Session Information
Session Type: ACR Plenary Session
Session Time: 11:00AM-12:30PM
Background/Purpose: The association between human leukocyte antigen (HLA) class I alleles and psoriatic disease indicates a potential role for the innate immune system in disease pathogenesis. HLA class I educates NK cells through interactions with killer cell Immunoglobulin receptors (KIRs) and by supplying peptides that bind HLA-E to form ligands for the more conserved CD94/NKG2A NK receptors. The peptides corresponding to residues −22 to −14 of the leader sequence of HLA-A, HLA-B, and HLA-C specifically bind to the binding site of HLA-E. In ~80% of HLA-B allotypes, methionine at position −21M (21M) is replaced by threonine. Methionine −21 delivers functional peptides, whereas threonine at this position (−21T) does not. This functional dimorphism divides the human population into three groups: −21M/M, M/T, and T/T, with decreased order of potency of the NK CD94/NKG2A+ receptor. We aimed to determine whether the distribution of the M and T haplotypes differed between patients with PsA, PsC and healthy controls.
Methods: Two sets of cohorts were included in this study: (a) A discovery cohort of 664 PsA patients, 1155 PsC patients and 3118 controls. Class I HLA alleles were imputed using SNP2HLA. Logistic regressions were used to obtain the association p values by PLINK. We performed three association analyses between HLA B -21 amino acid polymorphisms and different phenotypes: 1) PsC vs. controls; 2) PsA vs. controls; 3) PsA vs. PsC. Population stratification and sex were controlled by including the top seven principal components and sex as covariates in logistic regression. (b) A replication cohort of 1177 PsA patients, 659 PsC patients and 1096 controls with self reported European ethnicity from a large well-phenotyped single centre cohort. HLA typing was done by sequence specific oligonucleotide (SSO) probes using the reverse line blot technique with ambiguous results resolved using sequence specific primers (PCR-SSP). Association analyses as with the Discovery cohort were repeated. All analyses were conditioned on HLA-B*27.
Results:
PsC patients within our discovery cohort had a significantly lower prevalence of -21M compared to controls as well as those with PsA (Table 1). The results of the replication study showed similar results (Table 2).
Conclusion: The study provides indications for a potential role of NK cells in PsA pathogenesis, as well as provides a genetic marker that differentiates PsA from PsC.
Table 1. Results of the association study in the discovery cohort. |
|||||
Comparisons |
Freq of -21M in affected |
Freq -21M in controls |
OR (95% CI) |
P value |
P value (adjusted) |
PsC vs. Controls |
0.285 |
0.331 |
0.78 (0.70, 0.87) |
4.262e-05 |
4.411e-06 |
PsA vs. Controls |
0.337 |
0.331 |
1.04 (0.92, 1.18) |
0.671 |
0.546 |
PsA vs. PsC |
0.337 |
0.285 |
1.36 (1.17, 1.58) |
0.001 |
6.692e-05 |
Table 2. Results of the association study in the replication cohort. |
|||||
Comparisons |
Freq of -21M in affected |
Freq -21M in controls |
OR (95% CI) |
P value |
P value (adjusted) |
PsC vs. Control |
0.276 |
0.309 |
0.82 (0.70, 0.96) |
0.042 |
0.015 |
PsA vs. Control |
0.326 |
0.309 |
1.13 (0.99, 1.29) |
0.203 |
0.069 |
PsA vs. PsC |
0.326 |
0.276 |
1.40 (1.20, 1.63) |
0.002 |
2.268e-05 |
To cite this abstract in AMA style:
Chandran V, Li Q, Machhar R, Abji F, Ye JY, Nair R, Stuart P, Oikonomopoulou K, Elder JT, Gladman DD, Rahman P. −21 HLA-Class I Dimorphism Differentiates Psoriatic Arthritis (PsA) from Psoriasis without Psoriatic Arthritis (PsC) [abstract]. Arthritis Rheumatol. 2018; 70 (suppl 9). https://acrabstracts.org/abstract/%e2%88%9221-hla-class-i-dimorphism-differentiates-psoriatic-arthritis-psa-from-psoriasis-without-psoriatic-arthritis-psc/. Accessed .« Back to 2018 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/%e2%88%9221-hla-class-i-dimorphism-differentiates-psoriatic-arthritis-psa-from-psoriasis-without-psoriatic-arthritis-psc/