Integrative Clinical, Molecular, and Computational Analysis Identify Novel Biomarkers and Differential Profiles of Anti-TNF Response in Rheumatoid Arthritis

Artificial intelligence to predict treatment response in rheumatoid arthritis and spondyloarthritis: a scoping review

To analyse the types and applications of artificial intelligence (AI) technologies to predict treatment response in rheumatoid arthritis (RA) and spondyloarthritis (SpA). A comprehensive search in Medline, Embase, and Cochrane databases (up to August 2024) identified studies using AI to predict treatment response in RA and SpA. Data on study design, AI methodologies, data sources, and outcomes were extracted and synthesized. Findings were summarized descriptively. Of the 4257 articles identified, 89 studies met the inclusion criteria (74 on RA, 7 on SpA, 4 on Psoriatic Arthritis and 4 a mix of them). AI models primarily employed supervised machine learning techniques (e.g., random forests, support vector machines), unsupervised clustering, and deep learning. Data sources included electronic medical records, clinical biomarkers, genetic and proteomic data, and imaging. Predictive performance varied by methodology, with accuracy ranging from 60 to 70% and AUC values between 0.63 and 0.92. Multi-omics approaches and imaging-based models showed promising results in predicting responses to biologic DMARDs and JAK inhibitors but methodological heterogeneity limited generalizability. AI technologies exhibit substantial potential in predicting treatment responses in RA and SpA, enhancing personalized medicine. However, challenges such as methodological variability, data integration, and external validation remain. Future research should focus on refining AI models, ensuring their robustness across diverse patient populations, and facilitating their integration into clinical practice to optimize therapeutic decision-making in rheumatology.

Neutrophil Activation Markers and Rheumatoid Arthritis Treatment Response to the JAK1/2 Inhibitor Baricitinib

OBJECTIVE

Neutrophils play an important role in regulating immune and inflammatory responses in patients with rheumatoid arthritis (RA). We assessed whether baricitinib, a JAK1/JAK2 inhibitor, could reduce neutrophil activation and whether a neutrophil activation score could predict treatment response.

METHODS

Markers of neutrophil activation, calprotectin, and neutrophil extracellular traps (NETs) were analyzed using enzyme-linked immunosorbent assay in plasma from patients with RA (n = 271) and healthy controls (n = 39). For patients with RA, neutrophil activation markers were measured at baseline, 12 weeks, and 24 weeks after receiving placebo and 2 and 4 mg baricitinib. Whole-blood RNA analyses from multiple randomized baricitinib RA trials were performed to study neutrophil-related transcripts (n = 1,651).

RESULTS

Baseline levels of plasma neutrophil markers were elevated in patients with RA compared to healthy controls (P < 0.001). Receiving baricitinib reduced levels of soluble calprotectin at 12 and 24 weeks, especially in patients with RA responding to treatment, as determined by American College of Rheumatology 20% improvement criteria. Whole-blood RNA analysis revealed similar changes in the predominant neutrophil markers calprotectin and Fcα receptor I upon reception of baricitinib in three randomized clinical trials involving patients with at various stages of disease-modifying therapy. Clustering analysis of plasma activation markers showed elevated levels of calprotectin and NETs (eg, a neutrophil activation score, at baseline, could predict treatment response to baricitinib). In contrast, C-reactive protein levels could not distinguish between responders and nonresponders.

CONCLUSION

Neutrophil activation markers may add clinical value in predicting treatment response to baricitinib and other drugs targeting RA. This study supports personalized medicine in treating patients with RA, not only based on symptoms but also based on immunophenotyping.

Machine learning in the prediction of treatment response in rheumatoid arthritis: A systematic review

OBJECTIVE

This study aimed to investigate the current status and performance of machine learning (ML) approaches in providing reproducible treatment response predictions.

