 |
|
|
 |
Influence of HLA-B27 on the Clinical Presentation of Psoriatic Arthritis To the Editor: We were interested to read the report in The Journal by Dr. Queiro and colleagues regarding the influence of HLA antigens on the clinical presentation of psoriasis and psoriatic arthritis (PsA)1. We have also examined the relationship between HLA-B27 and age of onset and patterns of skin, nail, and joint disease in PsA. We would like to to compare Dr. Queiro's data with our study of UK patients with PsA. We studied 98 patients with PsA according to Moll and Wright2 (50 men, 48 women, mean age 45 ± 12 yrs) recruited from rheumatology outpatient clinics. A careful history was taken about the onset and patterns of psoriasis and arthritis. All patients had a full skin, nail, and musculoskeletal examination. HLA-B27 was detected by polymerase chain reaction sequence-specific primers. Data were analyzed using contingency tables and Student t tests. There were 20 (20.4%) patients with HLA-B27. We did not find a significant association between HLA-B27 and age of onset of skin or joint disease. The mean age of onset of psoriasis in the HLA-B27 positive group was 27 ± 15 years and in the HLA-B27 negative group 23 ± 11 years (p = 0.2). The mean age of onset of arthritis in the HLA-B27 positive group was 34 ± 13 years and in the negative group 32 ± 9 years (p = 0.4). However, we found other associations with respect to psoriasis patterns and HLA-B27. Palmoplantar pustulosis occurred in 3/20 (15%) HLA-B27 patients versus 1/78 (1.3%) in HLA-B27 negative patients (p = 0.006). This finding has been reported in patients with psoriasis without arthritis3, but, to our knowledge, has not been reported in a cohort of patients with PsA. Interestingly, scalp psoriasis occurred less frequently in those with HLA-B27; 14/20 (70%) of those with HLA-B27 had evidence of scalp psoriasis compared with 73/78 (93.5%) of those without HLA-B27 (p = 0.003). There were no other associations found between HLA-B27 and psoriasis type. HLA-B27 was not associated with the presence of nail disease. In summary, in contrast to the study of Spanish patients by Dr. Queiro and colleagues, our study of UK patients with PsA does not show an association between age of disease onset and HLA-B27. We accept that our conclusions may be subject to type II error, but believe that the relationship between HLA-B27 and onset of disease is unlikely to be of clinical significance in our population. However, our data do suggest that the presence of HLA-B27 may modulate other clinical aspects of psoriasis. NICOLA DALBETH, FRACP; JOANNE L. DOCKERTY, MB, ChB; LYN WILLIAMSON, MRCP, Department of Rheumatology, Nuffield Orthopaedic Centre, Oxford, UK. Acknowledgment Funded by an Oxford District Research Grant. Dr. Dalbeth is the Rose Hellaby Rheumatology Fellow 2002. REFERENCES 1. Queiro R, Torre JC, Gonzalez S, Lopez-Larrea C, Tinture T, Lopez-Lagunas I. HLA antigens may influence the age of onset of psoriasis and psoriatic arthritis. J Rheumatol 2003;30:505-7. 2. Moll JHM, Wright V. Psoriatic arthritis. Semin Arthritis Rheum 1973;3:55-78. 3. Zachariae H, Overgaard Peterson H, Kissmeyer Nielson F, Lamm L. HLA antigens in pustular psoriasis. Dermatologica 1977;154:73-7. Dr. Queiro replies To the Editor: On behalf of my colleagues, I deeply appreciate Dr. Dalbeth and colleagues' interest in our work1. The authors studied the relative role of the HLA-B27 antigen on the clinical expression of a substantial number of UK patients with psoriatic arthritis (PsA), and present some interesting conclusions. The first is that in their population HLA-B27 seems to have a marginal role in disease susceptibility, as only 20% of their series showed this marker. It would be of interest to know whether the distribution of this allele differs among the different subgroups of their patients with PsA, as the figure of 20% is similar to that published by us in psoriatic polyarthritis, but clearly inferior to the frequencies noted in our spondylitis or oligoarthritis subgroups2. On the other hand, the authors did not find a correlation