Abstract
Background The heterogenous presentation and variable clinical response of juvenile psoriatic arthritis (JPsA) to disease-modifying therapies suggests undiscovered subgroups within this disease. Nevertheless, JPsA is often studied under the umbrella of juvenile idiopathic arthritis (JIA), with few studies interrogating JPsA separately. To improve stratified treatment of this rare disease, such subgroups must be uncovered.
Objectives To identify novel, phenotypically consistent subgroups of children and young people (CYP) with JPsA at the point of first contact with paediatric rheumatology.
Methods CYP were initially selected if enrolled between January 2001 and December 2019 to the Childhood Arthritis Prospective Study, a UK, multicentre, prospective inception cohort of JIA. Those who had a physician’s diagnosis of JPsA at any time point through the 10-year follow-up were included, to allow for onset of psoriatic signs after initial diagnosis. At initial presentation to paediatric rheumatology, clinical features within the ILAR classification criteria for JPsA were collected: an active joint count and the presence or absence of psoriasis, dactylitis and nail abnormalities. Latent class analysis used these features to identify clusters of disease. Between one and ten clusters were tested and an optimal model selected based on statistical fit.
Results Of 1,753 CYP with JIA recruited to CAPS within the study period, a total of 161 CYP had ever had a diagnosis of JPsA (n=97 diagnosed as JPsA at initial presentation to paediatric rheumatology). The majority were female (61%), of white ethnicity (94%) and the median age at initial presentation was 10 years (IQR 6, 13).
The optimal latent class model identified two clusters of JPsA. An oligoarticular cluster (90%, median active joint count (IQR): 2 (1,5)) had a higher proportion of CYP affected by psoriasis (Cluster 1: 29%, Cluster 2: 14%). A polyarticular cluster (10%, median active joint count (IQR): 20 (16, 27)) had a higher proportion with nail abnormalities (Cluster 1: 8%, Cluster 2: 27%). There were similar proportions of dactylitis among the clusters (Cluster 1: 18%, Cluster 2: 15%) (Figure 1).
Conclusion This study identifies two clusters of JPsA at initial presentation to paediatric rheumatology with differences in key features used to classify this disease. Such subgroups may have different experiences of disease, and future analysis will explore characteristics, alongside disease impact and response to therapy for these groups.
Objectives To identify novel, phenotypically consistent subgroups of children and young people (CYP) with JPsA at the point of first contact with paediatric rheumatology.
Methods CYP were initially selected if enrolled between January 2001 and December 2019 to the Childhood Arthritis Prospective Study, a UK, multicentre, prospective inception cohort of JIA. Those who had a physician’s diagnosis of JPsA at any time point through the 10-year follow-up were included, to allow for onset of psoriatic signs after initial diagnosis. At initial presentation to paediatric rheumatology, clinical features within the ILAR classification criteria for JPsA were collected: an active joint count and the presence or absence of psoriasis, dactylitis and nail abnormalities. Latent class analysis used these features to identify clusters of disease. Between one and ten clusters were tested and an optimal model selected based on statistical fit.
Results Of 1,753 CYP with JIA recruited to CAPS within the study period, a total of 161 CYP had ever had a diagnosis of JPsA (n=97 diagnosed as JPsA at initial presentation to paediatric rheumatology). The majority were female (61%), of white ethnicity (94%) and the median age at initial presentation was 10 years (IQR 6, 13).
The optimal latent class model identified two clusters of JPsA. An oligoarticular cluster (90%, median active joint count (IQR): 2 (1,5)) had a higher proportion of CYP affected by psoriasis (Cluster 1: 29%, Cluster 2: 14%). A polyarticular cluster (10%, median active joint count (IQR): 20 (16, 27)) had a higher proportion with nail abnormalities (Cluster 1: 8%, Cluster 2: 27%). There were similar proportions of dactylitis among the clusters (Cluster 1: 18%, Cluster 2: 15%) (Figure 1).
Conclusion This study identifies two clusters of JPsA at initial presentation to paediatric rheumatology with differences in key features used to classify this disease. Such subgroups may have different experiences of disease, and future analysis will explore characteristics, alongside disease impact and response to therapy for these groups.
Original language | English |
---|---|
Pages | 661 |
Number of pages | 1 |
Publication status | Published - 2023 |
Event | European Alliance of Associations for Rheumatology - Milan, Italy Duration: 1 Jun 2023 → … |
Conference
Conference | European Alliance of Associations for Rheumatology |
---|---|
Country/Territory | Italy |
Period | 1/06/23 → … |
Keywords
- Epidemiology
- Artificial Intelligence
- Psoriatic arthritis