Multiple Sclerosis Resource Centre

Welcome to the Multiple Sclerosis Resource Centre. This website is intended for international healthcare professionals with an interest in Multiple Sclerosis. By clicking the link below you are declaring and confirming that you are a healthcare professional

You are here

Pharmacogenomics of interferon beta and glatiramer acetate response: A review of the literature

Autoimmunity Reviews, 2, 13, pages 178 - 186


Multiple sclerosis (MS) is one of the most common inflammatory and degenerative autoimmune diseases of the central nervous system with considerable heterogeneity in all aspects, including response to therapy. A number of disease modifying drugs, including traditional first line agents such as, interferon-beta (IFN-β) and glatiramer acetate (GA) are available for disease management. However, a considerable number of patients fail to achieve adequate response at therapeutic doses of IFN-β or GA. This variability in response to treatment has prompted the search for prognostic markers in order to personalize and optimize therapy so as to treat MS more efficiently. This review will summarize the existing literature examining the pharmacogenomics of IFN-β and GA response in MS patients.

Abbreviations: APC - antigen-presenting cell, BBB - blood–brain barrier, BDNF - brain-derived neurotrophic factor, CNS - central nervous system, DMD - disease modifying drug, EAE - experimental autoimmune encephalomyelitis, EDSS - expanded disability status scale, GA - glatiramer acetate, GWAS - genome wide association study, HCV - hepatitis C virus, HHV - human herpes virus, HLA - human leukocyte antigen, IFN - interferon, IFN-β - interferon-beta, ISG - interferon-stimulated gene, ISRE - interferon-stimulated response element, MHC - major histocompatibility complex, MRI - magnetic resonance imaging, MS - multiple sclerosis, NAb - neutralizing antibody, NR - non-responder, NT - neurotrophin, R - responder, RRMS - relapsing-remitting multiple sclerosis, SNP - single nucleotide polymorphism.

Keywords: Multiple sclerosis, Genetics, Pharmacogenomics, Interferon beta, Glatiramer acetate, Response.

1. Introduction

Multiple sclerosis (MS) is an inflammatory and degenerative disease characterised by autoimmune demyelination of the central nervous system (CNS). It is one of the most common CNS diseases in young adults affecting approximately 2.5 million people worldwide, and as with other autoimmune diseases MS is more prevalent in females than in males. The exact aetiology of MS is unknown, but it is believed to occur as a result of complex interactions between genetic and environmental factors [1] . While the human leukocyte antigen (HLA) class II region on chromosome 6 has been implicated as a MS susceptibility locus for many years, advancements in genomic technologies and data analysis in recent years have greatly facilitated the discovery of many other genetic risk factors for MS [2] . In the latest susceptibility study, the genomes of 14,498 MS patients and 24,091 healthy controls were screened for 161,311 autosomal variants, identifying 48 novel and confirming 49 previously known susceptibility loci for MS. This and other previous susceptibility studies have confirmed the role of immune system processes in MS and indicated a common underlying pathology with other autoimmune diseases [3] . Proposed environmental risk factors for MS, including low serum vitamin D, Epstein–Barr Virus infection, and smoking, also have the potential to affect immune responses [4] .

In the absence of a cure, current management of MS involves the administration of disease modifying drugs (DMDs) which are most effective if administered in the early stages of the disease to reduce the number of attacks and prevent the accumulation of disability by slowing disease progression. Traditional first line DMDs include interferon-beta (IFN-β) and glatiramer acetate (GA) which have been shown to reduce relapses, and delay progression of disability with minimal side effects [5] . However approximately 20–50% of patients do not respond well to these treatments. Typically clinical response criteria are based on a reduction in the frequency of relapses and/or prevention of disease progression as measured by the expanded disability status scale (EDSS) score. Magnetic resonance imaging (MRI) measures of response are also, although less frequently, incorporated into the criteria. Importantly, clinical evaluation requires a 1 to 2 year follow-up period. As such, non-responsive patients may face a significant delay in the administration of an effective alternative treatment with unnecessary exposure to side effects and ultimately suboptimal care with inefficient use of healthcare resources and associated significant economic burden [6] . With this in mind and given the growing pharmaceutical armoury in the fight against MS, there is a pressing need for the identification of sensitive and specific biomarkers that can be used to predict the risk versus benefit profile of these therapies for any given patient. Identification of such biomarkers will not only aid in the identification of the “right drug for the right patient” but may also offer insights into the pathogenesis of this complex disease.

The genetic composition of an individual may impact on their responsiveness to a medication. For example, a single nucleotide polymorphism (SNP) inIL28Bwas shown to affect response to IFN-alpha and ribavirin in patients with chronic hepatitis C viral (HCV) infection[7], [8], and [9]. As such it is tempting to propose that genetic polymorphisms may also determine response to IFN-β and GA in MS patients. In particular, it has been hypothesised that genetic variations in the biological pathways mediating the downstream effects of these agents may account for the differences in responsiveness observed clinically. Since MS is a complex polygenic disease and IFN-β and GA are pleiotropic agents, it is likely that a number of polymorphisms may determine the response phenotype. A number of studies have aimed to identify genetic variations responsible for the heterogeneous responses to these therapies, with particular focus on IFN-β. Many of these studies have been conducted using the traditional candidate gene approach, although in more recent years a number of genome wide association studies (GWAS) have also been published. This review summarizes the evidence from studies examining genetic polymorphisms as modifiers of clinical response to IFN-β and GA in MS patients.

