Background/Purpose: We have previously reported the association between HLA and KIR gene variants with the development of Arthritis Mutilans (AM) in psoriatic arthritis (PsA). However, severe joint damage in PsA may be further classified into disorganization, subluxation, pencil-in-cup change and ankylosis. We therefore sought to determine whether previously identified HLA and KIR variants associated with AM also associate with sub-phenotypes of AM.

Methods: Data on the presence of severe joint were obtained form a large cohort. Radiographs of the hands, feet and spine are obtained at baseline and 2-yearly intervals and scored according to the modified Steinbrocker method by at least 2 rheumatologists by consensus. Grade 4 is considered severe joint destruction Radiographs previously rated as 4 were reviewed and reclassified as disorganization (4.0), subluxation (4.1), pencil-in-cup change (4.2) and ankylosis (4.3). Data on the time when severe joint destruction was first observed was obtained on 555 Caucasian subjects with PsA satisfying CASPAR criteria who were seen in the clinic within 5 years of diagnosis. HLA typing was performed by PCR-SSO, and KIR typing by PCR-SSP. The principal outcome measure was the time to development of severe joint destruction after diagnosis of PsA. Parametric survival analyses with interval censoring using a Weibull model adjusted for age at diagnosis of PsA and sex were conducted. Gene variants previously reported to be associated with arthritis mutilans- HLA-A*11, -A*29, -B*27, -C*02, -C*03, -C*04, -DQB1*02 and KIR3DS1 were predictor variables.

Results: The 555 subjects 312 (56.2%) males, mean age at diagnosis of 40 (13.5) years, mean age at first visit 42 (13.3) years, mean duration of PsA 2.0 (1.7) years, mean tender joint count 7.9 (7.9), mean swollen joint count 4.9 (5.4)] had a median number of 3.4 (3.1) radiographic assessments during a median follow up of 6.9 (8.4) years. 102 (19%) subjects were observed to develop severe joint damage. No association between 4.0 and gene variants could be demonstrated. Univariate analyses adjusted for age and sex showed that HLA- C*02 (p=0.06) and -B*27 (p=0.004) were associated with an increased hazard of developing 4.1. Multivariate analysis with alleles significant at p<0.1 showed that only HLA-B*27 (HR 2.76, 95% CI (1.39, 5.49), p=0.004) was associated with increased hazard of developing 4.1. When analyzing 4.2, no association with HLA alleles and only a trend for association with KIR3DS1 (HR 1.72, p=0.08) was demonstrated. Association between 4.3 and HLA-DQB1*02 (p=0.03) was found on univariate analysis adjusted for age and sex. Multivariate analysis revealed that HLA- DQB1*02 (HR 1.86, 95% CI (1.09, 3.18)), p=0.02) was associated with increased hazard of developing 4.3, and there was a trend for association with KIR3DS1 (HR = 1.61, 95% CI (0.98, 2.64), p=0.06).

Conclusion: Differential HLA association with sub-phenotypes of severe damage was demonstrated. HLA-B*27 was associated with increased risk of subluxation, whereas – DQB1*02 was associated with increased risk of ankylosis.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/differential-association-between-human-leukocyte-antigen-hla-alleles-and-joint-subluxation-and-ankylosis-in-patients-with-psoriatic-arthritis/