- Email: [email protected]
Association of HLA-DR and HLA-DQ Genes With Susceptibility to Pulmonary Tuberculosis in Koreans: Preliminary Evidence of Associations With Drug Resistance, Disease Severity, and Disease Recurrence
Association of HLA-DR and HLA-DQ Genes With Susceptibility to Pulmonary Tuberculosis in Koreans: Preliminary Evidence of Associations With Drug Resistance, Disease Severity, and Disease Recurrence Hyun Soo Kim, Myoung Hee Park, Eun Young Song, Hyejin Park, Sung Youn Kwon, Sung Koo Han, and Young-Soo Shim ABSTRACT: Only 10% of persons infected with Mycobacterium tuberculosis develop clinical tuberculosis (TB), indicating the existence of host genetic factors regulating disease expression. We investigated the association of human leukocyte antigen (HLA) class II genes with the susceptibility to pulmonary TB in Koreans, with special emphasis on their association with drug resistance, disease severity, and disease recurrence. Human leukocyte antigens (HLA)-DRB1 and -DQB1 gene polymorphisms were analyzed in 160 Korean patients with pulmonary TB and 200 ethnically matched healthy controls. HLADRB1*0803 (25.0% vs. 14.5% in controls, OR ⫽ 1.97, p ⫽ 0.012, corrected p (pc) ⬎ 0.05) and DQB1*0601 (27.5% vs. 15.5% in controls, OR ⫽ 2.07, p ⫽ 0.005, pc ⬎ 0.05) were weakly associated with general susceptibility to TB. DRB1*0803 was significantly associated with drug resistance (30.9% vs. 14.5%, OR ⫽ 2.63, pc ⫽ ABBREVIATIONS HLA human leukocyte antigen PCR polymerase chain reaction
INTRODUCTION It is estimated that approximately one third of the world’s population is infected with Mycobacterium tuberFrom the Departments of Laboratory Medicine (H.S.K., M.H.P., E.Y.S., H.P.) and Internal Medicine (S.Y.K., S.K.H., Y.S.S.), Seoul National University College of Medicine, Seoul, Korea, and the Department of Laboratory Medicine (H.S.K.), Bundang Jesaeng General Hospital, Sungnam, Korea. Address reprint requests to: Dr. Myoung Hee Park, Department of Laboratory Medicine, Seoul National University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, Korea; E-mail: [email protected]. Supported by a grant of the Korea Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea (02-PJ1-PG3-20501-0010). Received July 20, 2005; accepted August 31, 2005. Human Immunology 66, 1074 –1081 (2005) © American Society for Histocompatibility and Immunogenetics, 2005 Published by Elsevier Inc.
0.047) and more advanced lung lesion (29.8% vs. 14.5%, OR ⫽ 2.50, pc ⫽ 0.022). DRB1*0803 showed the strongest association with disease recurrence, especially after curative treatment for the earlier infection (47.4% vs. 14.5%, OR ⫽ 5.31, pc ⫽ 0.00009). DQB1*0601, which is strongly linked to DRB1*0803 in this population showed similar changes in the patients as those of DRB1*0803. It is suggested that DRB1*0803 and DQB1*0601 alleles are associated with disease progression of TB in Koreans, exerting influence on the development of drug resistance, severe disease, and recurrent disease. Human Immunology 66, 1074 –1081 (2005). © American Society for Histocompatibility and Immunogenetics, 2005. Published by Elsevier Inc. KEYWORDS: Tuberculosis; HLA-DR; HLA-DQ
TB
tuberculosis
culosis. An interesting feature of mycobacterial infection is that clinical disease, tuberculosis (TB), develops in only 10% of those infected, which indicates the existence of host factors regulating disease expression [1]. Presence of racial differences in the susceptibility to TB and higher concordance rate for TB among monozygous than among dizygous twins are evidences that the host genetic factors play an important role in the development of TB in humans [1–3]. As one of the host genetic factors, associations of human leukocyte antigen (HLA) with susceptibility to 0198-8859/05/$–see front matter doi:10.1016/j.humimm.2005.08.242
Association of HLA-DR and HLA-DQ With Tuberculosis
1075
TABLE 1 Demographic data of patients with pulmonary tuberculosis
Age (mean ⫾ SD) Sex (men:women) Radiographic lung lesion Minimal Moderately advanced Far advanced Treatmentb Initial treatment Retreatment, total Retreatment, group I Retreatment, group II a b
Total (n ⫽ 160)
Drug sensitive (n ⫽ 79)
Drug resistanta (n ⫽ 81)
44.1 ⫾ 16.2 91:61
46.5 ⫾ 16.6 44:35
41.8 ⫾ 15.6 47:34
29 (18.1%) 65 (40.6%) 66 (41.3%)
26 (32.9%) 35 (44.3%) 18 (22.8%)
3 (3.7%) 30 (37.0%) 48 (59.3%)
106 (66.3%) 54 (33.8%) 38 (23.8%) 16 (10.0%)
64 (81.0%) 15 (19.0%) 15 (19.0%) 0 (0.0%)
42 (51.9%) 39 (48.1%) 23 (28.4%) 16 (19.8%)
Including 67 cases of multidrug resistance. Retreatment: patients with previous histories of curative treatment (group I) or inadequate treatment (group II).
TB have been studied in many ethnic groups. Although the results have been inconsistent among various ethnic groups [4], the association of HLA-DR2 or DRB1*15 (a split of DR2) with TB has been detected most commonly, especially in Asian populations including Asian Indian [5– 8], Indonesian [9], and five ethnic groups in Russia [10]. Association of HLA-DR2 or its splits with TB has also been observed in Polish (DRB1*16) and Mexican (DRB1*15) [11, 12]. Most of the studies on the HLA association with TB have been simply focused on its association with general susceptibility to TB, and association with disease severity or drug resistance has rarely been studied [5, 6]. The prevalence of active pulmonary TB in Koreans is still high (1%, 1,032 per 100,000 in 1995 survey) and drug resistance rate is nearly 10% [13]. Our purpose in this study was to investigate the association of HLA with pulmonary TB in Koreans with special emphasis on its association with drug resistance, disease severity, or disease recurrence. MATERIALS AND METHODS Subjects The study used a case-control design to compare between healthy controls and pulmonary TB patients. A total of 160 Korean patients (91 men and 69 women, age range 20 – 81 years, mean age 44.1 years) with pulmonary TB were enrolled at the Respiratory Clinic of the Seoul National University Hospital (Table 1). We designed a study including a much larger proportion of drugresistant patients than normally encountered (about 20% in our institute), so that about equal number of drug-sensitive and drug-resistant patients were recruited and about one third of the patients were recurrent cases with previous chemotherapy histories for TB.
The diagnosis was confirmed by the cultivation of M. tuberculosis in sputum culture with or without the presence of acid-fast bacilli in sputum smear and by standard clinical and radiologic investigations. No patient in this study had evidence of infection with human immunodeficiency virus, and anti– human immunodeficiency virus antibody was negative in all patients. Drug sensitivity test was performed as a referral test at the Korean National Tuberculosis Association against 11 drugs: isoniazid, rifampicin, ethambutol, pyrazinamide, streptomycin, kanamycin, prothionamide, cycloserine, para-aminosalicylate, ofloxacin, and enviomycin. According to the results of drug sensitivity test, patients whose isolates were all sensitive to tested drugs were classified as the drug-sensitive group and those whose isolates were resistant to one or more drugs tested as the drug-resistant group (including multidrug resistance with resistance to both isoniazid and rifampicin). For chemotherapy, four drug regimens including isoniazid and rifampicin were used for initial treatment and drug-sensitive patients. For drug-resistant patients, individualized regimen according to the drug sensitivity test was used, including at least four sensitive drugs. Pulmonary TB patients were categorized according to drug sensitivity, the extent of lung lesion (minimal, moderately advanced, and far advanced) [14], and the history of treatment (initially treated, retreated after curative treatment, and retreated after inadequate treatment) (Table 1). The study was approved by the institutional review board of the Seoul National University Hospital and informed consent was obtained from all subjects. Two hundred ethnically matched healthy individuals served as controls. They were 200 unrelated parents (101 women, 99 men) of Korean families, previously typed for
1076
HLA class II alleles and analyzed for class II haplotypes by segregation study [15]. HLA Typing Genomic DNA was extracted from peripheral blood by use of QIAamp blood kit (Qiagen, Hilden, Germany). HLA-DRB1 genotyping was carried out in two steps. First, low- to intermediate-resolution typing was performed by polymerase chain reaction (PCR)-sequence specific oligonucleotide hybridization method by using the Dynal RELITM SSO HLA-DRB Typing kit (Dynal Biotech Ltd., Wirral, UK). Then high-resolution DRB1 typing was carried out by group-specific amplification and PCR-single strand conformation polymorphism method [16]. HLA-DQB1 typing was performed using a combination of PCR-restriction fragment length polymorphism and PCR-single strand conformation polymorphism methods, as described previously [17]. Statistical Analysis Phenotype frequencies of the HLA-DRB1 and HLADQB1 alleles and DRB1-DQB1 haplotypes were compared by the 2 test or Fisher’s exact test as appropriate. The level of significance was set at p ⬍ 0.05, and odds ratios (ORs) with 95% confidence intervals (CI) were calculated for those comparisons showing significant p values. Where indicated as corrected p (pc), a Bonferroni correction was applied by multiplying the probability value by the number of comparisons made (i.e., 29 for HLA-DRB1 alleles, 15 for HLA-DQB1 alleles, and 48 for DRB1-DQB1 haplotypes). RESULTS HLA-DRB1 and HLA-DQB1 Associations With Total Group of TB Patients HLA-DRB1 and HLA-DQB1 frequencies in pulmonary TB patients and controls are listed in Tables 2 and 3. When compared with controls, TB patients showed increased frequencies of DRB1*0803 (OR ⫽ 1.97, p ⫽ 0.012, pc ⬎ 0.05), DQB1*0601 (OR ⫽ 2.07, p ⫽ 0.005, pc ⬎ 0.05), and DQB1*0609 (OR ⫽ 2.67, p ⫽ 0.023, pc ⬎ 0.05), and decreased frequency of DRB1*0101 (OR ⫽ 0.45, p ⫽ 0.031, pc ⬎ 0.05). HLA-DRB1 and HLA-DQB1 Associations With Subgroups of TB Patients Among the four HLA alleles listed previously, which showed weak associations with general susceptibility to TB, DRB1*0803 and DQB1*0601 alleles showed stronger associations with subgroups of TB patients in the context of drug resistance, disease severity, and disease recurrence (i.e., previous treatment history for pulmonary TB). Compared with controls (14.5%), the frequency of DRB1*0803 was significantly increased in the drug-
H.S. Kim et al.