2. Interferon beta

Interferons (IFNs) are a family of pleiotropic cytokines produced by cells in response to viral infections. These cytokines exhibit anti-viral, immunomodulatory and antitumor properties by regulating expression of hundreds of genes involved in crucial biological processes such as cell cycle progression, cell proliferation and apoptosis. While the exact mode of action of IFN-β in MS is likely to be complex and is not yet fully understood, it does appear to increase production of anti-inflammatory agents such as IL-10, decrease production of pro-inflammatory cytokines such as IL-17 and osteopontin, inhibit immune cell trafficking across the blood–brain barrier (BBB), and stimulate the production of trophic factors such as nerve growth factors. Additionally, it may also increase the production of CD56 bright natural killer cells, and promote apoptosis[10] and [11].

A number of IFN-β products, Betaferon® (IFN-β-1b), Avonex® (IFN-β-1a) and Rebif® (IFN-β-1a) are currently licensed for treating relapsing-remitting MS (RRMS), the most common form of MS. Several clinical trials have shown that these IFN-β products reduce the relapse rate, slow the appearance of new and enhancing lesions on MRI, and delay progression of disability[12] and [13].

However up to half of IFN-β treated patients continue to have exacerbations and increasing disability. A proportion of this non-responsiveness is likely explained by the development of IFN-β neutralizing antibodies (NAbs). These develop in up to a third of patients depending on the IFN-β product administered. Reduced bioactivity and thus clinical response seem to be particularly relevant for patients with persistently high titres [14] . However NAbs do not develop until 6–24 months after treatment onset and can disappear over time. Thus they are not necessarily an early predictor of treatment response. Additionally not all non-responsive patients develop NAbs, in fact only small proportions may be affected [15] . This suggests that NAbs contribute to reduced clinical efficacy of IFN-β but are not the only reason for non-responsiveness. Results of several candidate gene studies and a GWAS have identified genetic markers in the HLA locus (HLA-DRB1*04:01, *04:08)[16], [17], and [18]and non-HLA locus (TLR6- rs5743810) in male MS patients [19] , and an intergenic SNP, rs4961252 [17] , which influence development of NAbs in MS patients. While incorporation of NAb status in IFN-β pharmacogenomics studies is desirable, most studies to date have not included this information.

3. Candidate gene studies

The majority of IFN-β pharmacogenomic studies have examined “candidate” gene(s) relevant to MS or the pharmacological action of IFN-β, as potential modifiers of response. A summary of the findings from these studies is provided below. Additional details on the studies discussed, including the number of patients, their ethnicity, criteria used to classify patients as either responders (R) or non-responders (NR), and genes/SNPs examined, are summarized in Supplementary Table A.1 .

3.1. Immune cell activation and signalling

3.1.1. Human leukocyte antigen genes

Major histocompatibility complex (MHC) molecules are polymorphic glycoproteins present on cell membranes which play a critical role in adaptive immune responses. Class I MHC molecules are present in all nucleated cells of vertebrate species and are involved in presentation of peptide antigens to cytotoxic T-cells. Class II molecules on the other hand are found only on antigen-presenting cells (APC) and are responsible for presenting processed antigenic peptides to T helper cells. In humans, the human leukocyte antigen (HLA) genes, a cluster of genes located on the short arm of chromosome 6, encode the MHC molecules. The major HLA loci are HLA-A, HLA-B, and HLA-C (Class I) and HLA-DP, HLA-DQ, HLA-DR (Class II)[20] and [21].

The HLA-class II genes, specifically theDR2haplotype (DRB1*1501), have been consistently associated with MS susceptibility in nearly all populations tested [22] . Given the critical role of HLA genes in immune response and MS disease susceptibility, it is not surprising that the first MS pharmacogenomics study investigated the influence of HLA alleles, particularly the Class II alleles, on the clinical outcome of IFN-β treatment. In this study, 39 Italian MS patients (R = 22, NR = 17) treated with IFN-β-1a were followed up for 2 years. No significant association between HLA Class II alleles and IFN-β response was found [23] . Three other subsequent independent studies in Spanish MS patients also found no association between DR2 status and IFN-β response[24], [25], and [26]. Cunningham and colleagues sequenced the promoter regions of 100 genes containing IFN-stimulated response elements (ISREs) to identify variants which might modulate treatment response. 162 IFN-β treated Irish MS patients (R = 94, NR = 68) were analysed using a pooled DNA approach followed by validation in the individually typed DNA samples. Of the 100 genes screened, polymorphic markers were observed in a total of 32 genes (54 variants) including one in the HLA-DMA gene. However, no significant association between HLA class II genes and treatment response was found [27] . Kulakova and colleagues examined polymorphic variants from 9 candidate immune response genes including theDRB1locus in 253 Russian MS patients (R = 87, NR = 166) treated with IFN-β. No significant association with treatment response was observed [28] .

Class I genes have also been investigated but again no associations were found[26] and [27]. The findings of the above studies suggest that HLA genes are unlikely to be modifiers of response to IFN-β, per se. However as mentioned previously, HLA class II alleles (HLA-DRB1*04:01, *04:08, *16:01) have been associated with the development of NAbs against IFN-β in a large MS cohort (P < 0.05) [18] .

Other genes located within the MHC region involved in antigen presentation have also been investigated in the context of IFN-β response. A variation in the gene encoding low molecular weight polypeptide (LMP7) was associated with treatment outcome in Irish MS patients from the Cunningham et al. study (rs2071543, p = 0.002, OR: 6.37) [27] .

