A and PLK1, digitoxin and oxytetracycline were found to have the strongest affinity, respectively, with molecular docking. Subsequently, overlapping DEMs and miRNAs targeted by DECs obtained target DEMs (DETMs). Intersection of DEGs and genes targeted by DEMs obtained target DEGs (DETGs). Thus, a circRNA-miRNA-mRNA regulatory network was constructed from 16 circRNAs, 32 miRNAs, and 97 mRNAs. Three hub DECs have the largest number of regulated miRNAs and were verified through in vitro experiments. In addition, the expression level of 16 DECs was validated by RT-PCR. In conclusion, we constructed a circRNA-miRNA-mRNA regulatory network in OA and three new hub DECs, hsa_circ_0027914, hsa_circ_0101125, and hsa_circ_0102564, were identified as novel biomarkers for OA.

MicroRNA-183/96/182 cluster in immunity and autoimmunity

MicroRNAs (miRNAs) are crucial post-transcriptional regulators of gene expression in ubiquitous biological processes, including immune-related pathways. This review focuses on the miR-183/96/182 cluster (miR-183C), which contains three miRNAs, miR-183, -96, and -182, having almost identical seed sequences with minor differences. The similarity among seed sequences allows these three miRNAs to act cooperatively. In addition, their minor differences permit them to target distinct genes and regulate unique pathways. The expression of miR-183C was initially identified in sensory organs. Subsequently, abnormal expression of miR-183C miRNAs in various cancers and autoimmune diseases has been reported, implying their potential role in human diseases. The regulatory effects of miR-183C miRNAs on the differentiation and function of both innate and adaptive immune cells have now been documented. In this review, we have discussed the complex role of miR-183C in the immune cells in both normal and autoimmune backgrounds. We highlighted the dysregulation of miR-183C miRNAs in several autoimmune diseases, including systemic lupus erythematosus (SLE), multiple sclerosis (MS), and ocular autoimmune disorders, and discussed the potential for utilizing miR-183C as biomarkers and therapeutic targets of specific autoimmune diseases.

Peripheral blood mononuclear cell microRNAs are novel biomarkers for diagnosing and monitoring Crohn's disease

Crohn's disease is a recurrent, progressive, immune-mediated inflammatory disease and merely manifests non-specific symptoms at early stage. In this study, we isolated peripheral blood mononuclear cells (PBMCs) to determine whether PBMC miRNAs are reliable biomarkers for Crohn's disease diagnosing and monitoring. 5 Crohn's disease patients and 5 healthy controls were recruited to find differentially expressed miRNAs by next generation sequencing. Candidate PBMC miRNAs were further validated by qRT-PCR in another cohort consisting of 86 Crohn's disease patients and 39 healthy controls. We found PBMC miR-582-5p could diagnose Crohn's disease with the area under receiver operating characteristic curve (AUROC) of 0.701(95%CI 0.606-0.796, p < .001). While PBMC miR-96-5p was significantly higher in active Crohn's disease and correlated with both clinical (ρ = 0.376, p < .001) and endoscopic activity (ρ = 0.512, p = .015). Furthermore, PBMC miR-96-5p had a better performance in recognizing active Crohn's disease with AUROC of 0.727 (95%CI 0.609-0.844, p = .001) than C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and fecal calprotectin. In conclusion, PBMC miR-582-5p may be further utilized as a diagnostic biomarker, while miR-96-5p may be a novel and valuable biomarker in monitoring disease activity.

Artificial Intelligence in Rheumatoid Arthritis: Current Status and Future Perspectives: A State-of-the-Art Review

Investigation of the potential applications of artificial intelligence (AI), including machine learning (ML) and deep learning (DL) techniques, is an exponentially growing field in medicine and healthcare. These methods can be critical in providing high-quality care to patients with chronic rheumatological diseases lacking an optimal treatment, like rheumatoid arthritis (RA), which is the second most prevalent autoimmune disease. Herein, following reviewing the basic concepts of AI, we summarize the advances in its applications in RA clinical practice and research. We provide directions for future investigations in this field after reviewing the current knowledge gaps and technical and ethical challenges in applying AI. Automated models have been largely used to improve RA diagnosis since the early 2000s, and they have used a wide variety of techniques, e.g., support vector machine, random forest, and artificial neural networks. AI algorithms can facilitate screening and identification of susceptible groups, diagnosis using omics, imaging, clinical, and sensor data, patient detection within electronic health record (EHR), i.e., phenotyping, treatment response assessment, monitoring disease course, determining prognosis, novel drug discovery, and enhancing basic science research. They can also aid in risk assessment for incidence of comorbidities, e.g., cardiovascular diseases, in patients with RA. However, the proposed models may vary significantly in their performance and reliability. Despite the promising results achieved by AI models in enhancing early diagnosis and management of patients with RA, they are not fully ready to be incorporated into clinical practice. Future investigations are required to ensure development of reliable and generalizable algorithms while they carefully look for any potential source of bias or misconduct. We showed that a growing body of evidence supports the potential role of AI in revolutionizing screening, diagnosis, and management of patients with RA. However, multiple obstacles hinder clinical applications of AI models. Incorporating the machine and/or deep learning algorithms into real-world settings would be a key step in the progress of AI in medicine.