resistant group (30.9%, OR ⫽ 2.63, pc ⫽ 0.047), but not in the drug-sensitive group (Tables 2 and 4). Stratification for disease severity revealed that DRB1*0803 was significantly increased in more advanced (moderately or far advanced) group (29.8%, OR ⫽ 2.50, pc ⫽ 0.022), and this increase was also significant when compared with the minimal group (3.4%, OR ⫽ 11.87, p ⫽ 0.003). According to the history of TB treatment, the frequency of DRB1*0803 was significantly increased in the retreated group (37.0%, OR ⫽ 3.47, pc ⫽ 0.006), and this increase was also significant when compared with the initially treated group (18.9%, OR ⫽ 2.53, p ⫽ 0.012). When the retreated group was divided into group I (with previous history of curative treatment) and group II (with previous history of inadequate treatment), significant increase was observed only in the group I (47.4%, OR ⫽ 5.31, pc ⫽ 0.00009), and not in the group II (12.5%). Significant increase of DRB1*0803 frequency in the group I retreated patients was observed in both drug-sensitive (40.0%, OR ⫽ 3.93, p ⫽ 0.02) and drug-resistant (52.2%, OR ⫽ 6.43, pc ⫽ 0.003) patients. DQB1*0601, which is in strong linkage disequilibrium with DRB1*0803 in Koreans [15], showed similar changes in the frequencies as those found for DRB1*0803 (Tables 3 and 5). Compared with controls, significant increase of DQB1*0601 frequency was observed in drug-resistant (OR ⫽ 2.58, pc ⫽ 0.026), more advanced (OR ⫽ 2.48, pc ⫽ 0.01), and retreated (OR ⫽ 3.21, pc ⫽ 0.007) groups of patients. An especially marked increase was observed in the group I retreated patients (OR ⫽ 5.45, pc ⫽ 0.00003). In addition, patients with primary drug resistance (initially treated drug-resistant patients) showed an increased frequency of DQB1*0601 (OR ⫽ 2.44, p ⫽ 0.018). DRB1-DQB1 Haplotype Associations With TB Patients DRB1-DQB1 haplotypes in the controls were analyzed by family-based segregation study, as previously reported [15]. Because the DRB1 and DQB1 alleles are in strong linkage disequilibrium in this population, we could deduce the probable DRB1-DQB1 haplotypes in the patients (data not shown). Significant changes in the frequency of DRB1*0803-DQB1*0601 were observed in TB patients, and the level of significance was slightly lower than that of DRB1*0803 or DQB1*0601. Compared with controls (12.5%), the frequency of DRB1*0803-DQB1*0601 was increased in TB patients (21.3%, OR ⫽ 1.89, p ⫽ 0.026). Significant increase was observed in drug-resistant (25.9%, OR ⫽ 2.45, p ⫽ 0.006), more advanced (25.2%, OR ⫽ 2.36, p ⫽ 0.003), and retreated (31.5%, OR ⫽ 3.22, pc ⫽ 0.041) groups of patients. Especially marked increase was observed in the
Association of HLA-DR and HLA-DQ With Tuberculosis
1077
TABLE 2 HLA-DRB1 frequencies in patients with pulmonary tuberculosis and healthy controls Patients HLA-DRB1
Controls (n ⫽ 200) (%)
Total (n ⫽ 160) (%)
Drug sensitive (n ⫽ 79) (%)
Drug resistant (n ⫽ 81) (%)
DRB1*0101 0301 0401 0403 0404 0405 0406 0407 0410 0701 0802 0803 0901 1001 1101 1201 1202 1301 1302 1307 1401 1402 1403 1405 1406 1407 1501 1502 1602
28 (14.0) 7 (3.5) 0 (0.0) 10 (5.0) 4 (2.0) 32 (16.0) 22 (11.0) 2 (1.0) 4 (2.0) 27 (13.5) 12 (6.0) 29 (14.5) 33 (16.5) 9 (4.5) 13 (6.5) 18 (9.0) 19 (9.5) 6 (3.0) 31 (15.5) 0 (0.0) 11 (5.5) 2 (1.0) 5 (2.5) 15 (7.5) 0 (0.0) 1 (0.5) 34 (17.0) 9 (4.5) 0 (0.0)
11 (6.9)a 6 (3.8) 2 (1.3) 13 (8.1) 5 (3.1) 22 (13.8) 18 (11.3) 0 (0.0) 3 (1.9) 22 (13.8) 3 (1.9) 40 (25.0)b 29 (18.1) 4 (2.5) 19 (11.9) 12 (7.5) 7 (4.4) 4 (2.5) 32 (20.0) 1 (0.6) 6 (3.8) 0 (0.0) 2 (1.3) 9 (5.6) 1 (0.6) 1 (0.6) 21 (13.1) 14 (8.8) 2 (1.3)
5 (6.3) 5 (6.3) 1 (1.3) 8 (10.1) 1 (1.3) 10 (12.7) 13 (16.5) 0 (0.0) 1 (1.3) 11 (13.9) 2 (2.5) 15 (19.0) 16 (20.3) 3 (3.8) 9 (11.4) 6 (7.6) 3 (3.8) 2 (2.5) 15 (19.0) 0 (0.0) 3 (3.8) 0 (0.0) 2 (2.5) 1 (1.3) 0 (0.0) 1 (1.3) 13 (16.5) 6 (7.6) 1 (1.3)
6 (7.4) 1 (1.2) 1 (1.2) 5 (6.2) 4 (4.9) 12 (14.8) 5 (6.2) 0 (0.0) 2 (2.5) 11 (13.6) 1 (1.2) 25 (30.9)c 13 (16.0) 1 (1.2) 10 (12.3) 6 (7.4) 4 (4.9) 2 (2.5) 17 (21.0) 1 (1.2) 3 (3.7) 0 (0.0) 0 (0.0) 8 (9.9) 1 (1.2) 0 (0.0) 8 (9.9) 8 (9.9) 1 (1.2)
p ⫽ 0.031, OR ⫽ 0.45 (95% CI 0.22– 0.94). p ⫽ 0.012, OR ⫽ 1.97 (95% CI 1.15–3.35). c p ⫽ 0.002, pc ⫽ 0.047, OR ⫽ 2.63 (95% CI 1.42– 4.87). a
b
group I retreated patients (39.5%, OR ⫽ 4.57, pc ⫽ 0.002). DISCUSSION In this study, we have found that, although HLADRB1*0803 and HLA-DQB1*0601 alleles were weakly associated with general susceptibility to pulmonary TB in the Korean population, these alleles were strongly associated with drug resistance, disease severity, and disease recurrence, especially in patients with previous history of curative treatment. HLA associations with drug resistance or disease severity of pulmonary TB have only rarely been reported. In Asian Indians, DR2 association with pulmonary TB (relative risk ⫽ 1.8) was reported to be stronger in drug failure group of patients (relative risk ⫽ 3.7) [6]. In South Indians, DR2 association with pulmonary TB (relative risk ⫽ 2.2) was reported to be stronger in sputum smear–positive and far-advanced group of patients (relative risk ⫽ 3.3) [5].
Drug resistance, especially multidrug resistance of M. tuberculosis strains, is an increasing problem worldwide. The World Health Organization has warned about the growing prevalence of drug-resistant TB. The areas worst affected are eastern Europe and central Asia, and as many as 14% of new cases in these regions involve multidrugresistant strains of the bacterium [18]. Widely recognized risk factors for drug resistance include previous treatment for TB, residence in an area endemic for drug resistance, and close contact with a known case of drugresistant TB [19]. In addition to these risk factors, our study suggests that genetic factors of the host are associated with the development of drug resistance in TB patients. In the present study, drug-resistant TB was significantly associated with DRB1*0803 and DQB1*0601, whereas drug-sensitive TB patients did not show significant changes in the frequencies of these alleles. Drug resistance may be primary or acquired: primary drug resistance in patients with no prior TB treatment, and
1078
H.S. Kim et al.