The gene encoding for cathepsin S (CTSS) was also associated with IFN-β response (rs1136774, p = 0.02, OR: 0.38) [27] . It is also noteworthy thatCTSSis associated with immune-mediated disorders such as rheumatoid arthritis [29] and myasthenia gravis [30] including MS [31] . Higher levels of cathepsin S RNA and serum protein are reported to be aberrantly elevated in MS patients when compared to a control group (p = 3 × 10− 5and p = 0.002, respectively). Also the pre-treatment ratio of serum cathepsin S and its inhibitor cystatin C was higher in patients who responded well to treatment (n = 14) compared to patients who didn't respond (n = 10) (p = 0.003) [32] .

3.1.2. Complement regulatory proteins

CD46 is a type I transmembrane protein expressed on all nucleated human cells. It plays a critical role in regulating the host adaptive immune response and acts as a co-stimulatory molecule for human T-cells inducing their differentiation into specialized regulatory type I cells [33] . CD46 is also a receptor for several human pathogens including human herpesvirus 6 (HHV-6) which has been implicated as an environmental risk factor in the pathogenesis of MS[34] and [35]. Alvarez-Lafuente and colleagues genotyped five tagging SNPs (rs2796267, rs2724385, rs2796269, rs6657476 and rs859706) in theCD46gene in a group of 163 Spanish patients (R = 92, NR = 71) treated with IFN-β and followed up for 1 year. They found the frequency of the rs2724385 AT genotype was significantly lower, and the TT genotype frequency was significantly higher, in Rs compared to NRs (p = 0.007 and p = 0.006, respectively). The percentage of MS patients who responded to IFN-β was also higher among those with decreasedCD46mRNA expression than among those with increasedCD46mRNA (R = 65.9% vs 44.4%, respectively, p = 0.006) [36] .

3.1.3. Cytokines

Interleukin-10 (IL-10) is a potent anti-inflammatory and immunosuppressive cytokine known to modulate disease expression in MS patients as well as in experimental autoimmune encephalomyelitis (EAE), the animal model of MS [37] . Three SNPs in the promoter region of IL-10 (rs1800896, rs1800871 and rs1800872) were investigated in 25 IFN-β treated Caucasian Norwegian RRMS patients. Patients were grouped into 2 haplotype categories, GCC positive (GCC/GCC, GCC/ATA and GCC/ACC) and non-GCC positive (ACC/ACC, ATA/ATA and ACC/ATA). MRI-scans were performed at treatment initiation and then monthly for 6 months during treatment. A trend towards fewer lesions in non-GCC patients was observed, suggesting that these polymorphisms may influence initial response to IFN-β treatment (p = 0.05) [38] . O'Doherty et al. carried out a screen of 61 SNPs in 34 candidate genes in a total of 255 RRMS patients of Irish descent (R = 155, NR = 100) to identify potential modulators of IFN-β response. These variations were selected primarily based on the functional relevance of the genes and outcomes of published gene expression studies. This screen included the 3 aforementionedIL10SNPs from the Wergeland et al. investigation, but no significant associations with response to IFN-β treatment were found [39] . In the Cunningham study a − 162 C/T polymorphism, was discovered in the promoter ofIL10but was not associated with response [27] . Variants inIFNG, which encodes IFN-gamma (IFN-γ), have been implicated in a number of autoimmune conditions including MS[40] and [41]. As such, this gene has been investigated as a potential modifier of IFN-β response. Genetic analysis of a CA dinucleotide repeat in the first intron ofIFNGwas performed in a cohort of 110 RRMS patients treated with IFN-β for a period of 2 years (R = 39, NR = 71). This study found an association between the number of polymorphic CA repeats and the clinical outcome of IFN-β treatment, as determined by the number of relapses [42] . However, Kulakova and colleagues found no association with an intronic SNP inIFNG(rs2430561) and IFN-β response in their study [28] .

The chemotactic cytokine receptor 5 gene,CCR5, is primarily expressed on activated Th1-cells, macrophages, dendritic cells and microglia. Expression ofCCR5has been reported in actively demyelinating MS brain lesions and in EAE high levels of gene expression were found in the acute phase and low level expression during the recovery phase of the disease. A 32 bp deletion (rs333) inCCR5was associated with prolonged relapse-free intervals, less prominent abnormalities on T2-weighted MRI, and less rapid disease progression [43] , although no association between this variant and MS was observed in another study [44] . The same variant was associated with IFN-β response in Russian MS patients (p = 0.036, OR: 1.9) [28] .

Kulakova et al. also found an association with treatment response for a variant in the cytokine coding gene transforming growth factor beta-1 (TGFB1)(rs1800469, p = 0.0062, OR: 9.2) [28] . Other cytokines or their receptor genes which have been examined but for which no significant associations were detected include;IL-1b,IL-4,RANTES,IP9,TNFA,TNFB [27] ,IL1B,IL12RB2,IL10RB [39] ,IL7RAandTNF [28] .

Genes which code for proteins that mediate inflammation (PTGS2) [27] , and involved in co-stimulation (CD274andCTLA4) were also analysed, but no associations with IFN-β response were observed[39] and [28].

