Baseline predictors of different types of treatment success in rheumatoid arthritis

Development of a Machine Learning Model to Predict the Use of Surgery in Patients With Rheumatoid Arthritis

OBJECTIVE

One in five patients with rheumatoid arthritis (RA) rely on surgery to restore joint function. However, variable response to disease-modifying antirheumatic drugs (DMARDs) complicates surgical planning, and it is difficult to predict which patients may ultimately require surgery. We used machine learning to develop predictive models for the likelihood of undergoing an operation related to RA and which type of operation patients who require surgery undergo.

METHODS

We used electronic health record data to train two extreme gradient boosting machine learning models. The first model predicted patients' probabilities of undergoing surgery ≥5 years after their initial clinic visit. The second model predicted whether patients who underwent surgery would undergo a major joint replacement versus a less intensive procedure. Predictors included demographics, comorbidities, and medication data. The primary outcome was model discrimination, measured by area under the receiver operating characteristic curve (AUC).

RESULTS

We identified 5,481 patients, of whom 278 (5.1%) underwent surgery. There was no significant difference in the frequency of DMARD or steroid prescriptions between patients who did and did not have surgery, though nonsteroidal anti-inflammatory drug prescriptions were more common among patients who did have surgery (P = 0.03). The model predicting use of surgery had an AUC of 0.90 ± 0.02. The model predicting type of surgery had an AUC of 0.58 ± 0.10.

CONCLUSIONS

Predictive models using clinical data have the potential to facilitate identification of patients who may undergo rheumatoid-related surgery, but not what type of procedure they will need. Integrating similar models into practice has the potential to improve surgical planning.

American College of Rheumatology and Food and Drug Administration Summit: Summary of the Meeting May 17-18, 2022

The American College of Rheumatology and the US Food and Drug Administration co-sponsored a public meeting in May 2022 about challenges in the clinical development of drugs for rheumatoid arthritis (RA) and psoriatic arthritis (PsA), focusing on innovative clinical trial designs, outcome measures, and data collection methods. Recommendations include early dose-ranging studies and use of active comparators. Challenges and opportunities in assessing long-term safety by leveraging real-world data from electronic health records (EHRs) and claims data are discussed, along with insights from European registries and the evolving role of real-world evidence and artificial intelligence in regulatory evaluations. Endpoints for assessing disease activity and outcome measures used in RA and PsA trials are explored, emphasizing challenges in defining remission, assessing clinical response, and evaluating structural progression. The need for outcome measures that better reflect treatment targets and the potential of advanced imaging in future trials are highlighted. Challenges with placebo-controlled trials in RA are discussed and use of non-inferiority clinical trial design, in which new drugs are evaluated with active comparators, is proposed. Pragmatic trials in RA and PsA, employing decentralized approaches, are highlighted for their real-world relevance and administrative efficiencies. Strategies for identifying at-risk populations for RA and the challenges of using EHRs and insurance claims data in drug development are discussed. Registry data and digital health technologies show promise in bridging the gap between clinical trials and real-world effectiveness.

Patterns and determinants of response to novel therapies in juvenile and adult-onset polyarthritis

Biologic and targeted synthetic DMARDs (b/tsDMARDs) have revolutionized the management of multiple rheumatic inflammatory conditions. Among these, polyarticular JIA (pJIA) and RA display similarities in terms of disease pathophysiology and response pattern to b/tsDMARDs. Indeed, the therapeutic efficacy of novel targeted drugs is variable among individual patients, in both RA and pJIA. The mechanisms and determinants of this heterogeneous response are diverse and complex, such that the development of true 'precision'-medicine strategies has proven highly challenging. In this review, we will discuss pathophysiological, patient-specific, drug-specific and environmental factors contributing to individual therapeutic response in pJIA in comparison with what is known in RA. Although some biomarkers have been identified that stratify with respect to the likelihood of either therapeutic response or non-response, few have proved useful in clinical practice so far, likely due to the complexity of treatment-response mechanisms. Consequently, we propose a pragmatic, patient-centred and clinically based approach, i.e. personalized instead of biomarker-based precision medicine in JIA.

Disease activity drives transcriptomic heterogeneity in early untreated rheumatoid synovitis

OBJECTIVES

Transcriptomic profiling of synovial tissue from patients with early, untreated rheumatoid arthritis (RA) was used to explore the ability of unbiased, data-driven approaches to define clinically relevant subgroups.