TABLE 3 HLA-DQB1 frequencies in patients with pulmonary tuberculosis and healthy controls Patients HLA-DQB1
Controls (n ⫽ 200) (%)
Total (n ⫽ 160) (%)
Drug sensitive (n ⫽ 79) (%)
Drug resistant (n ⫽ 81) (%)
DQB1*0201 0202 0301 0302 0303 0401 0402 0501 0502 0503 0601 0602 0603 0604 0609
7 (3.5) 25 (12.5) 53 (26.5) 40 (20.0) 41 (20.5) 31 (15.5) 13 (6.5) 37 (18.5) 6 (3.0) 20 (10.0) 31 (15.5) 33 (16.5) 7 (3.5) 25 (12.5) 8 (4.0)
6 (3.8) 20 (12.5) 45 (28.1) 34 (21.3) 32 (20.0) 21 (13.1) 8 (5.0) 19 (11.9) 8 (5.0) 12 (7.5) 44 (27.5)a 20 (12.5) 4 (2.5) 16 (10.0) 16 (10.0)c
5 (6.3) 10 (12.7) 21 (26.6) 21 (26.6) 17 (21.5) 10 (12.7) 2 (2.5) 10 (12.7) 5 (6.3) 3 (3.8) 18 (22.8) 13 (16.5) 2 (2.5) 7 (8.9) 8 (10.1)
1 (1.2) 10 (12.3) 24 (29.6) 13 (16.0) 15 (18.5) 11 (13.6) 6 (7.4) 9 (11.1) 3 (3.7) 9 (11.1) 26 (32.1)b 7 (8.6) 2 (2.5) 9 (11.1) 8 (9.9)
p ⫽ 0.005, OR ⫽ 2.07 (95% CI 1.23–3.47). p ⫽ 0.002, pc ⫽ 0.026, OR ⫽ 2.58 (95% CI 1.41– 4.71). c p ⫽ 0.023, OR ⫽ 2.67 (95% CI 1.11– 6.40). a
b
TABLE 4 Changes of HLA-DRB1*0803 frequencies in various groups of pulmonary tuberculosis patients compared with controls Controls and patients
No. (%)
Controls (n ⫽ 200) Drug sensitivity (n ⫽ 160) Drug sensitive (n ⫽ 79) Drug resistant (n ⫽ 81) Radiographic lung lesion (n ⫽ 160) Minimal (n ⫽ 29) Moderately advanced (n ⫽ 65) Far advanced (n ⫽ 66) Moderately ⫹ far advanced (n ⫽ 131) Treatmentb (n ⫽ 160) Initial treatment (n ⫽ 106) Retreatment, total (n ⫽ 54) Retreatment, group I (n ⫽ 38) Retreatment, group II (n ⫽ 16) Drug sensitive (n ⫽ 79) Initial treatment (n ⫽ 64) Retreatment, group I (n ⫽ 15) Drug resistant (n ⫽ 81) Initial treatment (n ⫽ 42) Retreatment, group I (n ⫽ 23) Retreatment, group II (n ⫽ 16)
29 (14.5)
p Value
pc Value
Odds ratio (95% CI)
15 (19.0) 25 (30.9)
NS 0.002
0.047
2.63 (1.42–4.87)
1 (3.4) 18 (27.7) 21 (31.8) 39 (29.8)a
NS 0.016 0.002 0.0008
NS NS 0.022
2.26 (1.15–4.42) 2.75 (1.44–5.27) 2.50 (1.45–4.30)
20 (18.9) 20 (37.0)c 18 (47.4)d 2 (12.5)
NS 0.0002 0.000003 NS
0.006 0.00009
3.47 (1.76–6.83) 5.31 (2.51–11.23)
NS
3.93 (1.30–11.87)
0.003
6.43 (2.59–15.95)
9 (14.1) 6 (40.0)e
NS 0.020
11 (26.2) 12 (52.2)f 2 (12.5)
NS 0.0001 NS
Abbreviation: NS ⫽ not significant. a Versus minimal: p ⫽ 0.003, pc ⫽ NS, OR ⫽ 11.87 (95% CI 1.56 –90.33). b Retreatment: patients with previous histories of curative treatment (group I) or inadequate treatment (group II). c Versus initial treatment: p ⫽ 0.012, pc ⫽ NS, OR ⫽ 2.53 (95% CI 1.21–5.28). d Versus initial treatment: p ⫽ 0.0006, pc ⫽ 0.018, OR ⫽ 3.87 (95% CI 1.74 – 8.63). e Versus initial treatment: p ⫽ 0.032, pc ⫽ NS, OR ⫽ 4.07 (95% CI 1.17–14.23). f Versus initial treatment: p ⫽ 0.036, pc ⫽ NS, OR ⫽ 3.07 (95% CI 1.06 – 8.95).
Association of HLA-DR and HLA-DQ With Tuberculosis
1079
TABLE 5 Changes of HLA-DQB1*0601 frequencies in various groups of pulmonary tuberculosis patients compared with controls Controls and patients
No. (%)
Controls (n ⫽ 200) Drug sensitivity (n ⫽ 160) Drug sensitive (n ⫽ 79) Drug resistant (n ⫽ 81) Radiographic lung lesion (n ⫽ 160) Minimal (n ⫽ 29) Moderately advanced (n ⫽ 65) Far advanced (n ⫽ 66) Moderately ⫹ far advanced (n ⫽ 131) Treatmentb (n ⫽ 160) Initial treatment (n ⫽ 106) Retreatment, total (n ⫽ 54) Retreatment, group I (n ⫽ 38) Retreatment, group II (n ⫽ 16) Drug sensitive (n ⫽ 79) Initial treatment (n ⫽ 64) Retreatment, group I (n ⫽ 15) Drug resistant (n ⫽ 81) Initial treatment (n ⫽ 42) Retreatment, group I (n ⫽ 23) Retreatment, group II (n ⫽ 16)
31 (15.5)
p Value
pc Value
Odds ratio (95% CI)
18 (22.8) 26 (32.1)
NS 0.002
0.026
2.58 (1.41–4.71)
3 (10.3) 20 (30.8) 21 (31.8) 41 (31.3)a
NS 0.007 0.004 0.0007
NS NS 0.010
2.43 (1.26–4.65) 2.54 (1.34–4.85) 2.48 (1.46–4.23)
24 (22.6) 20 (37.0) 19 (50.0)c 1 (6.3)
NS 0.0005 0.000002 NS
0.007 0.00003
3.21 (1.64–6.28) 5.45 (2.60–11.45)
11 (17.2) 7 (46.7)d
NS 0.007
NS
4.77 (1.61–14.11)
13 (31.0) 12 (52.2)e 1 (6.3)
0.018 0.0002 NS
NS 0.003
2.44 (1.15–5.22) 5.95 (2.41–14.68)
Abbreviation: NS ⫽ not significant. a Versus minimal: p ⫽ 0.022, pc ⫽ NS, OR ⫽ 3.95 (95% CI 1.13–13.79). b Retreatment: patients with previous histories of curative treatment (group I) or inadequate treatment (group II). c Versus initial treatment: p ⫽ 0.002, pc ⫽ 0.024, OR ⫽ 3.42 (95% CI 1.56 –7.47). d Versus initial treatment: p ⫽ 0.035, pc ⫽ NS, OR ⫽ 4.22 (95% CI 1.26 –14.06). e Versus initial treatment: p ⫽ 0.021, pc ⫽ NS, OR ⫽ 3.27 (95% CI 1.17–9.18).
