Drs. Martorana and Vaglio contributed equally to this work.
Chronic periaortitis and HLA–DRB1*03: Another clue to an autoimmune origin
Article first published online: 6 FEB 2006
Copyright © 2006 by the American College of Rheumatology
Arthritis Care & Research
Volume 55, Issue 1, pages 126–130, 15 February 2006
How to Cite
Martorana, D., Vaglio, A., Greco, P., Zanetti, A., Moroni, G., Salvarani, C., Savi, M., Buzio, C. and Neri, T. M. (2006), Chronic periaortitis and HLA–DRB1*03: Another clue to an autoimmune origin. Arthritis & Rheumatism, 55: 126–130. doi: 10.1002/art.21698
- Issue published online: 6 FEB 2006
- Article first published online: 6 FEB 2006
- Manuscript Accepted: 1 AUG 2005
- Manuscript Received: 8 APR 2005
- Associazione Emma ed Ernesto Rulfo per la Genetica Medica
- Italian Ministry of Education, University and Research
- Chronic periaortitis;
- Retroperitoneal fibrosis;
- Autoimmune disease
Patients with chronic periaortitis (CP) often show clinical and laboratory findings of a systemic autoimmune disorder. The aim of the present study was to investigate the role of the HLA system in CP.
Low-resolution genotyping for HLA–A, HLA–B, and HLA–DRB1 loci and genotyping of TNFA(-238)A/G and TNFA(-308)A/G single nucleotide polymorphisms were performed in 35 consecutive patients with CP and 350 healthy controls.
The HLA–DRB1*03 allele frequency was strikingly higher in patients with CP than in controls (24.28% versus 9.14%; χ2 = 15.50, P = 0.000084, corrected P [Pcorr] = 0.0012, odds ratio [OR] 3.187, 95% confidence interval [95% CI] 1.74–5.83); the HLA–B*08 allele frequency was also higher in patients than in controls (17.14% versus 6.28%; χ2=11.12, P = 0.0008, Pcorr = 0.0269, OR 3.085, 95% CI 1.54–6.16). The A*01 allele frequency was significantly different (P = 0.0463), but the statistical significance was lost after correction for multiple testing (Pcorr = 0.5088). TNFA(-238)A allele and TNFA(-308)A allele frequencies were not significantly different (P = 0.512 and P = 0.445, respectively). Comparison of the main clinical and laboratory findings suggestive of a systemic autoimmune disease (e.g., acute-phase reactants, constitutional symptoms, other autoimmune diseases associated with CP) between the HLA–DRB1*03–positive and the HLA–DRB1*03–negative patients showed that the former group had significantly higher levels of C-reactive protein (P = 0.045) at disease onset, although this difference was not statistically significant after correction for multiple tests (Pcorr = 0.369).
The HLA system plays a role in susceptibility to CP. The strong association between CP and HLA–DRB1*03, an allele linked to a wide range of autoimmune conditions, further supports the view that CP may represent a clinical manifestation of an autoimmune disease.
Chronic periaortitis (CP) is a rare clinicopathologic entity characterized by the presence of a fibroinflammatory mass surrounding the abdominal aorta and the iliac arteries, which often extends into the retroperitoneal space to entrap the neighboring structures (e.g., ureters, inferior vena cava) (1). The spectrum of CP includes nonaneurysmatic (idiopathic retroperitoneal fibrosis [IRF]) and aneurysmatic (inflammatory abdominal aortic aneurysms [IAAAs] and perianeurysmal retroperitoneal fibrosis [PRF]) forms, which have common laboratory, histologic, and clinical findings (1, 2).
The pathogenesis of CP is still obscure. The leading theory suggests that it may be the consequence of a local inflammatory reaction to antigens in the aortic atherosclerotic plaques (2–4). In contrast, in some cases, findings such as the presence of constitutional symptoms, the high acute-phase reactant levels, the positivity for autoantibodies, and the association with other autoimmune diseases support the hypothesis that CP may be the manifestation of a systemic autoimmune disorder (5).
The genetic component of CP was thought to likely be small; no evidence of ethnic predisposition or familial clustering exists, and there are only anecdotal case reports of IRF in twins or siblings (6, 7). However, the genetic background of patients with CP has never been extensively investigated.
It is well known that genes of the HLA region contribute to disease susceptibility in several immunopathologic disorders, and in particular some HLA class II alleles have been found to be associated with a number of autoimmune conditions (8). In addition, the polymorphisms of genes belonging to HLA class III such as TNFA have also been shown to increase tumor necrosis factor α (TNFα) production and to increase susceptibility to autoimmune diseases (9).