MicroRNA-Mediated Epigenetic Regulation of Rheumatoid Arthritis Susceptibility and Pathogenesis

MicroRNAs (miRNAs) play crucial roles in regulating the transcriptome and development of rheumatoid arthritis (RA). Currently, a comprehensive map illustrating how miRNAs regulate transcripts, pathways, immune system differentiation, and their interactions with terminal cells such as fibroblast-like synoviocytes (FLS), immune-cells, osteoblasts, and osteoclasts are still laking. In this review, we summarize the roles of miRNAs in the susceptibility, pathogenesis, diagnosis, therapeutic intervention, and prognosis of RA. Numerous miRNAs are abnormally expressed in cells involved in RA and regulate target genes and pathways, including NF-κB, Fas-FasL, JAK-STAT, and mTOR pathways. We outline how functional genetic variants of miR-499 and miR-146a partly explain susceptibility to RA. By regulating gene expression, miRNAs affect T cell differentiation into diverse cell types, including Th17 and Treg cells, thus constituting promising gene therapy targets to modulate the immune system in RA. We summarize the diagnostic and prognostic potential of blood-circulating and cell-free miRNAs, highlighting the opportunity to combine these miRNAs with antibodies to cyclic citrullinated peptide (ACCP) to allow accurate diagnosis and prognosis, particularly for seronegative patients. Furthermore, we review the evidence implicating miRNAs as promising biomarkers of efficiency and response of, and resistance to, disease-modifying anti-rheumatic drugs and immunotherapy. Finally, we discuss the autotherapeutic effect of miRNA intervention as a step toward the development of miRNA-based anti-RA drugs. Collectively, the current evidence supports miRNAs as interesting targets to better understand the pathogenetic mechanisms of RA and design more efficient therapeutic interventions.

Circulating Mir-140 and leptin improve the accuracy of the differential diagnosis between psoriatic arthritis and rheumatoid arthritis: a case-control study

The differential diagnosis of psoriatic arthritis (PsA) and rheumatoid arthritis (RA) is difficult because of the lack of diagnostic clinical signs and reliable biomarkers. This study investigated microRNAs (miRNA) and adipokines as potential additional markers to discriminate PsA from RA. The expression profile of miRNA (miR-21, miR-140, miR-146a, miR-155, miR-181b, miR-223, miR-let-7e) and inflammatory cytokines (IL-1β, IL-6, IL-17a, IL-23a, TNF-α) from peripheral blood mononuclear cells of PsA and RA patients compared to healthy controls (HC) were evaluated by real-time PCR, and serum adipokines (adiponectin, chemerin, leptin, resistin, visfatin) and cytokines by ELISA assay. Univariable binary logistic regression was used to find the association between PsA and potential predictors. The gene expression of miRNA and cytokines and the serum levels of adipokines were found significantly different in PsA and RA patients compared to HC, as well as in PsA versus RA. MiR-140 gene expression resulted up-regulated in PsA patients and reduced in RA in comparison to HC, and, for the first time, significantly higher in PsA compared with RA. Serum levels of IL-23a and leptin were significantly increased in PsA and RA populations than in HC, as well as in PsA versus RA. Furthermore, circulating TNF-α was up-regulated in PsA and RA in comparison to controls, while resulted higher in RA than in PsA. Univariable binary logistic regression analysis found the above-mentioned markers associated to PsA versus RA. Our results first demonstrated an increased expression of circulating miR-140 and serum leptin in PsA patients compared to RA, which were identified as potential additional biomarkers to discriminate PsA from RA. Since the differential diagnosis of PsA and RA poses challenges in clinical practice, our data may help to enhance the diagnostic performance of PsA in daily practice.