3.2. IFN receptor and signalling pathway candidate gene analysis

IFN-β initiates transcription of target genes by binding to the type I IFN receptor (IFNAR1/IFNAR2 complex). Interaction with the receptor causes activation of Jak1 and Tyk2, leading to phosphorylation of Stat1 and Stat2. The latter two then complex with IFN-regulatory factor 9 (IRF9) to form IFN-stimulated gene factor 3 (ISGF3). ISGF3 translocates to the nucleus where it up-regulates expression of IFN-stimulated genes (ISGs) by binding to ISREs in their promoter regions (for overview and illustration of Type I IFN signalling see review [45] ). There are more than 300 such ISGs which code for proteins that are responsible for the pleiotropic biological effects of the IFN-β[46], [47], and [48]. In parallel, activation of Jak1 also initiates other signalling cascades via activation of the PI3K and MAPK pathways [49] .

These events can be restricted or regulated at several different levels by various negative regulators of the Jak/STAT pathway. There are three major classes of negative regulators: SOCS (suppressors of cytokine signalling), PIAS (protein inhibitors of activated STATs) and PTPs (protein tyrosine phosphatases) [50] .

3.2.1. IFNAR1 and IFNAR2 genes

Considering the critical role of the Type I IFN receptor in the initiation of IFN-β mediated events, it is no surprise thatIFNAR1andIFNAR2have been investigated as possible IFN-β response modifying genes. Genetic analysis of 8 SNPs and 2 microsatellites inIFNAR1(PROM-408, rs1041429, rs1012334, rs1012335, rs2257167, GTnrepeat andTTTArepeat) andIFNAR2(rs3153, rs1051393 and rs1131668) in a total of 105 Spanish MS patients (R = 57, NR = 48) found no significant association with response status. However, one variant rs1012334 showed a trend towards association with relapse-free status (p = 0.03) [51] . Several of these markers inIFNAR1(rs2257167 and PROM-408) andIFNAR2(rs1051393) were also analysed in another larger Spanish cohort of 147 MS patients (R = 104, NR = 43) but showed no association with IFN-β response [52] . In the Irish studies no association was found forIFNAR1variants rs1012334, PROM-408 [27] , andIFNAR2rs1131668 (also referred to as rs8127890) [39] . Furthermore, no association was found in Russian patients forIFNAR1rs1012335 [28] . However, the GTnrepeat microsatellite (rs55884088) analysed by Sriram et al. in the promoter region ofIFNAR1was associated with IFN-β response in Irish MS patients (p = 0.036) [27] .

3.2.2. Interferon-regulatory factor (IRF) genes

Genes which code for the IFN-regulatory factors (IRFs), a group of proteins that regulate transcription of IFNs [53] , have also been investigated in the context of IFN-β induced response.IRF5is associated with susceptibility to MS [54] and other autoimmune diseases including inflammatory bowel diseases [55] and systemic lupus erythematosus [56] . Two SNPs inIRF5(rs4728142 and rs3807306), which are associated with MS susceptibility, were analysed in 218 IFN-β treated patients from Spain (R = 106, NR = 112). The prevalence of human herpesvirus (HHV-6) infection, one of the possible environmental risk factors in the pathogenesis of MS, was also investigated in these patients. The results suggested an increased frequency of the rs4728142*A allele and rs3807306*T allele in IFN-β Rs, although not statistically significant (p = 0.27 and p = 0.09, respectively). The rs3807306*T allele was also associated with HHV-6 positive status (p = 0.05). The authors hypothesised that this rs3807306*T variant, also a risk allele for MS, may predispose to low level production of IRF5. In the presence of HHV-6 infection, the viral particles would easily interact with and inhibit this relatively low amount of IRF5, thus abolishing IFN production. As such patients with the T allele would be most likely to benefit from administration of exogenous IFN-β [57] .

Vosslamber and colleagues analysed severalIRF5gene variants (rs2004640, rs10954213, rs4728142 and a 30 bp insertion/deletion polymorphism in exon 6) and found that RRMS patients with the rs2004640-TT and rs4728142-AA genotypes responded poorly to IFN-β treatment compared with patients carrying the respective G-alleles (p = 0.0006 and p = 0.0023, respectively). Response was defined as the ratio of average gene expression during IFN therapy to the average gene expression before treatment for the 10 ISGs in 30 RRMS patients. Furthermore NRs (who had developed one or more new T2 lesions in any year or 12 month interval on IFN-β) (n = 42) were more likely to have the rs2004640-TT genotype compared to those not fulfilling this definition (n = 31) (p = 0.003). To validate these associations, SNPs—rs2004640 and rs4728142 were further genotyped in an independent group of 261 RRMS patients. The validation results confirmed that the rs2004640-TT genotype was associated with a shorter time to first relapse (p = 0.037). However, no association was observed with rs4728142 [58] . The results of these twoIRF5studies are somewhat conflicting given that the rs4728142-AA genotype was essentially a marker of reduced IFN-β activity in the Vosslamber study but appeared to be more frequent (though not significantly so) in Rs in the Vandenbroeck study [57] . Further investigation of this gene will be required to fully uncover its relevance or otherwise as a modulator of IFN-β response.

A SNP (rs17445836) located nearIRF8has also been investigated as a potential modifier of IFN-β induced response. This polymorphism was previously associated with susceptibility to MS and was shown to influence expression of IFN response genes[59] and [60]. rs17445836 was genotyped in 424 IFN-β treated MS patients, whereby response was defined in terms of the time to first event. An event was defined as either a clinical relapse, change in T2 hyperintense lesion burden or presence of a gadolinium-enhancing lesion on MRI, or increase of EDSS by 1 point sustained over 6 months. Patients who were homozygotes for the A allele displayed a significantly shorter time to event than carriers of the other genotypes (adjusted p = 0.017). However, this significant association could not be replicated in a follow-up cohort of NAb negative German patients (n = 211) in which event was defined only by the occurrence of a relapse [61] .

