Analysis of HLA antigens in Turkish sarcoidosis patients.

Background: Sarcoidosis is a systemic granulomatous disorder associated with high CD4+cell activity, without any detectable pathogen. Clustering in families occurs, and the existence of a genetic predisposition to sarcoidosis is widely accepted. There are differences among different ethnic groups.

Methods: We studied HLA polymorphisms in 64 Turkish patients with biopsy proven sarcoidosis. The control group was taken of 160 donor candidates of kidney transplantation within the same period.

Results: Fifty-one patients were female, and 13 were male. The mean age was 39 [+ or -] 6.1 years. Frequency of HLA A2, A9, A24 (9), A25, A69 (28), B12, B22, B38, B49 (21), DR4, and DR14 antigens were significantly higher, and frequencies of HLA B7 and DR7 were significantly less in sarcoidosis patients. Clustering in some families were also noted in our study.

Conclusions: This study implies a genetic predisposition to sarcoidosis in the Turkish population. Clustering in some families should be kept in mind.

Key Words: sarcoidosis, HLA antigens, ethnic group

**********

Sarcoidosis is a multisystemic granulomatous disease of unknown etiology. Prevalence of sarcoidosis in different ethnic groups, familial occurrence and its occurrence among monozygotic twins strongly suggest possible genetic predisposition. (1) Several reports from different countries indicate that certain HLA antigens are more frequent in patients with sarcoidosis. (2,3) In addition, an association between several HLA antigens and clinical features of the disease, such as age of onset, extrapulmonary involvement, radiological stage, response to steroid treatment and sex of patients, has been reported. (4,5) The purpose of this study was to evaluate the association between HLA typing and sarcoidosis patients in Turkey.

Materials and Methods

We studied HLA class I and II antigens in 64 sarcoidosis patients attending Cerrahpasa Medical Faculty. The control group was taken of 160 donor candidates of kidney transplantation within the same period. The age, sex, diagnostic methods, chest x-ray and disease stage at the time of diagnosis of sarcoidosis and extrapulmonary findings of the patients were obtained from the hospital records.

The diagnosis of sarcoidosis was based on clinical, radiological and histopathological criteria. Other granulomatous diseases were excluded, including the presence of noncaseating granulomas in at least one tissue specimen. The absence of mycobacteria and fungi in tissues and cultures were needed for diagnosis of sarcoidosis. Diagnosis was evidenced by histologic examination of transbronchial lung biopsy in 50 patients, mediastinoscopic biopsy in 9 patients, skin biopsy in 2 patients, axillary and epitrochlear lymph node biopsy in 2 patients, liver and bone-marrow biopsy in 1 patient.

Chest x-rays were classified into 3 stages according to De Remee (6): stage I, bilateral hilar lymphadenopathy (BHL); stage II, BHL plus parenchymal infiltration; and stage III, parenchymal infiltration without BHL.

Typing for HLA class I and II antigens was performed with the standard complement-dependent lymphomicrotoxicity test at Cerrahpasa Medical Faculty Blood Center-Immunology Laboratory. For this study, lymphocytes were tested with a panel of sera containing well-characterized HLA-specific alloantibodies. Each serum was placed in a microtiter well of a Terasaki plate (60-72 wells/plate). After a short incubation, rabbit serum was added as a source of complement, and the cells that had bound the alloantibody were lysed, making them permeable to the fluorochrome ethidium bromide. The wells containing the lysed cells were easily discriminated by microscopy. (7)

Associations between HLA markers and sarcoidosis were analyzed by [chi square] test and Fishers probability values, and p values were calculated for the degrees of freedom. In addition, probability values were corrected for multiple comparisons (number of antigens tested) and odds ratios calculated.

Results

Demographic data of control and study patients are shown in Table 1, and HLA typing for class I and II antigens in both groups are given in Table 2. Frequency of HLA A2, A9, A24 (9), A25, A69 (28) B12, B22, B38, DR4, and DR14 antigens were significantly higher in sarcoidosis patients. Frequencies of HLA B7 and DR7 were significantly less in sarcoidosis patients (Table 1). Interestingly, among those patients, there were two cases of identical HLA typing. Three sisters in a family were found to have the same HLA typing of A25 (10), All, B22, B35, DR14 (6), B6, DR51, DR52. Among these three sisters, two had sarcoidosis. HLA typing of another two sisters with sarcoidosis was A25 (10), A68 (21), B41, B35, DR11 (5), DR13. (6)