Machine learning approaches to improve disease management of patients with rheumatoid arthritis: review and future directions

INTRODUCTION

Although the management of rheumatoid arthritis (RA) has improved in major way over the last decades, this disease still leads to an important burden for patients and society, and there is a need to develop more personalized approaches. Machine learning (ML) methods are more and more used in health-related studies and can be applied to different sorts of data (clinical, radiological, or 'omics' data). Such approaches may improve the management of patients with RA.

AREAS COVERED

In this paper, we propose a review regarding ML approaches applied to RA. A scoping literature search was performed in PubMed, in September 2021 using the following MeSH terms: 'arthritis, rheumatoid' and 'machine learning'. Based on this search, the usefulness of ML methods for RA diagnosis, monitoring, and prediction of response to treatment and RA outcomes, is discussed.

EXPERT OPINION

ML methods have the potential to revolutionize RA-related research and improve disease management and patient care. Nevertheless, these models are not yet ready to contribute fully to rheumatologists' daily practice. Indeed, these methods raise technical, methodological, and ethical issues, which should be addressed properly to allow their implementation. Collaboration between data scientists, clinical researchers, and physicians is therefore required to move this field forward.

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.

The non-coding RNA interactome in joint health and disease

Non-coding RNAs have distinct regulatory roles in the pathogenesis of joint diseases including osteoarthritis (OA) and rheumatoid arthritis (RA). As the amount of high-throughput profiling studies and mechanistic investigations of microRNAs, long non-coding RNAs and circular RNAs in joint tissues and biofluids has increased, data have emerged that suggest complex interactions among non-coding RNAs that are often overlooked as critical regulators of gene expression. Identifying these non-coding RNAs and their interactions is useful for understanding both joint health and disease. Non-coding RNAs regulate signalling pathways and biological processes that are important for normal joint development but, when dysregulated, can contribute to disease. The specific expression profiles of non-coding RNAs in various disease states support their roles as promising candidate biomarkers, mediators of pathogenic mechanisms and potential therapeutic targets. This Review synthesizes literature published in the past 2 years on the role of non-coding RNAs in OA and RA with a focus on inflammation, cell death, cell proliferation and extracellular matrix dysregulation. Research to date makes it apparent that 'non-coding' does not mean 'non-essential' and that non-coding RNAs are important parts of a complex interactome that underlies OA and RA.

Plasma miRNAs improve the prediction of coronary atherosclerosis in patients with rheumatoid arthritis

OBJECTIVE

MicroRNAs (miRNAs) regulate gene expression and are disease biomarkers. Rheumatoid arthritis (RA) patients have accelerated atherosclerosis leading to excess cardiovascular morbidity and mortality, but traditional risk factors for cardiovascular risk stratification are inadequate. In the general population, miRNAs improve cardiovascular risk estimation beyond traditional risk factors. Our objective was to develop a miRNA panel that predicts coronary atherosclerosis in RA patients.

METHODS

Plasma small RNA next-generation sequencing (NGS) was performed on 161 RA patients whose Agatston scores for coronary artery calcium were previously measured. Random forest analysis of plasma NGS miRNA expression was used to determine which miRNAs best differentiated between those patients with and without coronary artery calcium. Top predictive miRNAs were assayed by quantitative PCR (qPCR). Elastic net regression was used to develop the most parsimonious models with qPCR-measured miRNA concentrations and clinical variables (age, sex, ACC/AHA 10-year risk score, DAS28 score, and diabetes) separately to predict the presence of coronary artery calcium and high coronary artery calcium. C-statistics were used to assess performance model performance.

RESULTS

The top miRNAs which differentiated those with and without coronary atherosclerosis based on random forest analysis included let-7c-5p, miR-30e-5p, miR-30c-5p, miR-4446-3p, miR-126-5p, miR-3168, miR-425-5p, miR-126-3p, miR-30a-5p, and miR-125a-5p. For coronary artery calcium prediction, addition of all miRNAs except miR-126-3p to clinical factors improved the c-statistic modestly from 0.86 to 0.87. For high coronary artery calcium prediction, addition of all miRNAs except miR-30c-5p to clinical factors improved the c-statistic from 0.75 to 0.80.

CONCLUSION

A plasma miRNA panel improved the prediction of high coronary artery calcium beyond traditional risk factors and RA disease activity. Further evaluation of the miRNA panel for prediction of coronary events in RA is necessary. Key Point • A plasma microRNA panel including let-7c-5p, miR-30a-5p, miR-30e-5p, miR-125a-5p, miR-126-3p, miR-126-5p, miR-425-5p, miR-3168, and miR-4446-3p improved the prediction of high coronary artery calcium beyond clinical factors in patients with rheumatoid arthritis.