acquired drug resistance in patients previously treated for TB. HLA class II alleles were strongly associated with acquired drug resistance in this study, especially in patients with previous history of curative treatment for TB (for DRB1*0803, OR ⫽ 6.43, pc ⫽ 0.003; for DQB1*0601, OR ⫽ 5.95, pc ⫽ 0.003). Our finding is further supported by a recent report in Asian Indians that the presence of HLA-DRB1*14 was an independent predictor of multidrug-resistant TB (OR ⫽ 8.2) [20]. It can be assumed that patients carrying particular HLA class II alleles are incapable of mounting adequate immune response to tubercle bacilli and thus are prone to develop drug resistance. The most interesting finding in this study was that DRB1*0803 and DQB1*0601 alleles were strongly associated with recurrent TB. Among patients with recurrent TB, the association was significant only in patients with previous history of curative treatment (for DRB1*0803, OR ⫽ 5.31, pc ⫽ 0.00009; for DQB1*0601, OR ⫽ 5.45, pc ⫽ 0.00003), but not in those with previous history of inadequate treatment. It is conceivable that the recurrence of TB after inadequate or incomplete treatment would not be related to the host genetic factors. Recurrence of TB after curative treatment could arise from reactivation of the earlier infection
or recent reinfection [21, 22]. The recurrence of TB in patients with previous treatment history is generally assumed to be due to inadequate treatment for the earlier infection, mostly because of poor compliance with drug therapy. However, in this study, we have found that recurrence of TB even after curative treatment is strongly associated with HLA class II alleles. To our knowledge, no other study has clearly shown the association of HLA or other genetic factors with recurrent TB in humans after curative treatment for the earlier infection. Because of the very strong linkage disequilibrium between DRB1*0803 and DQB1*0601 alleles in the Korean population [15], we could not determine which allele is primarily associated with TB susceptibility. In addition, DRB1*0803-DQB1*0601 haplotype showed similar changes in TB patients as those of DRB1*0803 and DQB1*0601. DRB1*0803 is present at relatively high frequency in East Asian populations including Korean, Japanese, and Chinese, but is rare in white or black populations [23], and its association with TB susceptibility has not been reported in other populations. Weak association of DR8 with TB has only rarely been reported in South Indians (OR ⫽ 5.99, p ⫽ 0.035) and Hong Kong Chinese (relative risk ⫽ 2.26, p ⫽ 0.028) [8, 24]. In contrast, DR2 or DRB1*15 (most commonly
1080
DRB1*1501) has been reported in many ethnic groups to be the main HLA allele associated with TB susceptibility [5–10, 12]. However, despite quite a high frequency of DR2 or DRB1*1501 in the Korean population, this allele did not show increased frequency in TB patients. Thus DRB1*0803 might not be primarily associated with TB susceptibility and the observed association might be secondary to the DQB1*0601 (a split of DQ1) association. This is partially supported by other reports in Asian Indians that DRB1*1501-DQB1*0601 haplotype was associated with TB susceptibility [8, 25]. DQB1*05, another split of DQ1, has also been reported to be associated with TB susceptibility in many ethnic groups: DQB1*0501 in Mexicans (together with DRB1*1501), DQB1*0502 in Asian Indians (together with DRB1*1501) and in Thais, DQB1*0503 in Cambodians, and DQB1*05 in Poles [12, 25–28]. It is possible that DQ1 and its splits DQB1*05 and DQB1*06 alleles are primarily associated with TB susceptibility. However, there is still another possibility that other major histocompatibility complex genes closely linked to the HLA-DRB1 and -DQB1 genes are primarily associated with TB susceptibility. The specific immune response to M. tuberculosis infection mainly depends on the activation of various T-cell subsets that produce cytokines, mainly interferon-␥ and tumor necrosis factor-␣, which orchestrate appropriate defense mechanism [29]. CD4 T cells, CD8 T cells, ␥␦ T cells, and CD1-restricted T cells are participating in the immune response to M. tuberculosis, among which CD4 T cells play the major role in containing TB at all stages of the disease [29, 30]. CD4 T cells are recognizing antigenic peptides in the context of HLA class II molecules, and thus particular HLA class II alleles might not be able to present appropriate antigenic peptides of tubercle bacilli to CD4 T cells. In the present study of Koreans, gene products of DRB1*0803, DQB1*0601, or other closely linked genes are suggested to have some defects in presenting appropriate antigenic peptides of tubercle bacilli. Lack of an effective immune response in carriers of DRB1*0803 or DQB1*0601 seems to be a plausible and unifying explanation for the increased proportion of severe disease, disease recurrence, and drug resistance among patients carrying these alleles. In summary, it is suggested that DRB1*0803 and DQB1*0601 alleles are associated with TB susceptibility in Koreans. DRB1*0803 and DQB1*0601 were only weakly associated with general susceptibility to TB. However, individuals possessing these alleles, after being infected, were strongly associated with the development of drug resistance, more advanced lung lesion, and disease recurrence even after curative treatment for the earlier infection. These findings are probably interrelated and the patients possessing these alleles might not be
H.S. Kim et al.
able to mount an effective immune response against M. tuberculosis.
REFERENCES 1. Bellamy R: Genetic susceptibility to tuberculosis in human populations. Thorax 53:588, 1998. 2. Stead WW, Senner JW, Reddick WT, Lofgren JP: Racial differences in susceptibility to infection by Mycobacterium tuberculosis. N Engl J Med 322:422, 1990. 3. Comstock GW: Tuberculosis in twins: a re-analysis of the Prophit survey. Am Rev Resp Dis 117:621, 1978. 4. Meyer CG, May J, Stark K: Human leukocyte antigens in tuberculosis and leprosy. Trends Microbiol 6:148, 1998. 5. Brahmajothi V, Pitchappan RM, Kakkanaiah VN, Sashidhar M, Rajaram K, Ramu S, Palanimurugan K, Paramasivan CN, Prabhakar R: Association of pulmonary tuberculosis and HLA in South India. Tubercle 72:123, 1991. 6. Rajalingam R, Mehra NK, Jain RC, Myneedu VP, Pande JN: Polymerase chain reaction-based sequence-specific oligonucleotide hybridization analysis of HLA class II antigens in pulmonary tuberculosis: relevance to chemotherapy and disease severity. J Infect Dis 173:669, 1996. 7. Selvaraj P, Uma H, Reetha AM, Kurian SM, Xavier T, Prabhakar R, Narayanan PR: HLA antigen profile in pulmonary tuberculosis patients & their spouses.Indian J Med Res 107:155, 1998. 8. Ravikumar M, Dheenadhayalan V, Rajaram K, Lakshmi SS, Kumaran PP, Paramasivan CN, Balakrishnan K, Pitchappan RM: Associations of HLA-DRB1, DQB1 and DPB1 alleles with pulmonary tuberculosis in south India. Tuber Lung Dis 79:309, 1999. 9. Bothamley GH, Beck JS, Schreuder GM, D’Amaro J, de Vries RR, Kardjito T, Ivanyi J: Association of tuberculosis and M. tuberculosis-specific antibody levels with HLA. J Infect Dis 159:549, 1989. 10. Khomenko AG, Litvinov VI, Chukanova VP, Pospelov LE: Tuberculosis in patients with various HLA phenotypes. Tubercle 71:187, 1990. 11. Dubaniewicz A, Lewko B, Moszkowska G, Zamorska B, Stepinski J: Molecular subtypes of the HLA-DR antigens in pulmonary tuberculosis. Int J Infect Dis 4:129, 2000. 12. Terán-Escandón D, Terán-Oritz L, Camarena-Olvera A, González-Avila G, Vaca-Marín MA, Granados J, Selman M: Human leukocyte antigen-associated susceptibility to pulmonary tuberculosis: molecular analysis of class II alleles by DNA amplification and oligonucleotide hybridization in Mexican patients. Chest 115:428, 1999. 13. Hong YP, Kim SJ, Lew WJ, Lee EK, Han YC: The seventh nationwide tuberculosis prevalence survey in Korea, 1995. Int J Tuberc Lung Dis 2:27, 1998. 14. American Thoracic Society: Diagnostic standards and Classification of Tuberculosis. 12th ed. New York, Na-
Association of HLA-DR and HLA-DQ With Tuberculosis
15.
16.
17.
18.
19.
20.
21.
22.
tional Tuberculosis and Respiratory Disease Association, 1969. Song EY, Park MH, Kang SJ, Park HJ, Kim BC, Tokunaga K, Akaza T, Juji T: HLA class II allele and haplotype frequencies in Koreans based on 107 families. Tissue Antigens 59:475, 2002. Bannai M, Tokunaga K, Lin L, Kuwata S, Mazda T, Amaki T, Fujisawa K, Juji T: Discrimination of human HLA-DRB1 alleles by PCR-SSCP (single-strand conformation polymorphism) method. Eur J Immunogenet 21:1, 1994. Park MH, Whang DH, Kang SJ: High resolution HLADQB1 typing by combination of PCR-RFLP and PCRSSCP. Hum Immunol 60:901, 1999. World Health Organization: Anti-tuberculosis drug resistance in the world: third global report. Geneva, WHO, 2004. Harkin TJ, Condos R: Management of multidrugresistant tuberculosis. In Rom WN, Garay SM (eds): Tuberculosis. 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. Sharma SK, Turaga KK, Balamurugan A, Saha PK, Pandey RM, Jain NK, Katoch VM, Mehra NK: Clinical and genetic risk factors for the development of multi-drug resistant tuberculosis in non-HIV infected patients at a tertiary care center in India: a case-control study. Infect Genet Evol 3:183, 2003. Gomez JE, McKinney JD: Persistence and drug resistance. In Rom WN, Garay SM (eds): Tuberculosis. 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. van Rie A, Warren R, Richardson M, Victor TC, Gie RP, Enarson DA, Beyers N, van Helden PD: Exogenous rein-
1081
23.
24.
25.
26.
27.
28.
29.
30.
fection as a cause of recurrent tuberculosis after curative treatment. N Engl J Med 341:1174, 1999. Imanishi T, Akaza T, Kimura A, Tokunaga K, Gojobori T: Allele and haplotype frequencies for HLA and complement loci in various ethnic groups. In Tsuji K, Aizawa M, Sasazuki T (eds): HLA 1991. Vol I. New York, Oxford University Press, 1992. Hawkins BR, Higgins DA, Chan SL, Lowrie DB, Mitchison DA, Girling DJ: HLA typing in the Hong Kong Chest Service/British Medical Research Council study of factors associated with the breakdown to active tuberculosis of inactive pulmonary lesions. Am Rev Respir Dis 138:1616, 1988. Mehra NK, Rajalingam R, Mitra DK, Taneja V, Giphart MJ: Variants of HLA-DR2/DR51 group haplotypes and susceptibility to tuberculoid leprosy and pulmonary tuberculosis in Asian Indians. Int J Lepr 63:241, 1995. Vejbaesya S, Chierakul N, Luangtrakool K, Srinak D, Stephens HA: Association of HLA class II alleles with pulmonary tuberculosis in Thais. Eur J Immunogenet 29:431, 2002. Goldfeld AE, Delgado JC, Thim S, Bozon MV, Uglialoro AM, Turbay D, Cohen C, Yunis EJ: Association of an HLA-DQ allele with clinical tuberculosis. JAMA 279: 226, 1998. Dubaniewicz A, Moszkowska G, Szczerkowska Z, Hoppe A: Analysis of DQB1 allele frequencies in pulmonary tuberculosis: preliminary report. Thorax 58:890, 2003. Ulrichs T, Kaufmann SH: Cell-mediated immune response. In Rom WN, Garay SM (eds): Tuberculosis. 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. Kaufmann SH: Protection against tuberculosis: cytokines, T cells, and macrophages. Ann Rheum Dis 61(Suppl 2):ii54, 2002.