Discussion

Previous studies point out that there may be a genetic predisposition to sarcoidosis. Predisposition to sarcoidosis has been reported previously to be associated with particular HLA antigens. (1,4) Same disease may be linked with different HLA antigens in different ethnic groups. In this study, the frequencies of HLA A2, A9, A24 (9), A25, B12, B22 and B38 class I antigens and HLA DR4 and DR14 class II antigens were found to be significantly higher than the control group. The antigens with the highest odd ratio were HLAB22, DR14, B12, and B38. In addition, HLA A9 antigen was only detected among sarcoidosis patients (12.5%). Also, two previous studies held in Turkey reported increased frequency of A9 antigen in sarcoidosis patients. (8,9)

The association of sarcoidosis with HLA antigens has been studied in various populations previously. (4,5,9-14) Some of the studies have been summarized in Table 3.

Several studies suggested that HLA B8 is associated with sarcoidosis in Caucasians, whereas such an association could not be demonstrated among Japanese people due to a low frequency of B8 antigen in these populations. (4) Although frequency of HLA B8 in our study was higher in sarcoidosis patients (12.5% versus 8.8%), it was not significant (P > 0.05).

While DRw6 was found to be increased in sarcoidosis patients in Denmark, (10) another work on 123 sarcoidosis patients from central Europe found HLA B8 and B13 antigens to be increased in the disease. (11) HLA DR17 was over represented among Scandinavian sarcoidosis patients and frequencies of DR14 and DR15 correlated well with the chronicity. (13) An Italian study demonstrated the link between sarcoidosis and HLA B8. (14) Another large scale demonstrated a positive correlation with HLA Al, B8, DR3 and a negative correlation with HLA B12 and DR4. (5) In that study, HLA B13 and B35 were associated with early age of onset and DR3 and DR4 with late age of onset. In addition, HLA DR3 was a good prognostic marker.

Two studies from Japan emphasized increased frequency of HLA DRw52 in sarcoidosis patients. (4,12)

There are several reports in literature linking HLA antigens and sarcoidosis, which is a granulomatous disease of unknown etiology. Though the etiology is unknown, the prevalence of sarcoidosis in different ethnic groups and familial occurrence of disease strongly suggests a genetic predisposition. In different populations, various HLA antigens are higher in frequency compared with control groups. Results of our investigation are consistent with other studies regarding HLA association between pathogenesis and clinical presentation of sarcoidosis. As outlined above, since the frequency of HLA antigens differ among nations, to draw a worldwide map of HLA antigen-sarcoidosis association, every country should establish its own HLA spectrum.

References

1. Schurmann M, Lympany PA, Reichel P, et al. Familial sarcoidosis is linked to the major histocompatibility complex region. Am J Respir Crit Care Med 2000;162:861-864.

2. Thomas PD, Hunninghake GW. Current concepts of the pathogenesis of sarcoidosis. Am Rev Respir Dis 1987;135:747-760.

3. Tiwari JL, Terasiki PI. HLA and Disease Association. New York: Springer-Verlag, 1985;380-381.

4. Abe S, Yamaguchi E, Makimura S, et al. Association of HLA DR with sarcoidosis: correlation with clinical course. Chest 1987;92:488-490.

5. Martinetti M, Tinelli C, Kolek V, et al. The sarcoidosis map: a joint survey of clinical and immunogenetic findings in two European countries. Am Respir Crit Care Med 1995;152:557-564.

6. De Remee RA. The roentgenographic staging of sarcoidosis: historic and contemporary perspectives. Chest 1983;83:128-133.

7. Terasaki PI, McCleland JD. Microdroplet assay of human serum cytotxins. Nature 1964;204:998-1000.

8. Akokan G, Celikoglu S, Goksel FM, et al. Antigens in Turkish patients with sarcoidosis. N Engl J Med 1977;296:759.

9. Celik G, Sen E, Ulger AF, et al. Human leukocyte antigens A and B in Turkish patients with sarcoidosis [in Spanish]. Arch Bronchoneumol 2004;40:449-452.

10. Odum N, Milman N, Jakobsen BK, et al. HLA class II (DR, DP, DQ) in patients with sarcoidosis: evidence of an increased frequency of DRw6. Exp Clin Immunogenet 1991;8:227-232.