More recently, the ubiquitin specific peptidase 18 (USP18) gene, a negative regulator of the type I IFN signalling pathway, was evaluated to verify its influence on the clinical outcome of IFN-β treatment. TwoUSP18SNPs (rs2542109 and rs9618216) were genotyped in a cohort 225 MS patients treated with IFN-β (R = 130, NR = 95). rs2542109 was significantly associated with IFN-β response status. The AA genotype was more frequent amongst Rs than NRs (p = 0.041, OR: 1.8) [62] .

No other studies have reported any associations in genes relating to the IFN type I receptor and its immediate signalling pathways. The following genes have been investigated with no evidence for association;ISG12,ISG56,ISG20,IFP53,SP100 [27] IRF4,IRF9,TYK2,PIAS1,SOCS3 [39] andIFNB1 [28] .

3.3. Interferon induced anti-viral response molecules

Expression of the human myxovirus resistance gene A (MxA) is known to be regulated by type I IFNs and is a sensitive biomarker of IFN-β bioactivity [63] . In the study conducted by Cunningham and colleagues, 2 variations inMxAgene, rs2071430 and rs17000900, were found to be significantly associated with IFN-β response (p = 0.015 and p = 0.018, respectively) [27] . However, Weinstock-Guttman et al. examined these SNPs in 37 NAb negative RRMS patients (R = 19, NR = 18) and found no significant associations with clinical response orMxAgene expression  [64] . Other IFN inducible genesTLR7 [27] andIDO1,GBP1, andPRKR [39] have been analysed in relation to IFN-β response, but no significant associations were found.

3.4. Pro-apoptotic molecules

TNF related apoptosis inducing ligand (TRAIL) and its receptors genes are reported to be involved in the pathogenesis of MS and other autoimmune diseases.TRAILis also induced by IFN-β in T-cells, natural killer cells and monocytes. Recently, López-Gómez et al. genotyped a total of 54 SNPs inTRAIL,TRAILR-1,TRAILR-2,TRAILR-3andTRAILR-4in 509 Spanish Caucasian RRMS patients (R = 213, NR = 296) who had been treated with IFN-β. Significant results were validated in a further 226 (R = 104, NR = 122) patients. The CC genotype of rs20576, located in exon 5 ofTRAILR-1, was associated with IFN-β response in both the original and validation cohorts (p = 0.056, OR: 0.41; p = 4.83 × 10− 3, OR: 0.19, respectively). The association remained significant when a subset of “NAb-free patients” (n = 350) were analysed (p = 1.17 × 10− 3, OR: 0.13), ruling out a confounding effect of NAbs [65] . Members of the cysteine protease family, which play a crucial role in apoptosis (CASP10,CASP3,CASP2,CASP7,CASP5,CASP8) [39] , and other regulators of apoptosis includingTSC22 [27] CFLAR,DUSP1 [39] , have also been investigated, but were not associated with IFN-β response.

3.5. Validated IFN-a response markers

Both IFN-α and IFN-β are known to share a common receptor. As such significant associations from pharmacogenomic studies examining IFN-α treatment response in HCV patients have been re-examined in MS patients treated with IFN-β. Candidate gene and GWA studies have shown several SNPs within theIL28Bgene region to be significantly associated with IFN-α treatment response in HCV infected patients from different populations[7], [8], [9], [66], and [67]. Malhotra and colleagues performed a candidate gene study to investigate whether the sameIL28Bvariants were also associated with IFN-β response in patients with MS. Neither of the twoIL28BSNPs (rs8099917 and rs12979860) analysed in 588 MS patients (R = 281, NR = 307) were associated with IFN-β response [68] .

3.6. Others

Other functionally relevant candidate genes have been evaluated as potential modifiers of clinical response to IFN-β, but no significant associations were found. These include genes involved in: cell adhesion;VCAM,ICAM1 [27] ITGB2,ITGA4 [39] , genes related to dopamine neurotransmission;DRD3,DBH,Th [39] , genes involved in specialized functions like keratinisation (SPRR2A), spermiogenesis (PRM3), regulation of neurotransmitter release (SYN2) [27] , and regulation of body weight (LEP) [39] , along with other genes involved in various critical biological processes (P21) [27] , (PRKCA) [39] .

3.7. Candidate gene–gene interactions

As it is hypothesised that multiple genetic variations may be critical in determining the heterogeneous nature of response to IFN-β treatment, several studies have examined the combined effects of candidate genes and their variations. Various statistical tools have been used to explore potential SNP–SNP/gene–gene interactions, and assess their collective impact on IFN-β induced response.

As mentioned previously, single marker analysis of 61 SNPs from 34 possible candidate genes in 255 Irish MS patients found no significant associations with IFN-β response. However, by applying the APsampler, a multi-locus/multi-level association analysis algorithm, they were able to identify allelic combinations that differed significantly between Rs and NRs. The most significant allelic combinations wereJAK2–IL10RB–GBP1–PIAS1(p = 0.0008) followed byJAK2–IL10–CASP3(p = 0.001) [39] .

