Association of HLA Class II Alleles with Disease Severity and Treatment Response in Iranian Patients with Myasthenia Gravis

Abstract

Myasthenia gravis is an autoimmune neuromuscular disease with a multifactorial etiology. A major part of the genetic susceptibility belongs to the HLA encoding genes. In this study, we investigated the role of HLA class II polymorphism in disease severity, and treatment response. In our 146 patients, 15 DRB1, 7 DQA1, and 9 DQB1 alleles, and 19 haplotypes were found. Adjusted p-values did not show any significant associations between these loci, disease severity and treatment outcome. Further studies in different populations with a larger number of patients are needed to determine the exact contribution of HLA class II alleles to MG prognosis.

INTRODUCTION

Myasthenia gravis is an autoimmune neuromuscular disease with a multifactorial etiologic background where complex interactions between genetic and environmental factors play role in its pathogenesis [1].

The global trend towards precision medicine has highlighted the role of immunogenetics in MG [1, 2]. Individualized approaches in MG help to recognize treatment-refractory cases or those prone to a more severe course of the disease, minimizing morbidity and mortality of MG [2].

A major part of the genetic contribution in MG belongs to the HLA encoding genes which have been the most investigated region in the whole vertebrate genome so far [3]. Different alleles and haplotypes of HLA have been associated with MG and its subgroups, some imparting susceptibility, and others conferring protective effects [3]. However, the role of HLA polymorphism in disease severity and treatment response is yet to be defined. In this study, we tried to find an answer to this question in a population of Iranian patients over a follow-up period of more than five years.

MATERIALS AND METHODS

In this retrospective cohort, 160 MG patients who had participated in the previous study of HLA typing [4] were enrolled. Patients with poor compliance to treatment, no regular follow-up visits, and incomplete medical records were excluded and ultimately 146 patients entered analysis.

Myasthenia Gravis Foundation of America (MGFA) clinical classification [5] was used to assess maximum disease severity throughout the disease. Patients were grouped as follows: Class I: Ocular; Class II: Mild weakness; Class III: Moderate weakness; Class IV: Severe weakness, and V: Defined by intubation.

Patients were categorized in different treatment groups defined as Acetylcholine esterase inhibitor alone, Prednisolone, 1 Immunosuppressive and more than one Immunosuppressive. The immunosuppressive response was evaluated in patients who tolerated optimal dose of immunosuppressives (Azathioprine: 2-3 mg/kg/day, Cyclosporine:4–6 mg/kg/day mg, Mycophenolate mofetil 2–2.5 g/day, and Rituximab: 1-2 g starting dose followed by 0.5–1 g after 6 months) for long enough period.

MGFA post-intervention status (PIS) [5] was used as the overall outcome indicator. To have more patients for each type of allele and haplotype, we categorized PIS as good (Complete Stable Remission, Pharmacologic Remission, Minimal Manifestation, and Improved) and poor (Unchanged, Worse, Exacerbation, and Died of MG). Those with poor MGFA PIS were considered as a failure to treat.

Statistical analysis

Statistical analysis was performed using SPSS software (version 16, IBM Corporation, New York, USA). Allele and haplotype frequencies were compared using Chi-square and Fisher’s exact test. Observed P-values were multiplied by the number of identified alleles and haplotypes to make corrected P-values (Pc). Corrected P-values less than 0.05 were considered statistically significant.

RESULTS

In this study 146 patients with MG were enrolled. Patients (88 females and 58 males) were 8 to 82 years of age (45±15) and were followed over 6 to 37 years (11±5.5). Most patients (136 patients, 96%) had generalized MG. Regarding the antibody status, 97 patients (67%) were positive for the acetylcholine receptor antibody, 22 patients (15%) for the muscle-specific tyrosine kinase (MuSK) antibody, and the remainder being double seronegative. Thymectomy was performed in 51 patients and pathology of thymoma ensued in 12 cases.

The majority of the patients (121 patients, 83%) were in good PIS (5.5%Complete Stable Remission, 36.3%Pharmacologic Remission, 41.1%Minimal Manifestation, and 11%Improved). Only 9 patients were in poor PIS.

General distributions in DRB1, DQA1, and DQB1 loci of HLA class II in MG patients revealed 15 different alleles of DRB1, 7 alleles of DQA1, 9 alleles of DQB1, and 19 haplotypes. Adjusted p-values did not show any significant associations between these loci, disease severity, different treatment options, and overall outcome (PIS). Distribution of HLA class II alleles and haplotypes based on the treatment groups and PIS is provided as a supplementary file.

DISCUSSION

Inter-individual differences in response to treatments are an area of concern in everyday practice. Many factors are contributory, among all genetic polymorphisms account for 20–30%of the variability. HLA alleles have been investigated as one of these pharmacogenomic biomarkers of drug response. HLA typing has significantly influenced the prescription of some specific drugs. For instance, testing for HLA-B*5701 is now recommended by the FDA before abacavir administration in HIV patients [6].

HLA is regarded as the strongest genetic contributor in neurologic conditions. HLA polymorphisms have been attributed to the pathogenesis of many neurologic diseases including those with well-characterized immunologic basis and those with a more degenerative nature [3]. Given the role of HLA in immunologic reactions, HLA gene polymorphisms are potentially implicated in the response to immunomodulatory treatments, as shown in rheumatoid arthritis [7]. Among neurologic conditions, multiple sclerosis has been the subject of most studies of this field. Certain alleles of HLA class II have been associated with response to relatively specific treatments (e.g. interferons and glatiramer acetate) as well as common immunosuppressive agents (e.g. azathioprine) in MS patients [8, 9].

This study is the first to investigate the possible role of HLA class II alleles in predicting disease severity and outcome in MG patients. Our previous study on the same population of patients revealed certain HLA class II alleles to be associated with anti-MuSK positive and thymomatous MG [4]. However, the present study did not show any significant relationship between HLA class II subtypes and MG severity. Likewise, MG patients with distinct alleles and haplotypes of HLA were equally responsive to treatment options.

In our study, owing to the many different HLA subtypes, a small number of patients were present for each type of allele and haplotype (as shown in the supplementary file). This might partly justify not reaching statistically significant results in our analysis. Moreover, HLA genes are extremely polymorphic with varying patterns of linkage disequilibrium [3]. Altogether, HLA gene polymorphism can still remain a hot point in immunogenetic studies with a larger number of patients with distinct genetic properties.