In the present study, a cohort of 35 patients with CP and a group of ethnically matched healthy controls were genotyped for HLA–A, HLA–B, and HLA–DRB1 loci, as well as TNFA(-238)A/G and TNFA(-308)A/G single nucleotide polymorphisms to explore the role of the HLA system in CP. Furthermore, we investigated the potential relationship between the patients' main clinical and laboratory data and the HLA findings.
PATIENTS AND METHODS
Patients and controls.
Informed oral consent was obtained from all the participants, and from parents for minors as appropriate. The protocol was approved by the local ethics committee of the University of Parma.
All the patients and the controls were Italian, white individuals; persons from Sardinia and from the other Italian genetic isolates were not included. The frequencies of the alleles were similar to other Italian control populations previously examined (10).
We enrolled 35 consecutive patients with CP (20 males and 15 females, median age 60 years, range 9–75 years) between March 1997 and May 2004. CP was diagnosed by means of computed tomography or magnetic resonance imaging. The diagnosis was histologically confirmed in 20 patients undergoing surgery (12 for ureterolysis, 6 for exploratory purposes, and 2 for repair of IAAAs). None of the patients had active infections or malignancies, had previously undergone major abdominal surgery, or were taking drugs known to cause CP.
Each patient also underwent physical examination, chest radiography, thyroid echotomography, and assessment of laboratory tests, which included routine parameters, erythrocyte sedimentation rate (ESR), C-reactive protein (CRP) level, and autoantibodies, as previously described (5). The control group consisted of 350 healthy individuals (176 men and 174 women, median age 32 years, range 20–55 years) without history of autoimmune/inflammatory diseases.
Genomic DNA was extracted from 10 ml EDTA peripheral blood samples using the QIAamp DNA Blood Mini Kit (Qiagen, Valencia, CA) and stored at −20°C until use. Low-resolution genotyping for HLA–A, HLA–B, and HLA–DRB1 loci was performed by sequence-specific primer–polymerase chain reaction (PCR; One Lambda, Canoga Park, CA). Genotyping of TNFA(-238)A/G and TNFA(-308)A/G single nucleotide polymorphisms was performed by sequence-specific priming using the Protrans Cyclerplate System Cytokines 2 (Protrans, Ketsch, Germany) according to the manufacturer's protocol. The PCR products were resolved by electrophoresis on 1.5% weight/volume agarose gel stained with ethidium bromide to determine their genotype.
Quality measures were adopted according to the consensus conference of the Centers for Disease Control and Prevention and the National Institutes of Health (11, 12). Each patient and each healthy control was assigned a code number. The samples were handled blindly in order of the code numbers by 2 investigators (DM and AZ) who did not know their corresponding genetic and clinical status. The degree of reproducibility between quality control replicates was 100%. The laboratory where HLA genotyping was performed is certified by the European Foundation for Immunogenetics.
The allele and haplotype frequencies of patients and controls were compared using the chi-square method or Fisher's exact test. The analyses were performed using SISA software (13). P values were corrected according to the number of specificities tested and the number of comparisons performed using Sidak's test for multiple comparisons (Pcorr = 1 − [1 − P]k, where k is the number of comparisons).
The power of the test was calculated with the Power for Association With Errors program (14, 15), assuming the allele to test as the first allele and the other alleles as the second allele. The patient and control groups were in Hardy-Weinberg equilibrium for all the alleles analyzed. The genotyping error rate was set at 0.01. The significance level of the test was set at 0.05.
Differences in the main clinical and laboratory characteristics of the HLA–DRB1*03–positive and HLA–DRB1*03–negative patients with CP were analyzed with Mann-Whitney U test, and differences in categorical variables were analyzed by Fisher's exact test. If the P value was significant, Sidak's test for multiple comparisons was then applied.
According to Woolf's method (16), 95% confidence intervals (95% CIs) were calculated for odds ratio (OR). A P value less than 0.05 was considered to be statistically significant.
Clinical manifestations and laboratory findings.
Of the 35 enrolled patients, 26 had IRF, 6 had PRF, and 3 had IAAAs. At the time of disease onset, localized pain (e.g., back, abdominal, lumbar) was found in 30 patients (86%) and constitutional symptoms (e.g., fatigue, anorexia, weight loss) in 27 patients (77%). Twenty-nine (83%) showed ureteral involvement, and 16 (55%) of these 29 developed acute renal failure. Other presenting clinical manifestations included varicocele and/or hydrocele (9 patients), lower limb edema (4 patients), and deep venous thrombosis (3 patients).