By applying the same statistical algorithm as O'Doherty and colleagues, Kulakova et al., found that patients carrying triallelic combinations ofCCR5*d + IFNAR1*G + IFNB1*T/T andCCR5*d + IFNAR1*G + IFNG*T (p = 0.017 and p = 0.035, respectively) may respond favourably to IFN-β treatment [28] . It is interesting to note that all significant combinations from this study included alleles/genotypes of theCCR5gene.

4. Genome wide association studies

To date, only 2 GWAS have been carried out to identify genetic determinants of response to IFN-β treatment in MS patients. The first of these studies was published in 2008 by Byun and colleagues. In an original cohort of 206 patients, response status was determined after 2 years of treatment follow-up. Patients with no relapses and no increase in EDSS were classified as Rs, while patients who had 2 or more relapses or an increase in EDSS by at least one point were classified as NRs. In the first stage, pooled DNA samples from the 206 patients (R = 99, NR = 107) were genotyped using an Affymetrix 100K GeneChip array. Each sample was included in 3 different pools. The 35 most significant associations from stage 1 were selected for validation on the basis of p-value (15 SNPs — 13 with p-value cut-off of < 0.0005 and 2 SNPs with p-value of < 0.00005 in 2 of 3 replicates), and robust cluster ranking (20 SNPs). These SNPs were genotyped individually in a validation cohort of 285 which consisted of the 206 patients from stage one and an additional 79 MS patients (R = 143, NR = 142). Of the shortlisted 35 SNPs, 18 remained significantly associated with response, with the top most significant associations inHAPLN1andGPC5(details of other significant intragenic SNPs are shown in Table 1 ).

Table 1 Validated intragenic associations from the pharmacogenomics GWAS by Byun and colleagues.

Gene Chromosome SNPs p value for joint analysis
HAPLN1 5 rs4466137 0.004
GPC5 13 rs10492503, rs9301789 0.007, 0.01
TAFA1 3 rs4855469 0.01
LOC442331 7 rs6944054 0.014
NPAS3 14 rs4128599 0.024
COL25A1 4 rs794143 0.037
CAST 3 rs10510779 0.042

Data summarized from Byun et al. [69] .

Results of gene ontology classification analysis highlighted an enrichment of gamma-aminobutyric acid and glutamate receptor genes. Both of these receptors are present in the CNS and are known to play an important role in neural transmission. SNPs in or near (75 Kb upstream or downstream) 112 candidate genes nominated by others (mainly Cunningham et al., 2005), although not necessarily genotyped previously, were also interrogated. A number of weak associations were detected includingADAR(rs4131514) (p = 0.0015) ( Table 2 ). There were no significant associations forIFNAR1,IFNAR2,LMP7,CTSSorMxA, which had been the focus of earlier candidate gene studies [69] .

Table 2 Significant associations from the study by Byun and colleagues, for SNPs in or near candidate genes listed by previous/other studies.

Chromosome Gene/Nearest Gene SNPs p value
1 ADAR rs4131514 0.0015
1 PTGS2 rs10494593, rs1924743 0.0192, 0.0363
2 PRKR rs728005 0.0212
3 SYN2 rs60834, rs795000 0.0484
4 CFI rs1990250, rs6814789 0.0055, 0.0093
9 OGN rs1121979 0.0119
11 TRAF6 rs10501154 0.0352
11 CASP1 rs580253 0.0369
12 IFN-γ rs10492199, rs10492198 0.0458, 0.0343
12 IL-22 rs10506559 0.0378
20 POLR3F rs3736775 0.0305

Data summarized from Byun et al. [69] .

An independent follow-up study aimed to replicate the significant associations identified by Byun et al. (specifically inHAPLN1andGPC5), in a similar ethnic population (R = 55, NR = 79). Only one marker,GPC5rs10492503 remained significant (p = 0.0005) whileHAPLN1variations did not show any significant difference between Rs and NRs [70] .

A second IFN-β pharmacogenomics response GWAS, published in 2009, was undertaken in 106 RRMS patients (R = 53, NR = 53) [71] . Similar to the previous GWAS, patient DNA samples were pooled. The response criteria were also similar. Genotyping was performed on a much denser Affymetrix 500K array. A total of 428,867 SNPs were successfully analysed, out of which 383 SNPs were selected for validation on the basis of ranking criteria generated using an in house silhouette statistic and clustering methods. These were individually genotyped in an independent patient cohort of 94 samples (R = 49, NR = 45). Eighteen out of 383 shortlisted SNPs maintained significance in the validation stage. The most significant association was observed for rs12557782 inGRIA3(p = 0.002, OR: 2.7). This gene codes for glutamate receptor 3, supporting the results of gene ontology analysis from the previous GWAS [69] . Details of the other top significant SNPs are given in Table 3 . Gene ontology analysis highlightedIFNAR2, as well asADARwhich is an ISG involved in host responses to viral infection [72] .

Table 3 Validated intragenic associations from the pharmacogenomics GWAS by Comabella and colleagues.

Gene Chromosome SNPs p value in validation cohort
GRIA3 X rs12557782 < .001 in women & 0.56 in men
ADAR 1 rs2229857 0.002
ZFAT 8 rs733254 0.002
CIT 12 rs7308076 0.003
IFNAR2 21 rs2248202 0.02
STARD13 13 rs9527281 0.03
ZFHX4 8 rs11787532 0.03

Data summarized from Comabella et al. [71] .

Taken together with the findings of the Byun study, this seems to support a role forADARas a modulator of IFN-β response. This study also investigated the effects of combinations of multiple SNPs on IFN-β response; however no significant combinations of alleles (pairs or triplets) were identified [71] .

