Human Leukocyte Antigen-G (HLA-G) 14-bp deletion polymorphism is associated with decreased risk of pulmonary fibrosis in Indonesian Javanese patients with multidrug-resistant tuberculosis

We enrolled 57 Indonesian Javanese patients with MDR-TB patients in this cohort study, TB-infected patients showing resistance to at least isoniazid and rifampisin, at Dr. Moewardi General Hospital, Central Java province, Indonesia, from May 2012 through January 2013. The patients came from various districts in Java island, mostly from Central Java province. There was no consecutive case observed, and none of the subjects were from the same family. Each patient was administered appropriate drugs (taking into account the patient’s drug resistance status) and was monitored from the time of MDR-TB diagnosis until sputum conversion (when a bacterial microscopic test has a negative result) or until January 2014. As control groups, 34 healthy individuals without TB and 69 individuals with a history of TB infection, although not MDR-TB, who were cured during the study period were enrolled to assess a possible association between HLA-G polymorphism status and susceptibility to either TB or MDR-TB. Like the group of MDR-TB patients, the control groups consisted of Javanese people, predominantly from Central Java province.

Blood samples were obtained from all patients to examine their CD4⁺ T cell counts and percentages, hematologic profiles (hemoglobin, hematocrit, erythrocytes, leukocytes, and thrombocytes), sodium (Na⁺), potassium (K⁺), interferon gamma (IFNg), and interleukin-10 (IL-10) levels. Patient sputum samples were evaluated initially at MDR-TB diagnosis and during routine follow-up.

Approval was obtained from the institutional ethical committee review boards of the Faculty of Medicine of Sebelas Maret University and Dr. Moewardi General Hospital, Central Java province, Indonesia (No: E.C. 129/VII/2009). Written informed consent was obtained from all study participants. All procedures were conducted according to the principles of the Declaration of Helsinki.

Genomic DNA from each patient was isolated from peripheral blood using a High Pure PCR Template Preparation Kit (Roche Diagnostics, Mannheim, Germany). For HLA-G 14-bp in/del polymorphism analysis, genomic samples were genotyped by polymerase chain reaction (PCR) using a FastStart HiFi PCR System dNTPack (Roche) with primer pairs that have been previously described [27]. The PCR products were electrophoresed in 3% agarose gels for 30 minutes at 100V and stained with ethidium bromide (EtBr). For the 14-bp insertion (+14bp), a PCR product of 214bp was obtained, while the deletion (–14bp) resulted in a PCR product of 200 bp. All samples were tested at least twice.

The chi-square (χ²) test was used to assess whether the HLA-G 14-bp in/del genotype distributions in the study population were in Hardy-Weinberg equilibrium. The χ² test, Fisher’s exact test, and the Kolmogorov-Smirnov test were used to analyze categorical variables. Logistic regression was performed to reveal associations between polymorphism status and TB/MDR-TB outcome and susceptibility. All data analyses used a 95% confidence interval (CI), and a two-tailed P < 0.05 was considered statistically significant. All statistical analyses were performed with SPSS Statistics for Windows software, version 21 (IBM Corp, Armonk, NY, USA).

The group of MDR-TB patients consisted of 26 (46%) men and 31 (54%) women aged 19-62 years (mean: 41.1 + 10.708). All patients suffered from intrapulmonary TB that was resistant to rifampicin and isoniazid. Other patients were also resistant to ethambutol (46%; 26/57) and streptomycin (54%; 31/57). In one patient, the resistance status to ethambutol and streptomycin was unknown. All patient bacterial cultures indicated the presence of M. tuberculosis. Data for patient body mass index (BMI) were obtained for 98% (56/57) of patients, with a mean BMI of 16.5 ± 2.82 kg/m². Of these, 41/56 (73%) patients were underweight (BMI < 18.5 kg/m²), whereas 15/56 (27%) patients were normal weight (BMI = 18.5-24.9 kg/m²). Sixteen (28%) patients smoked cigarettes, while other 41/57 (72%) patients did not. Patient blood samples were received from May 2012 until January 2014 and were sent for hematologic testing. Meanwhile, the control groups included 34 (33%) healthy participants without clinical symptom of TB and 69 (67%) participants with a history of TB, but not MDR-TB, who were cured during the study period. There was no difference between the group of MDR-TB patients and controls with respect to sex (P = 0.90), age (P = 0.33), and a history of smoking cigarettes (P = 0.89).