11. Lenhart K, Kolek V, Bartova A. HLA antigens associated with sarcoidosis. Dis Markers 1990;8:23-29.

12. Kunikane H, Abe S, Tsuneta Y, et al. Role of HLA DR antigens in Japanese patients with sarcoidosis. Am Rev Respir Dis 1987;135:688-691.

13. Berlin M, Fogdell-Hahn A, Olarup O, et al. HLA-DR predicts the prognosis in Scandinavian patients with pulmonary sarcoidosis. Am J Respir Crit Care Med 1997;156:1601-1605.

14. Pasturenzi L, Martinentti M, Cuccia M, et al. HLA class I, II and III polymorphism in Italian patients with sarcoidosis: the Pavia-Padora Sarcoidosis Study Group. Chest 1993;104:1170-1175.

When you get into a tight place and everything goes against you till it
seems you cannot hold on a minute longer, never give up, for that is
just the place or time that the tide will turn.
--Harriet Beecher Stowe

Muammer Bilir, MD, Sevtap Sipahi, MD, Erkan Yilmaz, PhD, Kenan Midilli, MD, Halil Yanardag, MD, Tulin Cagatay, MD, Sabriye Demirci, MD, Tuncer Karayel, MD, and Ergun Erdogan, MD

From the Departments of Internal Medicine, Transplantation-Immunology, and Microbiology and Clinical Microbiology, Cerrahpasa Medical Faculty, University of Istanbul, Istanbul, Turkey.

Reprint requests to Dr. Muammer Bilir, Internal Medicine, Cerrahpasa Medical Faculty, University of Istanbul, TR-34303 Istanbul, Turkey. Email: [e-mail]

Accepted September 22, 2006.

RELATED ARTICLE: Key Points

* Some epidemiological data including clustering in families suggest a genetic predisposition to sarcoidosis.

* A significant difference among the frequency of some HLA antigens is noted in Turkish sarcoidosis patients.

* The difference among the frequencies of HLA antigens in sarcoidosis is not universal and may differ in nations.

Table 1. Demographic data of control and study patients

                              Study patients         Control subjects

Number                        64                     160
Mean age [+ or -] SD (years)  39 [+ or -] 6.1         42 [+ or -] 5.2
Male/female                   13/51                   52/108
Clinical features
  Radiological stage          SI:17, SII:41, SIII:6
  Treatment                   Steroid: 36, no
                                treatment: 28
  Extrapulmonary involvement  Erythema nodosum: 10,
                                uveitis: 6, skin
                                lesions: 6, partial
                                facial paralysis: 1

Table 2. Frequency of HLA antigens in sarcoidosis patients and control
group

           Sarcoidosis
           patients     Control
           (N = 64)     (N = 160)
           n (%)        n (%)      P        OR       CI (%95)