METHODS

This systematic review was conducted in accordance with the PRISMA statement and the CHARMS checklist. We searched PubMed, Cochrane Library, Web of Science, Scopus, and EBSCO databases for cohort studies that derived and/or validated ML models focused on predicting rheumatoid arthritis (RA) treatment response. We extracted data and critically appraised studies based on the Transparent Reporting of a Multivariable Prediction Model for Individual Prognosis or Diagnosis (TRIPOD) and Prediction Model Risk of Bias Assessment Tool (PROBAST) guidelines.

RESULTS

From 210 unduplicated records identified by the literature search, we retained 29 eligible studies. Of these studies, 10 developed a predictive model and reported a mean adherence to the TRIPOD guidelines of 45.6 % (95 % CI: 38.3-52.8 %). The remaining 19 studies not only developed a predictive model but also validated it externally, with a mean adherence of 42.9 % (95 % CI: 39.1-46.6 %). Most of the articles had an unclear risk of bias (41.4 %), followed by a high risk of bias, which was present in 37.9 %.

CONCLUSIONS

In recent years, ML methods have been increasingly used to predict treatment response in RA. Our critical appraisal revealed unclear and high risk of bias in most of the identified models, suggesting that researchers can do more to address the risk of bias and increase transparency, including the use of calibration measures and reporting methods for handling missing data.

FUNDING

None.

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.

Neutrophilic Activity Biomarkers (Plasma Neutrophil Extracellular Traps and Calprotectin) in Established Patients with Rheumatoid Arthritis Receiving Biological or JAK Inhibitors: A Clinical and Ultrasonographic Study

INTRODUCTION

This study assesses the accuracy of neutrophil activation markers, including neutrophil extracellular traps (NETs) and calprotectin, as biomarkers of disease activity in patients with established rheumatoid arthritis (RA). We also analyse the relationship between NETs and various types of therapies as well as their association with autoimmunity.

METHODS

Observational cross-sectional study of patients with RA receiving treatment with biological disease-modifying antirheumatic drugs or Janus kinase inhibitors (JAK-inhibitors) for at least 3 months. Plasma calprotectin levels were measured using an enzyme-linked immunosorbent assay test kit and NETs by measuring their remnants in plasma (neutrophil elastase-DNA and histone-DNA complexes). We also assessed clinical disease activity, joint ultrasound findings and autoantibody status [reumatoid factor (RF), anti-citrullinated peptide/protein antibodies (ACPAs) and anti-carbamylated protein (anti-CarP)]. Associations between neutrophilic biomarkers and clinical or ultrasound scores were sought using correlation analysis. The discriminatory capacity of both neutrophilic biomarkers to detect ultrasound synovitis was analysed through receiver-operating characteristic (ROC) curves.

RESULTS

One hundred fourteen patients were included. Two control groups were included to compare NET levels. The active control group consisted of 15 patients. The second control group consisted of 30 healthy subjects. Plasma NET levels did not correlate with clinical disease status, regardless of the clinic index analysed or the biological therapy administered. No significant correlation was observed between NET remnants and ultrasound synovitis. There was no correlation between plasma NET and autoantibodies. In contrast, plasma calprotectin positively correlated with clinical parameters (swollen joint count [SJC] rho = 0.49; P < 0.001, Clinical Disease Activity Index [CDAI] rho = 0.30; P < 0.001) and ultrasound parameters (rho > 0.50; P < 0.001). Notably, this correlation was stronger than that observed with acute phase reactants.

CONCLUSION

While NET formation induced by neutrophils may play a role in RA pathogenesis, our study raises questions about the utility of NET remnants in peripheral circulation as a biomarker for inflammatory activity. In contrast, this study strongly supports the usefulness of calprotectin as a biomarker of inflammatory activity in patients with RA.

Towards Personalized Medicine in Rheumatoid Arthritis

Rheumatoid arthritis (RA) is a chronic, incurable, multisystem, inflammatory disease characterized by synovitis and extra-articular features. Although several advanced therapies targeting inflammatory mechanisms underlying the disease are available, no advanced therapy is universally effective. Therefore, a ceiling of treatment response is currently accepted where no advanced therapy is superior to another. The current challenge for medical research is the discovery and integration of predictive markers of drug response that can be used to personalize medicine so that the patient is started on "the right drug at the right time". This review article summarizes our current understanding of predicting response to anti-rheumatic drugs in RA, obstacles impeding the development of personalized medicine approaches and future research priorities to overcome these barriers.