between HLA-B27 and age at disease onset, in contrast to our results. It is well established that among psoriasis patients, HLA-Cw6 is associated with early onset of psoriasis and family aggregation (type I psoriasis). Thus, many HLA associations (B17 and DR7, for example) regarding the age at onset of psoriasis or arthritis published in previous years are better explained in the setting of the known phenomenon of linkage disequilibrium between these alleles and HLA-Cw63. For this reason, it has been reported that the strongest disease associations are with combinations of alleles at multiple loci (haplotype) rather than with an individual gene3. In addition, most populations are made up of different subpopulations with different allele frequencies, and such population stratification is a major confounding factor in association studies like the present one. HLA-B27 has been variably associated with a later onset of arthritis, or with an early onset of psoriasis and arthritis as we noted, or on the other hand, not correlated with age at onset of disease. As HLA-B27 is not associated with linkage disequilibrium with HLA-Cw6, the relative contribution of the HLA-B27 to this point will remain elusive. Other notable results of the Dalbeth group's work deserve consideration. They found a relationship between HLA-B27 and palmoplantar pustulosis, a finding also noted by others4. It is necessary to emphasize here in particular that this association is present in patients with anterior chest wall osteitis and sacroiliitis; moreover the strongest correlation of HLA-B27 is with bilateral sacroiliitis: therefore it would be necessary to carefully evaluate the pelvic radiographs of these subjects, since radiographic sacroiliitis may be present in the absence of symptoms5. Finally, the authors found that scalp psoriasis occurred less frequently in HLA-B27 positive than in B27 negative individuals. For some investigators, cervical spine involvement in PsA may occur more frequently in cases in which there is more severe scalp involvement with psoriasis6. In a previous work, we found that peripheral erosions and disease duration were predictive of cervical involvement in psoriatic spondyloarthropathy over time, and additionally, our B27 negative patients showed more peripheral erosions than our B27 positive patients7. Thus, our own data, in conjunction with those of Dr. Dalbeth, et al, add support to the association between peripheral erosions, cervical involvement, HLA-B27 negativity, and scalp psoriasis. We indeed concur with the authors that the presence of HLA-B27 may modulate other aspects of PsA and that this antigen may have a role in clinical expression, apart from its role in disease susceptibility. However, we should keep in mind that the association between a locus and a disease may reflect several situations, including a direct relationship between the marker allele and a phenotype or, alternatively, a linkage disequilibrium between the marker allele and a susceptibility locus. RUBEN QUEIRO, MD, Consultant Rheumatologist, Hospital San Agustín, Avilés-Asturias, Spain. E-mail:ruquei@mixmail.com REFERENCES 1. Queiro R, Torre JC, González S, López-Larrea C, Tinturé T, López-Lagunas I. HLA antigens may influence the age of onset of psoriasis and psoriatic artritis. J Rheumatol 2003;30:505-7. 2. Queiro R, Sarasqueta C, Torre JC, Tinturé T, López-Lagunas I. Spectrum of psoriatic spondyloarthropathy in a cohort of 100 Spanish patients. Ann Rheum Dis 2002;61:857-8. 3. Eastmond CJ. Psoriatic arthritis: genetics and HLA antigens. Baillieres Clin Rheumatol 1994;8:263-76. 4. Khan MF, Chaot AM. SAPHO syndrome. Rheum Dis Clin North Am 1992;18:225-46. 5. Braun J, Sieper J. The sacroiliac joint in the spondyloarthropathies. Curr Opin Rheumatol 1996;8:275-87. 6. Vasey FB, Espinoza LR. Psoriatic arthropathy. In: Calin A, editor. Spondyloarthropathies. New York: Grune and Stratton; 1984:151-85. 7. Queiro R, Sarasqueta C, Torre JC, Tinturé T, López-Lagunas I. Prevalence and predictors of cervical involvement in psoriatic spondyloarthropathy. J Clin Rheumatol 2002;8:23-9.
|