METHODS

RNASeq was performed on 74 samples, with disease activity data collected at inclusion. Principal components analysis (PCA) and unsupervised clustering were used to define patient clusters based on expression of the most variable genes, followed by pathway analysis and inference of relative abundance of immune cell subsets. Histological assessment and multiplex immunofluorescence (for CD45, CD68, CD206) were performed on paraffin sections.

RESULTS

PCA on expression of the (n=894) most variable genes across this series did not divide samples into distinct groups, instead yielding a continuum correlated with baseline disease activity. Two patient clusters (PtC1, n=52; PtC2, n=22) were defined based on expression of these genes. PtC1, with significantly higher disease activity and probability of response to methotrexate therapy, showed upregulation of immune system genes; PtC2 showed upregulation of lipid metabolism genes, described to characterise tissue resident or M2-like macrophages. In keeping with these data, M2-like:M1-like macrophage ratios were inversely correlated with disease activity scores and were associated with lower synovial immune infiltration and the presence of thinner, M2-like macrophage-rich synovial lining layers.

CONCLUSION

In this large series of early, untreated RA, we show that the synovial transcriptome closely mirrors clinical disease activity and correlates with synovial inflammation. Intriguingly, lower inflammation and disease activity are associated with higher ratios of M2:M1 macrophages, particularly striking in the synovial lining layer. This may point to a protective role for tissue resident macrophages in RA.

Rheumatoid arthritis study of the Egyptian College of Rheumatology (ECR): nationwide presentation and worldwide stance

To depict the spectrum of rheumatoid arthritis (RA) in Egypt in relation to other universal studies to provide broad-based characteristics to this particular population. This work included 10,364 adult RA patients from 26 specialized Egyptian rheumatology centers representing 22 major cities all over the country. The demographic and clinical features as well as therapeutic data were assessed. The mean age of the patients was 44.8 ± 11.7 years, disease duration 6.4 ± 6 years, and age at onset 38.4 ± 11.6 years; 209 (2%) were juvenile-onset. They were 8750 females and 1614 males (F:M 5.4:1). 8% were diabetic and 11.5% hypertensive. Their disease activity score (DAS28) was 4.4 ± 1.4 and health assessment questionnaire (HAQ) 0.95 ± 0.64. The rheumatoid factor (RF) and anti-cyclic citrullinated peptide (anti-CCP) were positive in 73.7% and 66.7% respectively. Methotrexate was the most used treatment (78%) followed by hydroxychloroquine (73.7%) and steroids (71.3%). Biologic therapy was received by 11.6% with a significantly higher frequency by males vs females (15.7% vs 10.9%, p = 0.001). The least age at onset, F:M, RF and anti-CCP positivity were present in Upper Egypt (p < 0.0001), while the highest DAS28 was reported in Canal cities and Sinai (p < 0.0001). The HAQ was significantly increased in Upper Egypt with the least disability in Canal cities and Sinai (p = 0.001). Biologic therapy intake was higher in Lower Egypt followed by the Capital (p < 0.0001). The spectrum of RA phenotype in Egypt is variable across the country with an increasing shift in the F:M ratio. The age at onset was lower than in other countries.

Failure and multiple failure for disease modifying antirheumatic drugs in rheumatoid arthritis: Real-life evidence from a tertiary referral center in Italy

BACKGROUND

Rheumatoid Arthritis (RA) is a chronic inflammatory disease with a heterogeneous treatments' clinical response. Goals of treatment are remission and low disease activity, which are not achieved in all patients despite the introduction of early treatment and the treat to target strategy.

OBJECTIVE

To investigate the causes of disease-modifying antirheumatic drugs (DMARDs) discontinuation and treatment failure and multiple failure for inefficacy, and to identify possible failure predictors' according to RA patient characteristics in a real-world setting.

METHODS

718 RA patients were retrospectively evaluated. Conventional synthetic (cs) and biologic (b)DMARDs treatments line/s, effectiveness, and reasons of discontinuations were evaluated. Patients failing to at least two csDMARDs or bDMARDs' drug for inefficacy were defined "csDMARDs multifailure" and "bDMARDs multifailure", respectively. Discontinuation of at least two cs- and bDMARDs was termed "global multifailure".