Personalized cardiovascular risk assessment in Rheumatoid Arthritis patients using circulating molecular profiles and their modulation by TNFi, IL6Ri, and JAKinibs

BACKGROUND & OBJECTIVES

This study aimed to: 1) analyze the inflammatory profile of Rheumatoid Arthritis (RA) patients, identifying clinical phenotypes associated with cardiovascular (CV) risk; 2) evaluate biologic and targeted-synthetic disease-modifying antirheumatic drugs (b-DMARDs and ts-DMARDs': TNFi, IL6Ri, JAKinibs) effects; and 3) characterize molecular mechanisms in immune-cell activation and endothelial dysfunction.

PATIENTS & METHODS

A total of 387 RA patients and 45 healthy donors were recruited, forming three cohorts: i) 208 RA patients with established disease but without previous CV events; ii) RA-CVD: 96 RA patients with CV events, and iii) 83 RA patients treated with b-DMARDs/ts-DMARDs for 6 months. Serum inflammatory profiles (cytokines/chemokines/growth factors) and NETosis/oxidative stress-linked biomolecules were evaluated. Mechanistic in vitro studies were performed on monocytes, neutrophils and endothelial cells (EC).

RESULTS

In the first RA-cohort, unsupervised clustering unveiled three distinct groups: cluster 3 (C3) displayed the highest inflammatory profile, significant CV-risk score, and greater atheroma plaques prevalence. In contrast, cluster 1 (C1) exhibited the lowest inflammatory profile and CV risk score, while cluster 2 (C2) displayed an intermediate phenotype. Notably, 2nd cohort RA-CVD patients mirrored C3's inflammation. Treatment with b-DMARDs or ts-DMARDs effectively reduced disease-activity scores (DAS28) and restored normal biomolecules levels, controlling CV risk. In vitro, serum from C3-RA or RA-CVD patients increased neutrophils activity and CV-related protein levels in cultured monocytes and EC, which were partially prevented by pre-incubation with TNFi, IL6Ri, and JAKinibs.

CONCLUSIONS

Overall, analyzing circulating molecular profiles in RA patients holds potential for personalized clinical management, addressing CV risk and assisting healthcare professionals in tailoring treatment, ultimately improving outcomes.

Neutrophils, neutrophil extracellular traps, and rheumatoid arthritis: An updated review for clinicians

Rheumatoid arthritis (RA) is a systemic autoimmune disease characterized by the presence of autoantibodies. Research on the pathogenic mechanisms involved in systemic autoimmune diseases has largely focused on the involvement of the adaptive immune system with dysregulated responses of T and B cells. However, in recent years, there is increasing evidence of the significant role played by the innate immune system, particularly neutrophils, in these diseases, particularly in RA. Neutrophil extracellular traps (NETs) are extracellular structures composed of remodeled and concentrated chromatin with DNA, histones, and neutrophil proteins, and were first described in 2004. It has been studied that NETs may play a pathogenic role in RA and could be a source of autoantigens, increasing the immune response in the form of autoantibodies in this disease. The possible role of NETs and other markers of neutrophil activation as biomarkers of activity in RA and other immune-mediated diseases has also been studied. This article reviews the role of NETs in RA. It discusses the role of neutrophils and the latest advances in NETs, especially their involvement in autoimmune phenomena in RA. Finally, a literature review is conducted on the determination of NETs in peripheral blood and their relationship as a biomarker of RA activity, as well as their potential role in disease monitoring.

Biomarkers to guide immunomodulatory treatment: where do we stand?

INTRODUCTION

This review summarizes current progress in the development of biomarkers to guide immunotherapy in oncology, rheumatology, and critical illness.