INTRODUCTION It is estimated that approximately one third of the world’s population is infected with Mycobacterium tuberFrom the Departments of Laboratory Medicine (H.S.K., M.H.P., E.Y.S., H.P.) and Internal Medicine (S.Y.K., S.K.H., Y.S.S.), Seoul National University College of Medicine, Seoul, Korea, and the Department of Laboratory Medicine (H.S.K.), Bundang Jesaeng General Hospital, Sungnam, Korea. Address reprint requests to: Dr. Myoung Hee Park, Department of Laboratory Medicine, Seoul National University College of Medicine, 28 Yeongeon-dong, Jongno-gu, Seoul 110-744, Korea; E-mail: [email protected]. Supported by a grant of the Korea Health 21 R&D Project, Ministry of Health & Welfare, Republic of Korea (02-PJ1-PG3-20501-0010). Received July 20, 2005; accepted August 31, 2005. Human Immunology 66, 1074 –1081 (2005) © American Society for Histocompatibility and Immunogenetics, 2005 Published by Elsevier Inc.
0.047) and more advanced lung lesion (29.8% vs. 14.5%, OR ⫽ 2.50, pc ⫽ 0.022). DRB1*0803 showed the strongest association with disease recurrence, especially after curative treatment for the earlier infection (47.4% vs. 14.5%, OR ⫽ 5.31, pc ⫽ 0.00009). DQB1*0601, which is strongly linked to DRB1*0803 in this population showed similar changes in the patients as those of DRB1*0803. It is suggested that DRB1*0803 and DQB1*0601 alleles are associated with disease progression of TB in Koreans, exerting influence on the development of drug resistance, severe disease, and recurrent disease. Human Immunology 66, 1074 –1081 (2005). © American Society for Histocompatibility and Immunogenetics, 2005. Published by Elsevier Inc. KEYWORDS: Tuberculosis; HLA-DR; HLA-DQ
TB
tuberculosis
culosis. An interesting feature of mycobacterial infection is that clinical disease, tuberculosis (TB), develops in only 10% of those infected, which indicates the existence of host factors regulating disease expression [1]. Presence of racial differences in the susceptibility to TB and higher concordance rate for TB among monozygous than among dizygous twins are evidences that the host genetic factors play an important role in the development of TB in humans [1–3]. As one of the host genetic factors, associations of human leukocyte antigen (HLA) with susceptibility to 0198-8859/05/$–see front matter doi:10.1016/j.humimm.2005.08.242
Association of HLA-DR and HLA-DQ With Tuberculosis
1075
TABLE 1 Demographic data of patients with pulmonary tuberculosis
Age (mean ⫾ SD) Sex (men:women) Radiographic lung lesion Minimal Moderately advanced Far advanced Treatmentb Initial treatment Retreatment, total Retreatment, group I Retreatment, group II a b
Total (n ⫽ 160)
Drug sensitive (n ⫽ 79)
Drug resistanta (n ⫽ 81)
44.1 ⫾ 16.2 91:61
46.5 ⫾ 16.6 44:35
41.8 ⫾ 15.6 47:34
29 (18.1%) 65 (40.6%) 66 (41.3%)
26 (32.9%) 35 (44.3%) 18 (22.8%)
3 (3.7%) 30 (37.0%) 48 (59.3%)
106 (66.3%) 54 (33.8%) 38 (23.8%) 16 (10.0%)
64 (81.0%) 15 (19.0%) 15 (19.0%) 0 (0.0%)
42 (51.9%) 39 (48.1%) 23 (28.4%) 16 (19.8%)
Including 67 cases of multidrug resistance. Retreatment: patients with previous histories of curative treatment (group I) or inadequate treatment (group II).
TB have been studied in many ethnic groups. Although the results have been inconsistent among various ethnic groups [4], the association of HLA-DR2 or DRB1*15 (a split of DR2) with TB has been detected most commonly, especially in Asian populations including Asian Indian [5– 8], Indonesian [9], and five ethnic groups in Russia [10]. Association of HLA-DR2 or its splits with TB has also been observed in Polish (DRB1*16) and Mexican (DRB1*15) [11, 12]. Most of the studies on the HLA association with TB have been simply focused on its association with general susceptibility to TB, and association with disease severity or drug resistance has rarely been studied [5, 6]. The prevalence of active pulmonary TB in Koreans is still high (1%, 1,032 per 100,000 in 1995 survey) and drug resistance rate is nearly 10% [13]. Our purpose in this study was to investigate the association of HLA with pulmonary TB in Koreans with special emphasis on its association with drug resistance, disease severity, or disease recurrence. MATERIALS AND METHODS Subjects The study used a case-control design to compare between healthy controls and pulmonary TB patients. A total of 160 Korean patients (91 men and 69 women, age range 20 – 81 years, mean age 44.1 years) with pulmonary TB were enrolled at the Respiratory Clinic of the Seoul National University Hospital (Table 1). We designed a study including a much larger proportion of drugresistant patients than normally encountered (about 20% in our institute), so that about equal number of drug-sensitive and drug-resistant patients were recruited and about one third of the patients were recurrent cases with previous chemotherapy histories for TB.
The diagnosis was confirmed by the cultivation of M. tuberculosis in sputum culture with or without the presence of acid-fast bacilli in sputum smear and by standard clinical and radiologic investigations. No patient in this study had evidence of infection with human immunodeficiency virus, and anti– human immunodeficiency virus antibody was negative in all patients. Drug sensitivity test was performed as a referral test at the Korean National Tuberculosis Association against 11 drugs: isoniazid, rifampicin, ethambutol, pyrazinamide, streptomycin, kanamycin, prothionamide, cycloserine, para-aminosalicylate, ofloxacin, and enviomycin. According to the results of drug sensitivity test, patients whose isolates were all sensitive to tested drugs were classified as the drug-sensitive group and those whose isolates were resistant to one or more drugs tested as the drug-resistant group (including multidrug resistance with resistance to both isoniazid and rifampicin). For chemotherapy, four drug regimens including isoniazid and rifampicin were used for initial treatment and drug-sensitive patients. For drug-resistant patients, individualized regimen according to the drug sensitivity test was used, including at least four sensitive drugs. Pulmonary TB patients were categorized according to drug sensitivity, the extent of lung lesion (minimal, moderately advanced, and far advanced) [14], and the history of treatment (initially treated, retreated after curative treatment, and retreated after inadequate treatment) (Table 1). The study was approved by the institutional review board of the Seoul National University Hospital and informed consent was obtained from all subjects. Two hundred ethnically matched healthy individuals served as controls. They were 200 unrelated parents (101 women, 99 men) of Korean families, previously typed for
1076
HLA class II alleles and analyzed for class II haplotypes by segregation study [15]. HLA Typing Genomic DNA was extracted from peripheral blood by use of QIAamp blood kit (Qiagen, Hilden, Germany). HLA-DRB1 genotyping was carried out in two steps. First, low- to intermediate-resolution typing was performed by polymerase chain reaction (PCR)-sequence specific oligonucleotide hybridization method by using the Dynal RELITM SSO HLA-DRB Typing kit (Dynal Biotech Ltd., Wirral, UK). Then high-resolution DRB1 typing was carried out by group-specific amplification and PCR-single strand conformation polymorphism method [16]. HLA-DQB1 typing was performed using a combination of PCR-restriction fragment length polymorphism and PCR-single strand conformation polymorphism methods, as described previously [17]. Statistical Analysis Phenotype frequencies of the HLA-DRB1 and HLADQB1 alleles and DRB1-DQB1 haplotypes were compared by the 2 test or Fisher’s exact test as appropriate. The level of significance was set at p ⬍ 0.05, and odds ratios (ORs) with 95% confidence intervals (CI) were calculated for those comparisons showing significant p values. Where indicated as corrected p (pc), a Bonferroni correction was applied by multiplying the probability value by the number of comparisons made (i.e., 29 for HLA-DRB1 alleles, 15 for HLA-DQB1 alleles, and 48 for DRB1-DQB1 haplotypes). RESULTS HLA-DRB1 and HLA-DQB1 Associations With Total Group of TB Patients HLA-DRB1 and HLA-DQB1 frequencies in pulmonary TB patients and controls are listed in Tables 2 and 3. When compared with controls, TB patients showed increased frequencies of DRB1*0803 (OR ⫽ 1.97, p ⫽ 0.012, pc ⬎ 0.05), DQB1*0601 (OR ⫽ 2.07, p ⫽ 0.005, pc ⬎ 0.05), and DQB1*0609 (OR ⫽ 2.67, p ⫽ 0.023, pc ⬎ 0.05), and decreased frequency of DRB1*0101 (OR ⫽ 0.45, p ⫽ 0.031, pc ⬎ 0.05). HLA-DRB1 and HLA-DQB1 Associations With Subgroups of TB Patients Among the four HLA alleles listed previously, which showed weak associations with general susceptibility to TB, DRB1*0803 and DQB1*0601 alleles showed stronger associations with subgroups of TB patients in the context of drug resistance, disease severity, and disease recurrence (i.e., previous treatment history for pulmonary TB). Compared with controls (14.5%), the frequency of DRB1*0803 was significantly increased in the drug-
H.S. Kim et al.