RESULTS

In total, 1422 csDMARDs and 714 bDMARDs treatment were analysed. Causes of csDMARDs discontinuation were intolerance (21.8%), inefficacy (20.2%), acute adverse reactions (5.3%) and severe infections (0.6%) while csDMARDs multifailure for inefficacy was observed in 5.7% of cases. Reasons of bDMARDs withdrawal were inefficacy (29%), intolerance (10.0%), acute adverse reaction (6.3%) and severe infections (1.5%). Altogether, 8.4% of patients were bDMARDs multifailure for inefficacy while 16.6% were global multifailure. Longstanding disease (≥ 12 months) and smoke habit, resulted as positive predictor of csDMARDs failure (OR 2.6 and OR 2.7, respectively). Thyreopathy was associated with both csDMARDs failure and global multifailure (OR 2.4 and OR 1.8, respectively). Higher prevalence of failure to at least one bDMARDs and global multifailure was detected in female than male (OR 2.3 and OR 2, respectively).

CONCLUSIONS

Different causes of drug discontinuation were observed on DMARDs treatments. Demographic and clinical features were identified as possible predictors of both cs- and bDMARDs treatment failure and multiple failure, underlining the need of a more personalized therapeutic approach to achieve treatment targets.

Clinical predictors of response to methotrexate in patients with rheumatoid arthritis: a machine learning approach using clinical trial data

BACKGROUND

Methotrexate is the preferred initial disease-modifying antirheumatic drug (DMARD) for rheumatoid arthritis (RA). However, clinically useful tools for individualized prediction of response to methotrexate treatment in patients with RA are lacking. We aimed to identify clinical predictors of response to methotrexate in patients with rheumatoid arthritis (RA) using machine learning methods.

METHODS

Randomized clinical trials (RCT) of patients with RA who were DMARD-naïve and randomized to placebo plus methotrexate were identified and accessed through the Clinical Study Data Request Consortium and Vivli Center for Global Clinical Research Data. Studies with available Disease Activity Score with 28-joint count and erythrocyte sedimentation rate (DAS28-ESR) at baseline and 12 and 24 weeks were included. Latent class modeling of methotrexate response was performed. The least absolute shrinkage and selection operator (LASSO) and random forests methods were used to identify predictors of response.

RESULTS

A total of 775 patients from 4 RCTs were included (mean age 50 years, 80% female). Two distinct classes of patients were identified based on DAS28-ESR change over 24 weeks: "good responders" and "poor responders." Baseline DAS28-ESR, anti-citrullinated protein antibody (ACPA), and Health Assessment Questionnaire (HAQ) score were the top predictors of good response using LASSO (area under the curve [AUC] 0.79) and random forests (AUC 0.68) in the external validation set. DAS28-ESR ≤ 7.4, ACPA positive, and HAQ ≤ 2 provided the highest likelihood of response. Among patients with 12-week DAS28-ESR > 3.2, ≥ 1 point improvement in DAS28-ESR baseline-to-12-week was predictive of achieving DAS28-ESR ≤ 3.2 at 24 weeks.

CONCLUSIONS

We have developed and externally validated a prediction model for response to methotrexate within 24 weeks in DMARD-naïve patients with RA, providing variably weighted clinical features and defined cutoffs for clinical decision-making.

What Factors Influence Treatment Effectiveness in Rheumatoid Arthritis: An Evidence-Based Approach to Multidimensional Measurement of Treatment Effectiveness

Purpose: The aim of the study was to examine the effects of socio-demographic characteristics, disease-related characteristics and health care use related-characteristics on the treatment effectiveness of rheumatoid arthritis patients, both separately and together. Methods: The sample of the study consisted of 440 rheumatoid arthritis patients for 99% confidence level, and this sample was reached based on the convenience sampling method. Patients who received at least one anti-TNF therapy were included in the study. Treatment effectiveness levels of rheumatoid arthritis patients were measured with a questionnaire. In the analysis of the study, four different regression models were established. In the first model, socio-demographic characteristics; in the second model, disease characteristics; in the third model, health care use characteristics: in the fourth model, the effect of all these variables on treatment effectiveness was examined. Results: In the study, smoking status, age (socio-demographic characteristics), drug regimen complexity, comorbidity status, education about the disease, disease duration (disease characteristics), and a number of admissions (health care use characteristics), were found to have a significant effect on treatment effectiveness. Conclusion: In the study, the factors affecting the treatment effectiveness were determined. Therefore, it is important to consider these factors revealed in this study in order to increase the treatment effectiveness in patients with rheumatoid arthritis.