AREAS COVERED

An extensive literature search was performed about biomarkers classifying patients' immune responses to guide immunotherapy in oncology, rheumatology, and critical illness. Surface markers, such as programmed death-ligand 1 (PD-L1), genetic biomarkers, such as tumor mutation load, and circulating tumor DNA are biomarkers associated with the effectiveness of immunotherapy in oncology. Genomics, metabolomics, and proteomics play a crucial role in selecting the most suitable therapeutic options for rheumatologic patients. Phenotypes and endotypes are a promising approach to detect critically ill patients with hyper- or hypo-inflammation. Sepsis trials using biomarkers such as ferritin, lymphopenia, HLA-DR expression on monocytes and PD-L1 to guide immunotherapy have been already conducted or are currently ongoing. Immunotherapy in COVID-19 pneumonia, guided by C-reactive protein and soluble urokinase plasminogen activator receptor (suPAR) has improved patient outcomes globally. More research is needed into immunotherapy in other critical conditions.

EXPERT OPINION

Targeted immunotherapy has improved outcomes in oncology and rheumatology, paving the way for precision medicine in the critically ill. Transcriptomics will play a crucial role in detecting the most suitable candidates for immunomodulation.

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.

Estrogen receptors, ERK1/2 phosphorylation and reactive oxidizing species in red blood cells from patients with rheumatoid arthritis

Red blood cells (RBCs) are recognized to be important pathogenetic determinants in several human cardiovascular diseases (CVD). Undergoing to functional alterations when submitted to risk factors, RBCs modify their own intracellular signaling and the redox balance, shift their status from antioxidant defense to pro-oxidant agents, become a potent atherogenic stimulus playing a key role in the dysregulation of the vascular homeostasis favoring the developing and progression of CVD. Rheumatoid arthritis (RA) is a chronic autoimmune disease associated with a significantly increased risk of cardiovascular mortality with a prevalence from two to five more likely in woman, mainly attributed to accelerated atherosclerosis. The purpose of this study was to correlate the RA disease activity and the RBCs functional characteristics. Thirty-two women (aged more than 18 years) with RA, and 25 age-matched healthy women were included in this study. The disease activity, measured as the number of swollen and painful joints (DAS-28), was correlated with 1) the expression of RBCs estrogen receptors, which modulate the RBC intracellular signaling, 2) the activation of the estrogen-linked kinase ERK_½, which is a key regulator of RBC adhesion and survival, and 3) the levels of inflammatory- and oxidative stress-related biomarkers, such as the acute-phase reactants, the antioxidant capacity of plasma, the reactive oxidizing species formation and 3-nitrotyrosine. All the biomarkers were evaluated in RA patients at baseline and 6 months after treatment with disease-modifying anti-rheumatic drugs (DMARDs). We found, for the first times, that in RA patients 1) the DAS-28 correlated with RBC ER-α expression, and did not correlate with total antioxidant capacity of plasma; 2) the RBC ER-α expression correlated with systemic inflammatory biomarkers and oxidative stress parameters, as well as ERK_½ phosphorylation; and 3) the DMARDs treatments improved the clinical condition measured by DAS-28 score decrease, although the RBCs appeared to be more prone to pro-oxidant status associated to the expression of survival molecules. These findings represent an important advance in the study of RA determinants favoring the developing of CVD, because strongly suggest that RBCs could also participate in the vascular homeostasis through fine modulation of an intracellular signal linked to the ER-α.

miRNA-Mediated Epigenetic Regulation of Treatment Response in RA Patients—A Systematic Review