resistant group (30.9%, OR ⫽ 2.63, pc ⫽ 0.047), but not in the drug-sensitive group (Tables 2 and 4). Stratification for disease severity revealed that DRB1*0803 was significantly increased in more advanced (moderately or far advanced) group (29.8%, OR ⫽ 2.50, pc ⫽ 0.022), and this increase was also significant when compared with the minimal group (3.4%, OR ⫽ 11.87, p ⫽ 0.003). According to the history of TB treatment, the frequency of DRB1*0803 was significantly increased in the retreated group (37.0%, OR ⫽ 3.47, pc ⫽ 0.006), and this increase was also significant when compared with the initially treated group (18.9%, OR ⫽ 2.53, p ⫽ 0.012). When the retreated group was divided into group I (with previous history of curative treatment) and group II (with previous history of inadequate treatment), significant increase was observed only in the group I (47.4%, OR ⫽ 5.31, pc ⫽ 0.00009), and not in the group II (12.5%). Significant increase of DRB1*0803 frequency in the group I retreated patients was observed in both drug-sensitive (40.0%, OR ⫽ 3.93, p ⫽ 0.02) and drug-resistant (52.2%, OR ⫽ 6.43, pc ⫽ 0.003) patients. DQB1*0601, which is in strong linkage disequilibrium with DRB1*0803 in Koreans [15], showed similar changes in the frequencies as those found for DRB1*0803 (Tables 3 and 5). Compared with controls, significant increase of DQB1*0601 frequency was observed in drug-resistant (OR ⫽ 2.58, pc ⫽ 0.026), more advanced (OR ⫽ 2.48, pc ⫽ 0.01), and retreated (OR ⫽ 3.21, pc ⫽ 0.007) groups of patients. An especially marked increase was observed in the group I retreated patients (OR ⫽ 5.45, pc ⫽ 0.00003). In addition, patients with primary drug resistance (initially treated drug-resistant patients) showed an increased frequency of DQB1*0601 (OR ⫽ 2.44, p ⫽ 0.018). DRB1-DQB1 Haplotype Associations With TB Patients DRB1-DQB1 haplotypes in the controls were analyzed by family-based segregation study, as previously reported [15]. Because the DRB1 and DQB1 alleles are in strong linkage disequilibrium in this population, we could deduce the probable DRB1-DQB1 haplotypes in the patients (data not shown). Significant changes in the frequency of DRB1*0803-DQB1*0601 were observed in TB patients, and the level of significance was slightly lower than that of DRB1*0803 or DQB1*0601. Compared with controls (12.5%), the frequency of DRB1*0803-DQB1*0601 was increased in TB patients (21.3%, OR ⫽ 1.89, p ⫽ 0.026). Significant increase was observed in drug-resistant (25.9%, OR ⫽ 2.45, p ⫽ 0.006), more advanced (25.2%, OR ⫽ 2.36, p ⫽ 0.003), and retreated (31.5%, OR ⫽ 3.22, pc ⫽ 0.041) groups of patients. Especially marked increase was observed in the
Association of HLA-DR and HLA-DQ With Tuberculosis
1077
TABLE 2 HLA-DRB1 frequencies in patients with pulmonary tuberculosis and healthy controls Patients HLA-DRB1
Controls (n ⫽ 200) (%)
Total (n ⫽ 160) (%)
Drug sensitive (n ⫽ 79) (%)
Drug resistant (n ⫽ 81) (%)
DRB1*0101 0301 0401 0403 0404 0405 0406 0407 0410 0701 0802 0803 0901 1001 1101 1201 1202 1301 1302 1307 1401 1402 1403 1405 1406 1407 1501 1502 1602
28 (14.0) 7 (3.5) 0 (0.0) 10 (5.0) 4 (2.0) 32 (16.0) 22 (11.0) 2 (1.0) 4 (2.0) 27 (13.5) 12 (6.0) 29 (14.5) 33 (16.5) 9 (4.5) 13 (6.5) 18 (9.0) 19 (9.5) 6 (3.0) 31 (15.5) 0 (0.0) 11 (5.5) 2 (1.0) 5 (2.5) 15 (7.5) 0 (0.0) 1 (0.5) 34 (17.0) 9 (4.5) 0 (0.0)
11 (6.9)a 6 (3.8) 2 (1.3) 13 (8.1) 5 (3.1) 22 (13.8) 18 (11.3) 0 (0.0) 3 (1.9) 22 (13.8) 3 (1.9) 40 (25.0)b 29 (18.1) 4 (2.5) 19 (11.9) 12 (7.5) 7 (4.4) 4 (2.5) 32 (20.0) 1 (0.6) 6 (3.8) 0 (0.0) 2 (1.3) 9 (5.6) 1 (0.6) 1 (0.6) 21 (13.1) 14 (8.8) 2 (1.3)
5 (6.3) 5 (6.3) 1 (1.3) 8 (10.1) 1 (1.3) 10 (12.7) 13 (16.5) 0 (0.0) 1 (1.3) 11 (13.9) 2 (2.5) 15 (19.0) 16 (20.3) 3 (3.8) 9 (11.4) 6 (7.6) 3 (3.8) 2 (2.5) 15 (19.0) 0 (0.0) 3 (3.8) 0 (0.0) 2 (2.5) 1 (1.3) 0 (0.0) 1 (1.3) 13 (16.5) 6 (7.6) 1 (1.3)
6 (7.4) 1 (1.2) 1 (1.2) 5 (6.2) 4 (4.9) 12 (14.8) 5 (6.2) 0 (0.0) 2 (2.5) 11 (13.6) 1 (1.2) 25 (30.9)c 13 (16.0) 1 (1.2) 10 (12.3) 6 (7.4) 4 (4.9) 2 (2.5) 17 (21.0) 1 (1.2) 3 (3.7) 0 (0.0) 0 (0.0) 8 (9.9) 1 (1.2) 0 (0.0) 8 (9.9) 8 (9.9) 1 (1.2)
p ⫽ 0.031, OR ⫽ 0.45 (95% CI 0.22– 0.94). p ⫽ 0.012, OR ⫽ 1.97 (95% CI 1.15–3.35). c p ⫽ 0.002, pc ⫽ 0.047, OR ⫽ 2.63 (95% CI 1.42– 4.87). a
b
group I retreated patients (39.5%, OR ⫽ 4.57, pc ⫽ 0.002). DISCUSSION In this study, we have found that, although HLADRB1*0803 and HLA-DQB1*0601 alleles were weakly associated with general susceptibility to pulmonary TB in the Korean population, these alleles were strongly associated with drug resistance, disease severity, and disease recurrence, especially in patients with previous history of curative treatment. HLA associations with drug resistance or disease severity of pulmonary TB have only rarely been reported. In Asian Indians, DR2 association with pulmonary TB (relative risk ⫽ 1.8) was reported to be stronger in drug failure group of patients (relative risk ⫽ 3.7) [6]. In South Indians, DR2 association with pulmonary TB (relative risk ⫽ 2.2) was reported to be stronger in sputum smear–positive and far-advanced group of patients (relative risk ⫽ 3.3) [5].
Drug resistance, especially multidrug resistance of M. tuberculosis strains, is an increasing problem worldwide. The World Health Organization has warned about the growing prevalence of drug-resistant TB. The areas worst affected are eastern Europe and central Asia, and as many as 14% of new cases in these regions involve multidrugresistant strains of the bacterium [18]. Widely recognized risk factors for drug resistance include previous treatment for TB, residence in an area endemic for drug resistance, and close contact with a known case of drugresistant TB [19]. In addition to these risk factors, our study suggests that genetic factors of the host are associated with the development of drug resistance in TB patients. In the present study, drug-resistant TB was significantly associated with DRB1*0803 and DQB1*0601, whereas drug-sensitive TB patients did not show significant changes in the frequencies of these alleles. Drug resistance may be primary or acquired: primary drug resistance in patients with no prior TB treatment, and
1078
H.S. Kim et al.
