Genetic association with articular damage in patients with juvenile idiopathic arthritis (JIA)

Introduction

Bone loss in inflammatory arthritis such as rheumatoid arthritis is partly due to aberrant expression of cytokines and bone homeostasis regulatory molecules, leading to excess bone resorption.

Objectives

To investigate if genetic factors affect the degree of cartilage and bone loss in JIA, irrespective of disease duration and treatment.

Methods

DNA was extracted from saliva samples from 80 JIA patients from Great Ormond Street Hospital, UK, 98 from Hôpital Necker, France, and 54 from Istituto Giannina Gaslini, Italy. Genetic variation was investigated using the tagging single nucleotide polymorphisms (tSNPs) approach. 17 candidate genes were selected for analysis: RANK, RANKL, osteoprotegerin (OPG), osteopontin, DKK-1, IL-1α, IL-1β, IL-1R1, IL-1R2, IL-1RN, IL-1RAP, IL-6, TGFβ1, IL-10, IL-19, IL-20, IL-24. 391 tSNPs were genotyped using the llumina GoldenGate assay genotyping platform and KBioscience, UK. JADI-A scores, wrist MRI bone erosion scores, and X-ray Poznanski scores were taken at presentation (baseline) and after 1 year followup.

Results

At baseline patients were divided into those with no damage (JADI or MRI score of 0) or with damage (any score>0). At 1 year follow-up patients were divided into those who had improved, unchanged, or worsened. Significant tSNPs from genetic association analysis using PLINK are presented below. Table 1.

Table 1

We observed weak correlations between JADI-A and MRI scores (Spearman’s r = 0.298, p<0.0001), JADI-A and Poznanski scores (Spearman’s r = -0.288, p=0.012), and Poznanski and MRI scores (Spearman’s r = -0.381, p=0.001). Disease duration, activity, and treatment were varied and were not significantly associated with the damage parameters in this cohort.

Conclusion

Our findings suggest that polymorphisms in cytokine and bone remodelling genes such as RANKL or OPG may be associated with the degree of articular damage in JIA. Given the weak correlations we found between JADI-A and MRI scores, it is not surprising that different tSNPs were found to associate with MRI damage than with JADI-A damage. Further studies including a larger cohort of patients are needed to validate these findings.

Disclosure of interest

None declared.