Evaluation of patient blood profiles was successfully conducted for 93% (53/57) of MDR-TB patients (Table 1). For the most part, subjects possessed normal CD4⁺ T cell counts and percentages (74% [42/57] had CD4⁺ T cell counts >500 cells/μL; 67% [38/57] had CD4⁺T cell percentages &#8805;29%). CD4⁺ T cell counts of 200-500 and <200 cells/μL were observed in 25% (14/57) and 2% (1/57) of patients, respectively, while 33% (19/57) of patients had CD4⁺ T cell percentages of 14%-28%. Among the total, 60% (34/57) and 98% (56/57) of patients demonstrated normal IFNγ and IL-10 levels (<10pg/ mL), respectively, while 40% (23/57) and 2% (1/57) of patients had excessive IFNγ and IL-10 levels (>10pg/mL), respectively.

Data for the thorax radiographical examination were obtained for 95% (54/57) of MDR-TB patients. Lung cavity infection was observed in 29/54 (54%)

patients, infiltration in 48/54 (89%), nodules in 1/54 (2%), fibrosis in 15/54 (28%), pleural effusion in 3/54 (6%), atelectasis in 5/54 (9%), and consolidation in 1/54 (2%). None of the patients suffered fibrothorax. During follow-up of MDR-TB patients, sputum conversion was observed mostly after two months of treatment (19/57, 33%). Other patients showed sputum conversion after three months (13/57, 23%), one month (11/57, 19%), four months (7/57, 12%), and five months (1/57, 2%) of treatment. Six (11%) patients died before sputum conversion.

In the total study population (n = 160; 57 MDR-TB patients, 69 controls with a history of TB infection, and 34 healthy controls), the D/D genotype was identified in 71 (44%) patients, I/D genotype in 68 (43%), and I/I genotype in 21 (13%) patients. The frequencies of the deletion (D) and insertion (I) alleles were 0.66 and 0.34, respectively. These frequencies complied with Hardy-Weinberg equilibrium (P > 0.05).

The number of MDR-TB patients with the D/D genotype was 19 (33%), I/D genotype was 30 (53%), and I/I genotype was 8 (14%), which resulted in D and I allele frequencies of 0.60 and 0.40, respectively. Among controls with a history of TB infection and healthy controls, the frequency of the D/D genotype was 30/69 (44%) vs 22/34 (65%), I/D genotype was 26/69 (38%) vs 12/34 (35%), and I/I genotype was 13/69 (19%) vs 0/34 (0%), respectively, with D and I allele frequencies of 0.62 vs 0.82 and 0.38 vs 0.18, respectively.

There was a difference in the proportions of the HLA-G 14-bp genotype between controls with a history of TB infection and healthy controls (P = 0.014); individuals carrying the D/D genotype less frequently had a history of TB infection (OR 0.4, 95% CI 0.179-0.981, P = 0.043), whereas the proportion of the HLA-G 14-bp genotype between controls with a TB history and the MDR-TB group was not different (P = 0.243).

The frequency of the HLA-G genotype was not different in terms of hematocrit, erythrocytes, leukocytes, thrombocytes, Na⁺, K⁺, or IFNγ levels (Table 1). Similarly, the HLA-G genotypic status proportion did not differ when taking into consideration patient drug resistance status (Table 2) and time to sputum conversion (P > 0.05).

The numbers of cured and uncured patients did not differ based on HLA-G genotype status (P > 0.05). Interestingly, the distribution frequency of pulmonary fibrosis was different when considering patient HLA-G genotype (P = 0.02) and was associated with the HLA-G 14-bp deletion allele (adjusted OR 0.1, 95% CI 0.014-0.639, P = 0.016). Moreover, patients with pulmonary fibrosis were less likely to be cured (adjusted OR 0.1, 95% CI 0.014-0.514, P = 0.007).