TABLE 3 HLA-DQB1 frequencies in patients with pulmonary tuberculosis and healthy controls Patients HLA-DQB1
Controls (n ⫽ 200) (%)
Total (n ⫽ 160) (%)
Drug sensitive (n ⫽ 79) (%)
Drug resistant (n ⫽ 81) (%)
DQB1*0201 0202 0301 0302 0303 0401 0402 0501 0502 0503 0601 0602 0603 0604 0609
7 (3.5) 25 (12.5) 53 (26.5) 40 (20.0) 41 (20.5) 31 (15.5) 13 (6.5) 37 (18.5) 6 (3.0) 20 (10.0) 31 (15.5) 33 (16.5) 7 (3.5) 25 (12.5) 8 (4.0)
6 (3.8) 20 (12.5) 45 (28.1) 34 (21.3) 32 (20.0) 21 (13.1) 8 (5.0) 19 (11.9) 8 (5.0) 12 (7.5) 44 (27.5)a 20 (12.5) 4 (2.5) 16 (10.0) 16 (10.0)c
5 (6.3) 10 (12.7) 21 (26.6) 21 (26.6) 17 (21.5) 10 (12.7) 2 (2.5) 10 (12.7) 5 (6.3) 3 (3.8) 18 (22.8) 13 (16.5) 2 (2.5) 7 (8.9) 8 (10.1)
1 (1.2) 10 (12.3) 24 (29.6) 13 (16.0) 15 (18.5) 11 (13.6) 6 (7.4) 9 (11.1) 3 (3.7) 9 (11.1) 26 (32.1)b 7 (8.6) 2 (2.5) 9 (11.1) 8 (9.9)
p ⫽ 0.005, OR ⫽ 2.07 (95% CI 1.23–3.47). p ⫽ 0.002, pc ⫽ 0.026, OR ⫽ 2.58 (95% CI 1.41– 4.71). c p ⫽ 0.023, OR ⫽ 2.67 (95% CI 1.11– 6.40). a
b
TABLE 4 Changes of HLA-DRB1*0803 frequencies in various groups of pulmonary tuberculosis patients compared with controls Controls and patients
No. (%)
Controls (n ⫽ 200) Drug sensitivity (n ⫽ 160) Drug sensitive (n ⫽ 79) Drug resistant (n ⫽ 81) Radiographic lung lesion (n ⫽ 160) Minimal (n ⫽ 29) Moderately advanced (n ⫽ 65) Far advanced (n ⫽ 66) Moderately ⫹ far advanced (n ⫽ 131) Treatmentb (n ⫽ 160) Initial treatment (n ⫽ 106) Retreatment, total (n ⫽ 54) Retreatment, group I (n ⫽ 38) Retreatment, group II (n ⫽ 16) Drug sensitive (n ⫽ 79) Initial treatment (n ⫽ 64) Retreatment, group I (n ⫽ 15) Drug resistant (n ⫽ 81) Initial treatment (n ⫽ 42) Retreatment, group I (n ⫽ 23) Retreatment, group II (n ⫽ 16)
29 (14.5)
p Value
pc Value
Odds ratio (95% CI)
15 (19.0) 25 (30.9)
NS 0.002
0.047
2.63 (1.42–4.87)
1 (3.4) 18 (27.7) 21 (31.8) 39 (29.8)a
NS 0.016 0.002 0.0008
NS NS 0.022
2.26 (1.15–4.42) 2.75 (1.44–5.27) 2.50 (1.45–4.30)
20 (18.9) 20 (37.0)c 18 (47.4)d 2 (12.5)
NS 0.0002 0.000003 NS
0.006 0.00009
3.47 (1.76–6.83) 5.31 (2.51–11.23)
NS
3.93 (1.30–11.87)
0.003
6.43 (2.59–15.95)
9 (14.1) 6 (40.0)e
NS 0.020
11 (26.2) 12 (52.2)f 2 (12.5)
NS 0.0001 NS
Abbreviation: NS ⫽ not significant. a Versus minimal: p ⫽ 0.003, pc ⫽ NS, OR ⫽ 11.87 (95% CI 1.56 –90.33). b Retreatment: patients with previous histories of curative treatment (group I) or inadequate treatment (group II). c Versus initial treatment: p ⫽ 0.012, pc ⫽ NS, OR ⫽ 2.53 (95% CI 1.21–5.28). d Versus initial treatment: p ⫽ 0.0006, pc ⫽ 0.018, OR ⫽ 3.87 (95% CI 1.74 – 8.63). e Versus initial treatment: p ⫽ 0.032, pc ⫽ NS, OR ⫽ 4.07 (95% CI 1.17–14.23). f Versus initial treatment: p ⫽ 0.036, pc ⫽ NS, OR ⫽ 3.07 (95% CI 1.06 – 8.95).
Association of HLA-DR and HLA-DQ With Tuberculosis
1079
TABLE 5 Changes of HLA-DQB1*0601 frequencies in various groups of pulmonary tuberculosis patients compared with controls Controls and patients
No. (%)
Controls (n ⫽ 200) Drug sensitivity (n ⫽ 160) Drug sensitive (n ⫽ 79) Drug resistant (n ⫽ 81) Radiographic lung lesion (n ⫽ 160) Minimal (n ⫽ 29) Moderately advanced (n ⫽ 65) Far advanced (n ⫽ 66) Moderately ⫹ far advanced (n ⫽ 131) Treatmentb (n ⫽ 160) Initial treatment (n ⫽ 106) Retreatment, total (n ⫽ 54) Retreatment, group I (n ⫽ 38) Retreatment, group II (n ⫽ 16) Drug sensitive (n ⫽ 79) Initial treatment (n ⫽ 64) Retreatment, group I (n ⫽ 15) Drug resistant (n ⫽ 81) Initial treatment (n ⫽ 42) Retreatment, group I (n ⫽ 23) Retreatment, group II (n ⫽ 16)
31 (15.5)
p Value
pc Value
Odds ratio (95% CI)
18 (22.8) 26 (32.1)
NS 0.002
0.026
2.58 (1.41–4.71)
3 (10.3) 20 (30.8) 21 (31.8) 41 (31.3)a
NS 0.007 0.004 0.0007
NS NS 0.010
2.43 (1.26–4.65) 2.54 (1.34–4.85) 2.48 (1.46–4.23)
24 (22.6) 20 (37.0) 19 (50.0)c 1 (6.3)
NS 0.0005 0.000002 NS
0.007 0.00003
3.21 (1.64–6.28) 5.45 (2.60–11.45)
11 (17.2) 7 (46.7)d
NS 0.007
NS
4.77 (1.61–14.11)
13 (31.0) 12 (52.2)e 1 (6.3)
0.018 0.0002 NS
NS 0.003
2.44 (1.15–5.22) 5.95 (2.41–14.68)
Abbreviation: NS ⫽ not significant. a Versus minimal: p ⫽ 0.022, pc ⫽ NS, OR ⫽ 3.95 (95% CI 1.13–13.79). b Retreatment: patients with previous histories of curative treatment (group I) or inadequate treatment (group II). c Versus initial treatment: p ⫽ 0.002, pc ⫽ 0.024, OR ⫽ 3.42 (95% CI 1.56 –7.47). d Versus initial treatment: p ⫽ 0.035, pc ⫽ NS, OR ⫽ 4.22 (95% CI 1.26 –14.06). e Versus initial treatment: p ⫽ 0.021, pc ⫽ NS, OR ⫽ 3.27 (95% CI 1.17–9.18).
acquired drug resistance in patients previously treated for TB. HLA class II alleles were strongly associated with acquired drug resistance in this study, especially in patients with previous history of curative treatment for TB (for DRB1*0803, OR ⫽ 6.43, pc ⫽ 0.003; for DQB1*0601, OR ⫽ 5.95, pc ⫽ 0.003). Our finding is further supported by a recent report in Asian Indians that the presence of HLA-DRB1*14 was an independent predictor of multidrug-resistant TB (OR ⫽ 8.2) [20]. It can be assumed that patients carrying particular HLA class II alleles are incapable of mounting adequate immune response to tubercle bacilli and thus are prone to develop drug resistance. The most interesting finding in this study was that DRB1*0803 and DQB1*0601 alleles were strongly associated with recurrent TB. Among patients with recurrent TB, the association was significant only in patients with previous history of curative treatment (for DRB1*0803, OR ⫽ 5.31, pc ⫽ 0.00009; for DQB1*0601, OR ⫽ 5.45, pc ⫽ 0.00003), but not in those with previous history of inadequate treatment. It is conceivable that the recurrence of TB after inadequate or incomplete treatment would not be related to the host genetic factors. Recurrence of TB after curative treatment could arise from reactivation of the earlier infection
or recent reinfection [21, 22]. The recurrence of TB in patients with previous treatment history is generally assumed to be due to inadequate treatment for the earlier infection, mostly because of poor compliance with drug therapy. However, in this study, we have found that recurrence of TB even after curative treatment is strongly associated with HLA class II alleles. To our knowledge, no other study has clearly shown the association of HLA or other genetic factors with recurrent TB in humans after curative treatment for the earlier infection. Because of the very strong linkage disequilibrium between DRB1*0803 and DQB1*0601 alleles in the Korean population [15], we could not determine which allele is primarily associated with TB susceptibility. In addition, DRB1*0803-DQB1*0601 haplotype showed similar changes in TB patients as those of DRB1*0803 and DQB1*0601. DRB1*0803 is present at relatively high frequency in East Asian populations including Korean, Japanese, and Chinese, but is rare in white or black populations [23], and its association with TB susceptibility has not been reported in other populations. Weak association of DR8 with TB has only rarely been reported in South Indians (OR ⫽ 5.99, p ⫽ 0.035) and Hong Kong Chinese (relative risk ⫽ 2.26, p ⫽ 0.028) [8, 24]. In contrast, DR2 or DRB1*15 (most commonly
1080
DRB1*1501) has been reported in many ethnic groups to be the main HLA allele associated with TB susceptibility [5–10, 12]. However, despite quite a high frequency of DR2 or DRB1*1501 in the Korean population, this allele did not show increased frequency in TB patients. Thus DRB1*0803 might not be primarily associated with TB susceptibility and the observed association might be secondary to the DQB1*0601 (a split of DQ1) association. This is partially supported by other reports in Asian Indians that DRB1*1501-DQB1*0601 haplotype was associated with TB susceptibility [8, 25]. DQB1*05, another split of DQ1, has also been reported to be associated with TB susceptibility in many ethnic groups: DQB1*0501 in Mexicans (together with DRB1*1501), DQB1*0502 in Asian Indians (together with DRB1*1501) and in Thais, DQB1*0503 in Cambodians, and DQB1*05 in Poles [12, 25–28]. It is possible that DQ1 and its splits DQB1*05 and DQB1*06 alleles are primarily associated with TB susceptibility. However, there is still another possibility that other major histocompatibility complex genes closely linked to the HLA-DRB1 and -DQB1 genes are primarily associated with TB susceptibility. The specific immune response to M. tuberculosis infection mainly depends on the activation of various T-cell subsets that produce cytokines, mainly interferon-␥ and tumor necrosis factor-␣, which orchestrate appropriate defense mechanism [29]. CD4 T cells, CD8 T cells, ␥␦ T cells, and CD1-restricted T cells are participating in the immune response to M. tuberculosis, among which CD4 T cells play the major role in containing TB at all stages of the disease [29, 30]. CD4 T cells are recognizing antigenic peptides in the context of HLA class II molecules, and thus particular HLA class II alleles might not be able to present appropriate antigenic peptides of tubercle bacilli to CD4 T cells. In the present study of Koreans, gene products of DRB1*0803, DQB1*0601, or other closely linked genes are suggested to have some defects in presenting appropriate antigenic peptides of tubercle bacilli. Lack of an effective immune response in carriers of DRB1*0803 or DQB1*0601 seems to be a plausible and unifying explanation for the increased proportion of severe disease, disease recurrence, and drug resistance among patients carrying these alleles. In summary, it is suggested that DRB1*0803 and DQB1*0601 alleles are associated with TB susceptibility in Koreans. DRB1*0803 and DQB1*0601 were only weakly associated with general susceptibility to TB. However, individuals possessing these alleles, after being infected, were strongly associated with the development of drug resistance, more advanced lung lesion, and disease recurrence even after curative treatment for the earlier infection. These findings are probably interrelated and the patients possessing these alleles might not be
H.S. Kim et al.