A1         17 (26.6)    30 (18.8)  NS        1.56    0.79-3.10
A2#        36 (56.3)    54 (33.8)   0.002#   2.52#   1.39-4.56#
A3          6 (9.4)     29 (18.1)  NS        0.46    0.18-1.18
A9#         8 (12.5)     0         <0.001#  --       --
A10         6 (9.4)     19 (11.9)  NS        0.76    0.29-2.02
A11         8 (12.5)    20 (12.5)  NS        1       0.41-2.40
A19         4 (6.3)     13 (8.1)   NS        0.75    0.23-2.40
A23         1 (1.6)     11 (6.9)   NS        0.21    0.02-1.70
A24 (9)#   18 (28.1)    15 (9.4)   <0.001#   3.78#   1.76-8.09#
A25#        9 (14.1)     5 (3.1)    0.004#   5.07#   1.62-15.79#
A26         0 (0)        5 (3.1)   NS
A28         3 (4.7)     10 (6.3)   NS        0.73    0.19-2.77
A29         2 (3.1)      2 (1.3)   NS        2.54    0.35-18.49
A30         8 (12.5)    18 (11.3)  NS        1.12    0.46-2.74
A31         2 (3.1)     10 (6.3)   NS        0.48    0.10-2.27
A32         1 (1.6)      7 (4.4)   NS        0.34    0.04-2.87
A36         2 (3.1)      0         NS
A68         2 (3.1)      0         NS
A69 (28)#   3 (4.7)      0          0.023#
B7#         1 (1.6)     17 (10.6)   0.024#   0.134#  0.01-0.92
B8          8 (12.5)    14 (8.8)   NS        1.49    0.59-3.74
B12#       21 (32.8)     5 (3.1)   <0.001#  15.14#   5.39-42.49#
B13         1 (1.6)     13 (8.1)   NS        0.17    0.02-1.40
B14         3 (4.7)     10 (6.3)   NS        0.73    0.19-2.77
B15         0 (0)        6 (3.8)   NS
B16         1 (1.6)      7 (4.4)   NS        0.34    0.04-5.87
B17         2 (3.1)     18 (11.3)  NS        0.25    0.05-1.13
B18         5 (7.8)     14 (8.8)   NS        0.88    0.30-2.56
B21         3 (4.7)      6 (3.8)   NS        1.26    0.30-5.20
B22#       13 (20.3)     2 (1.3)   <0.001#  20.13#   4.39-92.23#
B27         2 (3.1)      2 (1.3)   NS        2.54    0.35-18.49
B35        20 (31.3)    55 (34.4)  NS        0.86    0.46-1.61
B37         2 (3.1)      2 (1.3)   NS        2.54    0.35-18.49
B38#        5 (7.8)      1 (0.6)    0.008#  13.47#   1.54-117.75#
B41         2 (3.1)      0         NS
B42         1 (1.6)      0         NS
B44         5 (7.8)     18 (11.3)  NS        0.66    0.23-1.88
B48         1 (1.6)      0         NS
B49 (21)#   3 (4.7)      0          0.023#
B51 (5)    18 (28.1)    45 (28.1)  NS        1.0     0.52-1.90
B55 (22)    1 (1.6)      0         NS
B60         3 (4.7)      2 (1.3)   NS        3.88    0.63-23.82
B62         3 (4.7)      1 (0.6)   NS        7.82    0.79-76.63
B67         1 (1.6)      5 (3.1)   NS        0.49    0.05-4.29
DR1        12 (18.8)    28 (17.5)  NS        1.08    0.51-2.30
DR2         2 (3.1)     16 (10.0)  NS        0.29    0.06-1.30
DR3         2 (3.1)     13 (8.1)   NS        0.36    0.06-1.66
DR4#       23 (35.9)    33 (20.6)   0.017#   2.15#   1.14-4.08#
DR6         2 (3.1)      1 (0.6)   NS        5.12    0.45-57.58
DR7#        2 (3.1)     27 (16.9)   0.006#   0.15#   0.03-0.68#
DR8         3 (4.7)     16 (10.0)  NS        0.44    0.12-1.57
DR9         1 (1.6)      6 (3.8)   NS        0.40    0.04-3.45
DR10        5 (7.8)     16 (10.0)  NS        0.76    0.27-2.17
DR11       34 (53.1)    70 (43.8)  NS        1.45    0.81-2.60
DR12        2 (3.1)     15 (9.4)   NS        0.31    0.06-1.40
DR13        4 (6.3)     14 (8.8)   NS        0.69    0.22-2.19
DR14#      11 (17.2)     2 (1.3)   <0.001#  16.39#   3.52-76.36#
DR15       21 (32.8)    40 (25.0)  NS        1.46    0.77-2.75
DR18        8 (12.5)    17 (10.6)  NS        1.20    0.49 2.94

Bold data are statistically significant.

Note: Data indicated with # are statistically significant.

Table 3. A summary of some studies investigating HLA types in
sarcoidosis

                                                              HLA types
                                                              with
                  Origin of    No. of    HLA types with       decreased
Study (ref.)      patients     patients  increased frequency  frequency

Abe et al. (4)    Japan         58       DRw52
Martinetti et     Italy/Czech  233       A1, B8, DR3          B12, DR4
  al. (5)
Celik et  (9)     Turkey        83       A9, B5, B8*          A24, A26,
                                                                B62*
Odum et al. (10)  Denmark       41       DRw6                 DR3**
Lenhart et        Moravia      123       B8 and B13
  al. (11)
Kunikane et       Japan         53       DRw52
  al. (12)
Berlin et         Scandinavia  122       DR17 (3)
  al. (13)
Pasturenzi et     Italy        107       B8
  al. (14)
Current study     Turkey        64       A2, A9, A24 (9),     B7, DR7
                                           A25, A69 (28),
                                           B12, B22, B38,
                                           B49 (21), DR4,
                                           DR14

*The differences failed to remain after Bonferroni correction;
**P = 0.07, ie, not statistically significant.