The median ESR was 56 mm/hour (range 10–124 mm/hour, normal value 1–15 mm/hour) and the median CRP level was 23 mg/liter (range 4–182 mg/liter, normal value <5 mg/liter). Antinuclear antibodies were found in 14 patients (40%), antithyroglobulin and/or antithyroid microsome antibodies in 11 patients (31%), antismooth muscle antibodies in 5 patients (14%), rheumatoid factor in 5 patients (14%), and antineutrophil cytoplasmic antibodies in 3 patients (2 with cytoplasmic and 1 with perinuclear pattern).
Seventeen (49%) patients also had autoimmune/inflammatory diseases involving other organs/structures: autoimmune thyroiditis, diagnosed on the basis of antithyroid antibody positivity and echotomographic findings, was found in 11 (31%) patients, 2 of whom developed hypothyroidism. Two patients developed seropositive rheumatoid arthritis, 2 seronegative undifferentiated arthritis, 1 membranous glomerulonephritis, and 1 had experienced autoimmune thrombocytopenia prior to the onset of CP.
The HLA alleles frequencies in patients and controls are listed in Table 1. The HLA–DRB1*03 allele was present in 17 (48.57%) of 35 patients with CP and in 58 (16.57%) of 350 controls, whereas the allele frequency was 24.28% (17 of 70) in patients and 9.14% (64 of 700) in controls (χ2 = 15.50, P = 0.000084, Pcorr = 0.0012, OR 3.187, 95% CI 1.74–5.83). The HLA–B*08 allele frequency was 17.14% in patients and 6.28% in controls (χ2 = 11.12, P = 0.0008, Pcorr = 0.0269, OR 3.085, 95% CI 1.54–6.16). HLA–A*01 allele was also significantly different (P = 0.0463), but this value became nonsignificant after Sidak's test correction for multiple tests (Pcorr = 0.5088); TNFA(-238)A allele and TNFA(-308)A allele frequencies were not significantly different (P = 0.512 and P = 0.445, respectively). The power value for allelic tests for HLA–DRB1*03 allele was 0.976 without errors and 0.963 assuming 1% of error in allele typing; for HLA–B*08 and HLA–A*01 alleles the power values were 0.912 and 0.505 without errors and 0.87 and 0.473 with errors, respectively.
|A*01||15/70 (21.43)||90/700 (12.85)||3.97||0.0463||0.5088||1.848||1.00–3.41|
|B*08||12/70 (17.14)||44/700 (6.28)||11.12||0.0008||0.0269†||3.085||1.54–6.16|
|DRB1*03||17/70 (24.28)||64/700 (9.14)||15.50||0.000084||0.0012†||3.187||1.74–5.83|
|TNFA(-238)A||7/70 (10)||89/700 (12.71)||0.43||0.512||0.762||0.763||0.34–1.72|
|TNFA(-308)A||14/70 (20)||115/700 (16.43)||0.58||0.445||0.692||1.272||0.68–2.36|
Comparison of HLA–DRB1*03–positive and HLA–DRB1*03–negative patients
The most relevant clinical and laboratory findings of the HLA–DRB1*03–positive and HLA–DRB1*03–negative patients are shown in Table 2. The comparison of these characteristics showed that the CRP levels at the time of disease onset were significantly higher in the HLA–DRB1*03–positive patients than in the HLA–DRB1*03–negative patients in uncorrected tests, although the statistical significance was lost after Sidak's correction test for multiple comparisons. Other findings such as the presence of constitutional symptoms, the association with autoimmune thyroiditis, the abnormal ESR, and the positivity for autoantibodies also occurred more frequently in the HLA–DRB1*03–positive group, but the differences were not statistically significant.
|Patient characteristics||HLA–DRB1*03–positive (17 patients)||HLA–DRB1*03–negative (18 patients)||P|
|Age at disease onset, median (range) years||60 (9–75)||55 (37–72)||0.21|
|Constitutional symptoms, no. (%)†||15 (88)||12 (67)||0.23|
|Ureteral involvement, no. (%)||15 (88)||14 (78)||0.66|
|ESR at disease onset, median (range) mm/hour||59 (38–124)||50 (10–114)||0.20|
|CRP level at disease onset, median (range) mg/liter‡||41 (15–182)||15 (4–105)||0.045§|
|Autoimmune thyroiditis, no. (%)||7 (41)||4 (22)||0.29|
|Other autoimmune/inflammatory diseases||1 rheumatoid arthritis, 1 undifferentiated arthritis, 1 autoimmune thrombocytopenia||1 rheumatoid arthritis, 1 undifferentiated arthritis, 1 membranous glomerulonephritis|
|Positive antinuclear antibodies, no. (%)||8 (47)||6 (33)||0.50|
|Patients positive for other autoantibodies¶||6 (35)||5 (28)||0.72|
CP encompasses a spectrum of diseases of unknown etiology whose common histopathologic characteristic is the presence of a chronic inflammatory tissue, which usually spreads from the adventitia of the abdominal aorta and the iliac arteries to the retroperitoneum, where it most likely elicits an exuberant fibrogenic reaction (Figure 1) (1, 2). In some cases the disease may also involve other vascular territories, such as the thoracic aorta and its main branches (17). Although a local inflammatory response to aortic atherosclerotic plaque antigens such as oxidized low-density lipoproteins and ceroid has been claimed as the main pathogenetic mechanism, the association between CP and systemic autoimmune/inflammatory disorders has been acknowledged since the earliest reports of the disease (18). In addition, CP frequently presents with constitutional symptoms, and in the early stages its laboratory findings reflect a persistent acute-phase reaction (5); treatment with corticosteroids, alone or in combination with other immunosuppressive agents, usually achieves prompt and durable clinical responses in patients with CP. These observations are likely to suggest that CP may result from a systemic immune-mediated disorder rather than from a local exaggerated reaction to atherosclerosis.