able to mount an effective immune response against M. tuberculosis.
REFERENCES 1. Bellamy R: Genetic susceptibility to tuberculosis in human populations. Thorax 53:588, 1998. 2. Stead WW, Senner JW, Reddick WT, Lofgren JP: Racial differences in susceptibility to infection by Mycobacterium tuberculosis. N Engl J Med 322:422, 1990. 3. Comstock GW: Tuberculosis in twins: a re-analysis of the Prophit survey. Am Rev Resp Dis 117:621, 1978. 4. Meyer CG, May J, Stark K: Human leukocyte antigens in tuberculosis and leprosy. Trends Microbiol 6:148, 1998. 5. Brahmajothi V, Pitchappan RM, Kakkanaiah VN, Sashidhar M, Rajaram K, Ramu S, Palanimurugan K, Paramasivan CN, Prabhakar R: Association of pulmonary tuberculosis and HLA in South India. Tubercle 72:123, 1991. 6. Rajalingam R, Mehra NK, Jain RC, Myneedu VP, Pande JN: Polymerase chain reaction-based sequence-specific oligonucleotide hybridization analysis of HLA class II antigens in pulmonary tuberculosis: relevance to chemotherapy and disease severity. J Infect Dis 173:669, 1996. 7. Selvaraj P, Uma H, Reetha AM, Kurian SM, Xavier T, Prabhakar R, Narayanan PR: HLA antigen profile in pulmonary tuberculosis patients & their spouses.Indian J Med Res 107:155, 1998. 8. Ravikumar M, Dheenadhayalan V, Rajaram K, Lakshmi SS, Kumaran PP, Paramasivan CN, Balakrishnan K, Pitchappan RM: Associations of HLA-DRB1, DQB1 and DPB1 alleles with pulmonary tuberculosis in south India. Tuber Lung Dis 79:309, 1999. 9. Bothamley GH, Beck JS, Schreuder GM, D’Amaro J, de Vries RR, Kardjito T, Ivanyi J: Association of tuberculosis and M. tuberculosis-specific antibody levels with HLA. J Infect Dis 159:549, 1989. 10. Khomenko AG, Litvinov VI, Chukanova VP, Pospelov LE: Tuberculosis in patients with various HLA phenotypes. Tubercle 71:187, 1990. 11. Dubaniewicz A, Lewko B, Moszkowska G, Zamorska B, Stepinski J: Molecular subtypes of the HLA-DR antigens in pulmonary tuberculosis. Int J Infect Dis 4:129, 2000. 12. Terán-Escandón D, Terán-Oritz L, Camarena-Olvera A, González-Avila G, Vaca-Marín MA, Granados J, Selman M: Human leukocyte antigen-associated susceptibility to pulmonary tuberculosis: molecular analysis of class II alleles by DNA amplification and oligonucleotide hybridization in Mexican patients. Chest 115:428, 1999. 13. Hong YP, Kim SJ, Lew WJ, Lee EK, Han YC: The seventh nationwide tuberculosis prevalence survey in Korea, 1995. Int J Tuberc Lung Dis 2:27, 1998. 14. American Thoracic Society: Diagnostic standards and Classification of Tuberculosis. 12th ed. New York, Na-
Association of HLA-DR and HLA-DQ With Tuberculosis
15.
16.
17.
18.
19.
20.
21.
22.
tional Tuberculosis and Respiratory Disease Association, 1969. Song EY, Park MH, Kang SJ, Park HJ, Kim BC, Tokunaga K, Akaza T, Juji T: HLA class II allele and haplotype frequencies in Koreans based on 107 families. Tissue Antigens 59:475, 2002. Bannai M, Tokunaga K, Lin L, Kuwata S, Mazda T, Amaki T, Fujisawa K, Juji T: Discrimination of human HLA-DRB1 alleles by PCR-SSCP (single-strand conformation polymorphism) method. Eur J Immunogenet 21:1, 1994. Park MH, Whang DH, Kang SJ: High resolution HLADQB1 typing by combination of PCR-RFLP and PCRSSCP. Hum Immunol 60:901, 1999. World Health Organization: Anti-tuberculosis drug resistance in the world: third global report. Geneva, WHO, 2004. Harkin TJ, Condos R: Management of multidrugresistant tuberculosis. In Rom WN, Garay SM (eds): Tuberculosis. 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. Sharma SK, Turaga KK, Balamurugan A, Saha PK, Pandey RM, Jain NK, Katoch VM, Mehra NK: Clinical and genetic risk factors for the development of multi-drug resistant tuberculosis in non-HIV infected patients at a tertiary care center in India: a case-control study. Infect Genet Evol 3:183, 2003. Gomez JE, McKinney JD: Persistence and drug resistance. In Rom WN, Garay SM (eds): Tuberculosis. 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. van Rie A, Warren R, Richardson M, Victor TC, Gie RP, Enarson DA, Beyers N, van Helden PD: Exogenous rein-
1081
23.
24.
25.
26.
27.
28.
29.
30.
fection as a cause of recurrent tuberculosis after curative treatment. N Engl J Med 341:1174, 1999. Imanishi T, Akaza T, Kimura A, Tokunaga K, Gojobori T: Allele and haplotype frequencies for HLA and complement loci in various ethnic groups. In Tsuji K, Aizawa M, Sasazuki T (eds): HLA 1991. Vol I. New York, Oxford University Press, 1992. Hawkins BR, Higgins DA, Chan SL, Lowrie DB, Mitchison DA, Girling DJ: HLA typing in the Hong Kong Chest Service/British Medical Research Council study of factors associated with the breakdown to active tuberculosis of inactive pulmonary lesions. Am Rev Respir Dis 138:1616, 1988. Mehra NK, Rajalingam R, Mitra DK, Taneja V, Giphart MJ: Variants of HLA-DR2/DR51 group haplotypes and susceptibility to tuberculoid leprosy and pulmonary tuberculosis in Asian Indians. Int J Lepr 63:241, 1995. Vejbaesya S, Chierakul N, Luangtrakool K, Srinak D, Stephens HA: Association of HLA class II alleles with pulmonary tuberculosis in Thais. Eur J Immunogenet 29:431, 2002. Goldfeld AE, Delgado JC, Thim S, Bozon MV, Uglialoro AM, Turbay D, Cohen C, Yunis EJ: Association of an HLA-DQ allele with clinical tuberculosis. JAMA 279: 226, 1998. Dubaniewicz A, Moszkowska G, Szczerkowska Z, Hoppe A: Analysis of DQB1 allele frequencies in pulmonary tuberculosis: preliminary report. Thorax 58:890, 2003. Ulrichs T, Kaufmann SH: Cell-mediated immune response. In Rom WN, Garay SM (eds): Tuberculosis. 2nd ed. Philadelphia, Lippincott Williams & Wilkins, 2004. Kaufmann SH: Protection against tuberculosis: cytokines, T cells, and macrophages. Ann Rheum Dis 61(Suppl 2):ii54, 2002.