The Potential Importance of MicroRNAs as Novel Indicators How to Manage Patients with Juvenile Idiopathic Arthritis More Effectively

Small, noncoding sequences of ribonucleic acid called microRNAs (miRNAs, miR) are functioning as posttranscriptional regulators of gene expression. As they draw increasing attention of rheumatologists, there is a growing body of evidence concerning specific molecules that may affect the long-term care of patients with inflammatory arthritides. Findings involving children with juvenile idiopathic arthritis (JIA) are still limited though. The aim of the study was to browse the available data on microRNAs which may be utilized as potential biomarkers helpful in diagnosing and monitoring JIA patients. The review contains a brief summary on the most studied sequences: miR-16, miR-125a-5p, miR-146a, miR-155, and miR-223. It is complemented with other miRNAs possibly relevant for JIA (miR-145, miR-23b, miR-27a, and miR-204) and discussion on challenges for using miRNAs in pediatric rheumatology (particularly, issues regarding specificity of biomarkers and measurements involving synovial fluid).

Study on Omentin-1 and miR-502-3p in osteoporotic fracture

OBJECTIVES

To explore the expression and correlation of Omentin-1 and miR-502-3p in serum of patients with osteoporotic fracture (OPF).

METHODS

Sixty OPF patients diagnosed and treated in our hospital from June 2018 to December 2019 were included in group A. Fifty-six osteoporosis patients without fractures were included in group B. Omentin-1 and miR-502-3p levels were detected by enzyme-linked immunosorbent assay (ELISA) and real-time quantitative PCR (qRT-PCR). Their predictive value for diagnostic efficiency was assessed by ROC curve. Spearman's rank correlation test was used for correlation analysis. The risk factors related to the prognosis of OPF were analyzed by Logistic univariate and multivariate analysis.

RESULTS

The expression of Omentin-1 and miR-502-3p in group A was markedly lower than in group B (P<0.001). Spearman correlation analysis showed that in OPF, there was a negative correlation between serum Omentin-1 and TNF-α (r=0.8579, P<0.001), a negative correlation between serum miR-502-3p and TNF-α (r= 0.8653, P<0.001), and a positive correlation between serum Omentin-1 and miR-502-3p (r= 0.8764, P<0.001).

CONCLUSIONS

Omentin-1 and miR-502-3p were down-regulated in serum of patients with OPF, both of which could be used as potential biomarkers for the diagnosis and disease evaluation of OPF.

isomiRs-Hidden Soldiers in the miRNA Regulatory Army, and How to Find Them?

Numerous studies on microRNAs (miRNA) in cancer and other diseases have been accompanied by diverse computational approaches and experimental methods to predict and validate miRNA biological and clinical significance as easily accessible disease biomarkers. In recent years, the application of the next-generation deep sequencing for the analysis and discovery of novel RNA biomarkers has clearly shown an expanding repertoire of diverse sequence variants of mature miRNAs, or isomiRs, resulting from alternative post-transcriptional processing events, and affected by (patho)physiological changes, population origin, individual's gender, and age. Here, we provide an in-depth overview of currently available bioinformatics approaches for the detection and visualization of both mature miRNA and cognate isomiR sequences. An attempt has been made to present in a systematic way the advantages and downsides of in silico approaches in terms of their sensitivity and accuracy performance, as well as used methods, workflows, and processing steps, and end output dataset overlapping issues. The focus is given to the challenges and pitfalls of isomiR expression analysis. Specifically, we address the availability of tools enabling research without extensive bioinformatics background to explore this fascinating corner of the small RNAome universe that may facilitate the discovery of new and more reliable disease biomarkers.

Dysregulated microRNA expression in rheumatoid arthritis families-a comparison between rheumatoid arthritis patients, their first-degree relatives, and healthy controls

OBJECTIVE

Recent studies have demonstrated an altered expression of certain microRNAs in patients with rheumatoid arthritis (RA) as well as their first-degree relatives (FDRs) compared to healthy controls (HCs), suggesting a role of microRNA in the progression of the disease. To corroborate this, a set of well-characterized RA families originating from northern Sweden were analyzed for differential expression of a selected set of microRNAs.

METHOD

MicroRNA was isolated from frozen peripheral blood cells obtained from 21 different families and included 26 RA patients, 22 FDRs, and 21 HCs. Expression of the selected microRNAs miR-22-3p, miR-26b-5p, miR-34a-3p, miR-103a-3p, miR-142-3p, miR-146a-5p, miR-155, miR-346, and miR-451a was determined by a two-step quantitative real-time polymerase chain reaction (qRT-PCR). Statistical analysis including clinical variables was applied.