This study aimed to evaluate the role of microRNAs (miRNA) as biomarkers of treatment response in rheumatoid arthritis (RA) patients through a systematic review of the literature. The MEDLINE and Embase databases were searched for studies including RA-diagnosed patients treated with disease-modifying antirheumatic drugs (DMARDs) that identify miRNAs as response predictors. Review inclusion criteria were met by 10 studies. The main outcome of the study was the response to treatment, defined according to EULAR criteria. A total of 839 RA patients and 67 healthy donors were included in the selected studies. RA patients presented seropositivity for the rheumatoid factor of 74.7% and anti-citrullinated C-peptide antibodies of 63.6%. After revision, 15 miRNAs were described as treatment response biomarkers for methotrexate, anti-tumour necrosis factor (TNF), and rituximab. Among treatments, methotrexate presented the highest number of predictor miRNAs: miR-16, miR-22, miR-132, miR-146a and miR-155. The most polyvalent miRNAs were miR-146a, predicting response to methotrexate and anti-TNF, and miR-125b, which predicts response to infliximab and rituximab. Our data support the role of miRNAs as biomarkers of treatment response in RA and point to DMARDs modifying the miRNAs expression. Nevertheless, further studies are needed since a meta-analysis that allows definitive conclusions is not possible due to the lack of studies in this field.

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.

Identification of SLAMF1 as an immune-related key gene associated with rheumatoid arthritis and verified in mice collagen-induced arthritis model

Background

Rheumatoid arthritis (RA) is the most common inflammatory arthropathy. Immune dysregulation was implicated in the pathogenesis of RA. Thus, the aim of the research was to determine the immune related biomarkers in RA.

Methods

We downloaded the gene expression data of RA in GSE89408 and GSE45291 from Gene Expression Omnibus public database (GEO). Differentially expressed genes (DEGs) were identified between RA and control groups. Infiltrating immune cells related genes were obtained by ssGSEA and weighted gene co-expression network analysis (WGCNA). We performed functional enrichment analysis of differentially expressed immunity-related genes (DEIRGs) by “clusterProfiler” R package, key genes screening by protein-protein interaction (PPI) network of DEIRGs. And mice collagen-induced arthritis (CIA) model was employed to verify these key genes.

Results

A total of 1,885 up-regulated and 1,899 down-regulated DEGs were identified in RA samples. The ssGSEA analysis showed that the infiltration of 25 cells was significantly different. 603 immune related genes were obtained by WGCNA, and 270 DEIRGs were obtained by taking the intersection of DEGs and immune related genes. Enrichment analyses indicated that DEIRGs were associated with immunity related biological processes. 4 candidate biomarkers (CCR7, KLRK1, TIGIT and SLAMF1) were identified from the PPI network of DEIRGs and literature research.

In mice CIA model, the immunohistochemical stain showed SLAMF1 has a significantly high expression in diseased joints. And flow cytometry analysis shows the expression of SLAMF1 on CIA mice-derived CTL cells, Th, NK cells, NKT cells, classical dendritic cell (cDCs) and monocytes/macrophages was also significantly higher than corresponding immune cells from HC mice.

Conclusion

Our study identified SMLAF1 as a key biomarker in the development and progression of RA, which might provide new insight for exploring the pathogenesis of RA.

Machine learning-based prediction of relapse in rheumatoid arthritis patients using data on ultrasound examination and blood test

Recent effective therapies enable most rheumatoid arthritis (RA) patients to achieve remission; however, some patients experience relapse. We aimed to predict relapse in RA patients through machine learning (ML) using data on ultrasound (US) examination and blood test. Overall, 210 patients with RA in remission at baseline were dichotomized into remission (n = 150) and relapse (n = 60) based on the disease activity at 2-year follow-up. Three ML classifiers [Logistic Regression, Random Forest, and extreme gradient boosting (XGBoost)] and data on 73 features (14 US examination data, 54 blood test data, and five data on patient information) at baseline were used for predicting relapse. The best performance was obtained using the XGBoost classifier (area under the receiver operator characteristic curve (AUC) = 0.747), compared with Random Forest and Logistic Regression (AUC = 0.719 and 0.701, respectively). In the XGBoost classifier prediction, ten important features, including wrist/metatarsophalangeal superb microvascular imaging scores, were selected using the recursive feature elimination method. The performance was superior to that predicted by researcher-selected features, which are conventional prognostic markers. These results suggest that ML can provide an accurate prediction of relapse in RA patients, and the use of predictive algorithms may facilitate personalized treatment options.

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.