JIA Patients Show Differences That May Affect Future Disease Activity, Study Says

JIA Patients Show Differences That May Affect Future Disease Activity, Study Says
0
(0)

An analysis using machine learning identified distinct patient groups and disease activity trajectories among young patients in the U.K. with juvenile idiopathic arthritis (JIA).

Its findings revealed that nearly a quarter of patients report persistently poor well-being despite having a low number of active joints or disease activity.

Understanding the characteristics of patient subgroups, defined by their manifestations or disease progression, could improve care in JIA by tailoring healthcare services and treatment plans, the researchers said. 

The study, “Patient-reported wellbeing and clinical disease measures over time captured by multivariate trajectories of disease activity in individuals with juvenile idiopathic arthritis in the UK: a multicentre prospective longitudinal study,” appeared in the journal The Lancet Rheumatology.

JIA symptoms and progression vary widely from patient to patient. Tools such as the clinical Juvenile Arthritis Disease Activity Score (cJADAS) are used to assess disease activity and assign composite scores, with high scores indicating more severe activity.

Due to disease variability, some groups of JIA patients may experience divergent patterns of disease activity over time. However, clinical measures of disease activity do not always reflect those experiences. In fact, JIA is classified by clinicians as clinically inactive in nearly one-quarter of patients under the age of 16 despite ongoing symptoms.

Characterizing disease activity patterns and outcomes may aid in developing personalized treatments leading to better outcomes. Researchers at the University of Manchester led a study that used unsupervised machine learning approaches to identify clusters of JIA patients with similar disease patterns.

At disease onset and over three years, groups of JIA patients were identified using models that spot out those following similar trajectories based on disease activity measures. The cJADAS10 components included in the analysis were an active joint count up to 10, physician’s global assessment (PGA), and patient or parent global evaluation (PGE) of well-being.

The study included children under age 16 enrolled in the U.K. Childhood Arthritis Prospective Study (CAPS), recruited from 2001 to 2015 to allow for at least three years of follow-up. Data were collected yearly, with exams given every six months from 2001 to 2010.

Of the 1,184 participants, 773 (65%) were female and 411 (35%) were male, with a median age of 7.4 at study start. Disease activity data were available for 1,140 (96%) children at study start and 949 (80%) at the three-year follow-up. Of 1,179 patients with JIA subtype classifications, 594 (50%) were diagnosed with oligoarthritis.

The model identified five unique patient clusters at the initial presentation. The “all-low” cluster of 554 patients (47% of all participants) had lower scores in all disease activity components, the “all-high” cluster of 182 patients (15%) had higher scores in all components.

The remaining three groups had low-to-moderate scores on two components, and a high score on the others. Specifically, the high active joint count cluster had 189 patients (16%), the high PGE cluster had 161 (14%), and the high PGA cluster had 98 (8%).

The trajectory model classified six groups based on follow-up disease activity data. In four of the six groups, all three disease activity components followed the same trajectory in how they changed from study start to year three. 

In the remaining two groups, individual disease activity components followed different trajectories. The low-persistent group of 62 patients (14% of participants) had low joint counts and PGA scores at initial presentation, while the high-persistent group of 96 patients (8%) had high joint counts and PGA scores at the onset. Significantly, in patients from both groups (a total of 22%), joint counts and PGA improved over time, but the poor well-being scores at onset remained stable throughout follow-up.

Compared to those in the other four groups, patients in these two groups were older at first presentation and had a higher social deprivation score, which quantifies socioeconomic disadvantages. Of the 594 oligoarthritis patients, 74% were either in disease remission or with low disease activity at three years. In turn, of the 85 low-persistent patients, 26% had enthesitis-related arthritis and 22% had psoriatic arthritis, while 27% of high-persistent patients had rheumatoid factor-positive polyarthritis. 

The 554 patients in the all-low cluster at initial presentation were most frequently in the “low” trajectories, with 50% reaching disease remission and 32% low disease activity. The high active joint count cluster (189 participants) was most frequent in the high trajectories, while the high PGE cluster was primarily in the low-persistent and the low-to-disease remission trajectory groups.

“These data show that although disease severity at onset of JIA is important across several JIA core outcomes, it does not entirely predict future disease course, particularly for those who have persistently poor wellbeing despite improvement in clinically detected arthritis,” the scientists wrote.

“Our study also shows that novel unsupervised machine learning methods applied to traditional epidemiological data represent an exciting step toward stratified management for children and young people with JIA,” they added.

Aisha Abdullah received a B.S. in biology from the University of Houston and a Ph.D. in neuroscience from Weill Cornell Medical College, where she studied the role of microRNA in embryonic and early postnatal brain development. Since finishing graduate school, she has worked as a science communicator making science accessible to broad audiences.
Total Posts: 11

José holds a PhD in Neuroscience from Universidade of Porto, in Portugal. He has also studied Biochemistry at Universidade do Porto and was a postdoctoral associate at Weill Cornell Medicine, in New York, and at The University of Western Ontario in London, Ontario, Canada. His work has ranged from the association of central cardiovascular and pain control to the neurobiological basis of hypertension, and the molecular pathways driving Alzheimer’s disease.

×
Aisha Abdullah received a B.S. in biology from the University of Houston and a Ph.D. in neuroscience from Weill Cornell Medical College, where she studied the role of microRNA in embryonic and early postnatal brain development. Since finishing graduate school, she has worked as a science communicator making science accessible to broad audiences.
Latest Posts
  • machine learning and JIA
  • Genome analysis, JIA subtype
  • Epstein-Barr virus
  • gene variants, methotrexate (MTX)

How useful was this post?

Click on a star to rate it!

Average rating 0 / 5. Vote count: 0

No votes so far! Be the first to rate this post.

As you found this post useful...

Follow us on social media!

We are sorry that this post was not useful for you!

Let us improve this post!

Tell us how we can improve this post?