Drivers of non-zero physician global scores during periods of inactive disease in juvenile idiopathic arthritis

Towards stratified treatment of JIA: machine learning identifies subtypes in response to methotrexate from four UK cohorts

BACKGROUND

Methotrexate (MTX) is the gold-standard first-line disease-modifying anti-rheumatic drug for juvenile idiopathic arthritis (JIA), despite only being either effective or tolerated in half of children and young people (CYP). To facilitate stratified treatment of early JIA, novel methods in machine learning were used to i) identify clusters with distinct disease patterns following MTX initiation; ii) predict cluster membership; and iii) compare clusters to existing treatment response measures.

METHODS

Discovery and verification cohorts included CYP who first initiated MTX before January 2018 in one of four UK multicentre prospective cohorts of JIA within the CLUSTER consortium. JADAS components (active joint count, physician (PGA) and parental (PGE) global assessments, ESR) were recorded at MTX start and over the following year. Clusters of MTX 'response' were uncovered using multivariate group-based trajectory modelling separately in discovery and verification cohorts. Clusters were compared descriptively to ACR Pedi 30/90 scores, and multivariate logistic regression models predicted cluster-group assignment.

FINDINGS

The discovery cohorts included 657 CYP and verification cohorts 1241 CYP. Six clusters were identified: Fast improvers (11%), Slow Improvers (16%), Improve-Relapse (7%), Persistent Disease (44%), Persistent PGA (8%) and Persistent PGE (13%), the latter two characterised by improvement in all features except one. Factors associated with clusters included ethnicity, ILAR category, age, PGE, and ESR scores at MTX start, with predictive model area under the curve values of 0.65-0.71. Singular ACR Pedi 30/90 scores at 6 and 12 months could not capture speeds of improvement, relapsing courses or diverging disease patterns.

INTERPRETATION

Six distinct patterns following initiation of MTX have been identified using methods in artificial intelligence. These clusters demonstrate the limitations in traditional yes/no treatment response assessment (e.g., ACRPedi30) and can form the basis of a stratified medicine programme in early JIA.

FUNDING

Medical Research Council, Versus Arthritis, Great Ormond Street Hospital Children's Charity, Olivia's Vision, and the National Institute for Health Research.

Current Validated Clinical and Patient Reported Disease Outcome Measures in Juvenile Idiopathic Arthritis

Juvenile idiopathic arthritis (JIA) is a common chronic illness in childhood and comprises seven categories based on the International League of Associations for Rheumatology (ILAR) criteria. Accurate assessment and measurement of the clinical, functional, and quality of life outcomes of patients with JIA are paramount for understanding the disease course and formulating effective treatment strategies. Interest in the development and use of outcome measurements specifically focused on rheumatologic conditions has greatly expanded over the last two decades, adding to and improving upon the established disease measures. Furthermore, many of these measures have been validated using the widely accepted Outcome Measures in Rheumatology (OMERACT) core principles of instrument validation, allowing researchers and clinicians to gain confidence in these tools. This review summarizes the current validated disease outcome measures in JIA, including clinical, imaging, patient-reported, and functional outcome measurement tools, and highlights ongoing work that continues to refine and improve upon the available tools. The clinical disease outcome measures discussed in this review include physician global assessment (PhGA), American College of Rheumatology (ACR, Wallace) criteria for clinical inactive disease and clinical remission, juvenile arthritis disease activity scores (JADAS), juvenile spondyloarthritis disease activity index (JSPaDA), juvenile arthritis damage index (JADAI), and the ACR pediatric response scores. The imaging outcome measures discussed include the Dijkstra composite scores, childhood arthritis radiographic score of the hip (CARSH), and Poznanski Score. The patient-reported disease outcome measures discussed include patient global assessment (PtGA), patient-reported outcome measurement information system for JIA (PROMIS), juvenile arthritis parent/child centered disease assessment index (JAPAI, JACAI), juvenile arthritis multidimensional assessment report (JAMAR), and the Pediatric quality of life inventory rheumatology module (PedsQL). The functional outcome tools discussed include the Childhood Health Assessment Questionnaire (CHAQ), juvenile arthritis functionality scale and index (JAFS and JASI), and Juvenile Arthritis Functional Assessment Report and Scale (JAFAS and JAFAR).

Paediatric rheumatologists do not score the physician's global assessment of juvenile idiopathic arthritis disease activity in the same way

OBJECTIVES

To assess the heterogeneity in factors affecting physician's global assessment of disease activity (PhGA) and in PhGA scoring of multiple JIA patient's case scenarios.

METHODS

An electronic web-based questionnaire of factors potentially considered in PhGA was sent worldwide to members of PRINTO and the Pediatric Rheumatology Care and Outcomes Improvement Network (PR-COIN). The respondents were asked to rate from 0 to 100 the relevance of 17 factors possibly affecting PhGA scoring and to derive a PhGA score of 17 detailed JIA patient cases. The median and interquartile range was used to measure the heterogeneity in the scoring. To demonstrate the consistency among the PhGA scores of the patient cases provided by multiple physicians, we assessed the inter-rater reliability using intra-class correlation.

RESULTS

The questionnaire was completed by 491 respondents. A large individual variation was observed in the impact of different factors on PhGA when assessing JIA. For non-systemic JIA the presence of fever had the largest variation and swollen joint count had the smallest. For sJIA, the largest variation was seen in the presence of erosions and the smallest in the presence of fever. The intra-class correlation of the group for PhGA scoring of patient cases was 0.53 (95% CI 0.38, 0.72).

CONCLUSIONS

In a sample of worldwide respondents, the scoring of the PhGA is divergent. Consensus on PhGA scoring guidelines is required to obtain a consistent assessment of patients.