5. IFN-β summary

Traditional candidate gene studies performed on the basis of prior knowledge of biological and/or functional relevance of plausible candidate gene(s) have generally proven to be of limited success in identifying reliable determinants of IFN-β response in MS. “Hypothesis-free” GWAS, utilizing advances in SNP array technologies, have the advantage of interrogating hundreds of thousands of SNPs in one single experiment. GWAS performed to uncover genetic modifiers of IFN-β response unequivocally suggest that not a single gene or pathway but multiple genes and pathways may play a role in IFN-β response. TheGPC5association has been confirmed in at least one other independent study, and both GWAS suggested a role for glutamate receptors andADAR. In spite of generating somewhat consistent and/or “plausible” results, these studies are not without their limitations. One of the biggest challenges, particularly for pharmacogenomics studies, is recruitment of patients with comprehensive clinical information such that the number in each comparator group is large enough to have sufficient statistical power. Of course the effect sizes may not be as small as for susceptibility studies, which seem to be the case withGPC5. However it is likely that both of the published IFN-β pharmacogenomics GWAS were underpowered. Both investigations also used pooled patient samples for their stage one. While this has the advantage of reducing the cost and time to genotype thousands of SNPs, lack of uniform mixing during the pooling of equimolar samples may lead to loss of individual genotype data and hence reduce the power of the study[73] and [74]. Pooling of DNA also hampers the quality checks to eliminate SNPs with low genotype success rate and high heterozygosity rate. Moreover, without individual genotype data it is not possible to assess for population stratification or cryptic relatedness within the cohort, and imputation cannot be carried out. Finally, arrays used in both investigations had limited genomic coverage. Byun et al. used the Affymetrix 100K array which covers 31% of the Caucasian genome with 100,000 SNPs and Comabella et al. used the Affymetrix 500K array which covers 65% of the Caucasian genome [75] . Therefore, high powered GWAS (or indeed whole genome sequencing) performed using larger well characterised patient cohorts, and arrays with greater genomic coverage are needed to better probe and identify genetic factors underlying IFN-β response.

6. Glatiramer acetate

Glatiramer acetate (Copaxone®) formerly known as copolymer-1 is another first line disease modifying drug available to treat MS. GA is a random polymer of glutamic acid, lysine, alanine, and tyrosine, with a structure similar to myelin basic protein. GA was shown to significantly lower the number of lesions and exacerbations compared with patients who received placebo[76] and [77]and was approved in the U.S. and Europe in 1996 and 2000, respectively. As with IFN-β, the mechanism of action of GA is not completely understood. The results of in vitro and in vivo systems suggest that GA has multiple effects on the host's innate and adaptive immune system. By binding to MHC class II molecules, GA can prevent the presentation of other antigens and stop subsequent T-cell activation. GA also appears to have various other effects on T-cells, B-cells and monocytes. Broadly speaking, GA promotes the production of anti-inflammatory cytokines, as well as increasing the number of regulatory T-cells and reducing the number of Th17 cells. Furthermore, it has been demonstrated that GA treatment promotes neurogenesis, neuroprotection and remyelination in MS lesions by inducing secretion of neurotrophic factors such as brain-derived neurotrophic factor (BDNF) and neurotrophin (NT)-3 and 4 [78] . Response to GA varies with approximately 50% of patients not responding to therapy[23] and [76]. Hence it is necessary to identify Rs and NRs, and consider alternative treatment strategies for non-responding MS patients.

6.1. Genetic markers of GA treatment response

To date, very few studies have evaluated the pharmacogenomics of response to GA treatment. In the first such study, Fusco et al. investigated the relationship between HLA class II alleles (DRB1 and DQB1) and GA response in a total of 44 patients (R = 22, NR = 22). MS patients who were DRB1*1501 positive were more likely to respond to GA treatment compared to patients who were DRB1*1501 negative (p = 0.016) [23] . In a study conducted by Grossman and colleagues, HLA-DRB*1501 and 61 SNPs from 27 relevant candidate genes were analysed in 174 patients from two multi-centre GA clinical trials (European/Canadian; 49 treated: 52 placebo, and US; 36 treated: 37 placebo) to identify potential genetic modifiers of GA treatment response. All patients were classified as Rs or NRs to GA treatment based on the end-point determinants of the original clinical trials they were part of (European/Canadian: T1 enhancing lesions; US: relapse rates). Seven genes were associated with treatment response in at least one of the cohorts ( Table 4 ). None of these associations were detected in the placebo groups indicating that these associations were related to true drug effects. A SNP inTRB@(rs71878) was significantly associated with response in both cohorts. T-cell receptor beta (TCRB) proteins which are coded by theTRB@gene cluster are known to play a critical role in immunological processes controlling MS pathoetiology and may play a role in GA signal transduction [79] . Grossman et al. also identified 2 polymorphisms within theCTSSgene (rs2275235 and rs1415148) which were significantly associated with GA response even after permutation correction in the European/Canadian trial cohort. The most significant association from the Fusco et al. study, the HLA-DRB1*1501 variation, did not show any significance in this study. The authors suggested that variations in ethnicity and geographic locations may account for this, as the frequency of HLA alleles tends to vary across different populations [80] .

Table 4 Genetic variations significantly associated with response to GA treatment in patients from 2 different clinical trials.