RESULTS

Out of the nine selected microRNAs that previously have been linked to RA, we confirmed four after adjusting for age and gender, i.e., miR-22-3p (p = 0.020), miR-26b-5p (p = 0.018), miR-142-3p (p = 0.005), and miR-155 (p = 0.033). Moreover, a significant trend with an intermediate microRNA expression in FDR was observed for the same four microRNAs. In addition, analysis of the effect of corticosteroid use showed modulation of miR-103a-3p expression.

CONCLUSIONS

We confirm that microRNAs seem to be involved in the development of RA, and that the expression pattern in FDR is partly overlapping with RA patients. The contribution of single microRNAs in relation to the complex network including all microRNAs and other molecules is still to be revealed. Key Points • Expression levels of miR-22-3p, miR-26b-5p, miR-142-3p, and miR-155 were significantly altered in RA patients compared to those in controls. • In first-degree relatives, a significant trend with an intermediate microRNA expression in FDR was observed for the same four microRNAs.

MiR-221-5p is involved in the regulation of inflammatory responses in acute gouty arthritis by targeting IL-1β

AIM

Gout is caused by the accumulation of deposited monosodium urate (MSU) crystals in the joints. Recent studies have shown that interleukin-1β (IL-1β) is a key inflammatory mediator of acute gouty arthritis (AGA), and its level is regulated by microRNAs (miRNAs). The purpose of this study was to study the role of miR-221-5p in the pathogenesis of AGA.

METHODS

One hundred patients with AGA and 94 healthy individuals were recruited. The expression of serum miR-221-5p was determined by quantitative real-time polymerase chain reaction. The receiver operating curve (ROC) was applied for diagnostic value analysis. A luciferase reporter assay was performed to confirm the interaction of miRNA and the 3'-untranslated region (UTR) of IL-1β. Enzyme-linked immunosorbent assay was used to detect serum and proinflammatory factors.

RESULTS

miR-221-5p had lower expression in the serum of AGA patients. The area under the curve was 0.884, the sensitivity was 82.0%, and the specificity was 80.9%. Serum miR-221-5p was negatively correlated with the expression levels of visual analog scale and IL-1β. Cell experiments showed that overexpression of miR-221-5p significantly inhibited the expression of inflammatory factors tumor necrosis factor-α, IL-8, and IL-1β, while down-regulation of miR-221-5p was the opposite. Luciferase analysis showed that IL-1β was the target gene of miR-221-5p.

CONCLUSIONS

This study confirmed that miR-221-5p regulates the production of inflammatory cytokines during the pathogenesis of AGA. These results suggested that miR-221-5p could be used as a potential therapeutic target for the treatment of AGA.

The Endogenous Plasma Small RNAome of Rheumatoid Arthritis

OBJECTIVE

Small RNA (sRNA) sequencing has revealed new sRNA classes beyond microRNAs (miRNAs). These sRNAs can regulate genes and act as biomarkers. The aim of this study was to determine if the endogenous plasma sRNA landscape is altered in patients with rheumatoid arthritis (RA) compared with control subjects and to determine its association with disease-related parameters in RA.

METHODS

sRNA sequencing was performed on plasma from 165 RA and 90 control subjects who were frequency-matched for age, race, and sex. Endogenous sRNAs, such as miRNAs, isomiRs, sRNAs derived from small nuclear RNAs (snDRs), small nucleolar RNAs (snoDRs), Y RNAs (yDRs), transfer-derived RNAs (tDRs), long noncoding RNAs (lncDRs) as well as miscellaneous sRNAs (miscRNAs), were quantified using Tools for Integrative Genome analysis of Extracellular sRNAs (TIGER). Individual and categories of sRNAs were compared between RA and controls, and significantly altered sRNAs and sRNA categories were correlated with disease activity and general laboratory measures in RA.

RESULTS

Patients with RA had more miRNAs (1.42-fold, P = 0.01), more tDRs (1.14-fold, P = 0.04), and fewer yDRs (-1.41-fold, P = 0.009) compared with control subjects. Disease duration was inversely associated with yDRs. Disease-related parameters, such as Disease Activity Score-28 (DAS28), swollen joint count, and inflammatory markers were significantly positively associated with tDRs and miscRNAs, and miR-22-3p and related sequences and isomiRs were most significantly associated with DAS28.

CONCLUSION

Endogenous plasma sRNAs are altered in RA compared with control subjects. Although individual miRNAs have been well studied and many are excellent biomarkers in RA, several non-miRNA sRNAs were significantly associated with disease-related parameters as classes and may represent novel biomarkers for RA.