The hypothesis that immuno-genetic factors play a role in the pathogenesis of CP has been considered by several researchers, but most of these studies involved single cases or small case series (19–21). In a number of case reports, IRF was associated with HLA–B27–positive ankylosing spondylitis, thus HLA–B27 has been proposed as a potential genetic marker of the disease (19, 20); so far, however, this has not been confirmed. Another report found no association between IRF and HLA antigens, but only 8 patients were studied (21). A larger study, performed on IAAAs, showed that the HLA–DRB1*15 and HLA–DRB1*0404 alleles were more frequent in patients with IAAAs compared with healthy controls (22).
In the present study, we evaluated the frequencies of HLA–A, HLA–B, and HLA–DRB1 alleles in patients with CP and in controls. In addition to HLA class I and class II typing, we have also analyzed the class III TNFA(-238)A/G and TNFA(-308)A/G single nucleotide polymorphisms in the regulatory sequence of the TNFα promoter.
Our results show that the HLA system plays a role in the susceptibility to CP; moreover, the observation that CP is significantly associated with HLA–DRB1*03 allele may have potential pathogenetic implications. It is well known that HLA–DRB1*03 allele is a genetic marker of autoimmunity: a large number of autoimmune diseases are associated with HLA–DRB1*03 allele, including systemic lupus erythematosus, myasthenia gravis, type 1 diabetes mellitus, and autoimmune thyroid disease (23). HLA–DRB1*03 allele is in linkage disequilibrium with HLA–A*01, HLA–B*08, and TNFA(-308)A alleles, which could explain the decreasing association with HLA–A*01 and HLA–B*08 alleles, the susceptibility to CP being primarily associated with HLA–DRB1*03.
Several HLA-mediated mechanisms potentially contribute to the pathogenesis of autoimmune diseases. The HLA system is crucial in positive and negative selection of T cells in the thymus, thus particular HLA molecules can select self-reactive T cell clones; a preferential binding and presentation to potentially self-reactive T cells of peptides derived from autoantigens by the disease-associated HLA molecules may be another possible mechanism (24). Furthermore, it has also been demonstrated that patients with particular alleles such as HLA–DRB1*03 and HLA–B*08 have a spontaneously increased production of TNFα, which is a central regulator of the synthesis of a number of cytokines and adhesion molecules; this may enhance the acute-phase immune response in an antigen-independent way (25).
In our study, a large number of patients presented with constitutional symptoms, high ESR and CRP levels, and also frequently showed positive autoantibodies and autoimmune diseases involving other organs or structures (e.g., autoimmune thyroiditis and rheumatoid arthritis), thus confirming previous findings (5). These manifestations seemed more frequent in the HLA–DRB1*03–positive patients than in the HLA–DRB1*03–negative patients; in particular, the CRP levels at disease onset were higher in patients than in controls, although this difference lost its statistical significance after the correction for multiple tests. This finding suggests that the HLA system not only contributes to disease susceptibility in patients with CP, but may also influence the intensity of the inflammatory/autoimmune response. Replicate studies and larger cohorts of patients are needed to validate our HLA findings and to further investigate the correlations between the HLA profile and the clinical data.
In conclusion, the present study shows that the HLA system plays a role in conferring susceptibility to CP; the strikingly significant association with HLA–DRB1*03, an allele linked to many autoimmune conditions, further supports the hypothesis that CP may belong to the family of autoimmune diseases.
We gratefully acknowledge Dr. Umberto Maggiore for his statistical support, and Drs. Cornelia Weyand and Wei Ma-Krupa for their insightful review of the manuscript.