SNPs Gene Chromosome p value
European/Canadian clinical trial
rs2275235, rs1415148 CTSS 3 0.0009, 0.0018
rs470929 MBP 18 0.0038
rs982764 FAS 10 0.03
rs71878 TRB@ 7 0.039
rs2001791, rs1129055 CD86 3 0.04, 0.04
rs956730 IL1R1 2 0.049
US clinical trial
rs71878 TRB@ 7 0.006
rs946685 IL12RB2 1 0.045

Data summarized from Grossman et al. [80] .

In a retrospective study, Gross and colleagues investigated HLA-DRB1*1501 allele status with GA treatment response. They genotyped rs3135388, a known proxy marker for HLA-DRB1*1501 (r2 = 0.97) in a total of 332 RRMS patients treated with GA. Their data suggested that patients homozygous for the A allele responded better to treatment than patients who were either heterozygotes (adjusted p = 0.015) or homozygotes for the G allele (adjusted p = 0.048) [61] . In a study of 214 Russian RRMS patients (R = 130, NR = 84), variants from nine candidate genes, HLA class IIDRB1variants,CCR5(rs333),IFNAR1(rs1012335),IFNB1(rs1051922),TNF(rs1800629),IFNG(rs2430561),TGFB1(rs1800469),CTLA4(rs231775) andIL7RA(rs6897932), were examined in relation to GA treatment induced response. No individual polymorphism was significantly associated with response. However, using the APsampler algorithm, allelic combinations ofDRB1*15 + TGFB1*T + CCR5*d + IFNAR1*G, followed byDRB1*15 + TGFB1*T + CCR5*d were found to be significantly more frequent in NRs than in Rs (permutation p = 0.0056 and p = 0.013, respectively) [81] .

In a recent study, Dhib-Jalbut and colleagues genotyped DR15, DR17, DQ2, and DQ6 alleles in 64 RRMS patients treated with GA (R = 30, NR = 34) followed up for at least 2 years. MS patients who were carriers of DR15 or DQ6 (p = 0.020 and p = 0.014, respectively) and patients who lacked the DR17 and DQ2 alleles (p = 0.012 and p = 0.002, respectively) responded favourably to GA treatment. Also patients carrying the DR15–DQ6 haplotypes and who lacked the DR17–DQ2 haplotype responded well to treatment (adjusted p = 0.0437 and p = 0.0462, respectively). The results of prognostic models developed using multiple haplotype status revealed that a combined DR15-DQ6 positive but DR17-DQ2 negative status was a strong predictor of favourable clinical response to GA (71% were Rs), and DR15-DQ6 negative but DR17-DQ2 positive haplotype status was a strong predictor of poor clinical response (17% were Rs), while the presence or absence of either DR15-DQ6 and DR17-DQ2 status were neutral predictors of response (39% were Rs) to GA treatment [82] .

López-Gómez and colleagues genotyped the IFN-β response modifyingTRAILR1variant (rs20576) in 77 GA treated patients (R = 29, NR = 48) but found no association with response [65] .

In conclusion, several studies have indicated a beneficial effect of DR15[23], [61], and [82], which seems consistent with the mechanism of action of GA. However one study did not find such an association [81] . These findings need to be replicated in much larger independent GA treated cohorts.

7. Conclusion

In recent years there has been an increased focus on identification of a potential biomarker for responsiveness to DMDs in MS. Except for the role ofGPC5in IFN-β response, results have on the whole been inconsistent or lack further validation. A critical limitation of most pharmacogenomics studies to date is that they have insufficient power to detect small true positive associations. This is partly because of the difficultly in acquiring detailed drug related information retrospectively. The use of comprehensive databases such as MSbase [83] should help researchers to easily obtain detailed clinical data for future MS pharmacogenomic studies. Furthermore, unlike in HCV infection where response to IFN-α can be easily quantified, response to DMDs in MS is more difficult to define and little consensus exists as to the most appropriate criteria to use. Use of extreme phenotype criteria such as those proposed by the UEPHA-MS network [84] should help to avoid misclassification and facilitate comparison between different centres and different patient populations. Incorporation of placebo-treated patient groups would also help to confirm that associations in the treatment group are indeed true drug response effects. However, for ethical reasons this is probably only feasible for emerging treatments in the context of clinical trials. While MS pharmacogenomics studies to date have been limited in terms of the genetic variants examined, continually improving genomic technologies should over the coming years, help to extend our understanding of the genetic factors influencing the efficacy of these DMDs and thus aid the identification of a biomarker(s) of response. Such biomarkers will not only aid in the identification of likely Rs or NRs, but also provide new insights into the mechanisms of action of these drugs and in doing so potentially identify new drug targets.

Although a number of new, potentially more effective DMDs have been approved in recent years, and others are likely to be approved in the near future, the serious adverse effects of these medications and limited experience post marketing will still leave room for the traditional agents which have been in use for the last two decades. With increased choice, a diversity of risk/benefit profiles and costs, comes a greater need for rationalised selection of these first line agents. As such the field of MS pharmacogenomics is likely to expand over the next decade.

Take-home messages


  • First line DMDs, IFN-β and GA reduce relapses, and delay disease progression with minimal side effects. However approximately 20–50% of patients do not respond well to these treatments.
  • Identification of genetic variants which predict clinical response to IFN-β or GA would facilitate timely alternative treatment in likely ‘NRs’.
  • To date, most MS pharmacogenomic studies have focused on IFN-β, and have examined candidate genes. Two GWAS have been carried out and one of these studies highlighted a variant inGPC5as a modifier of response. This was confirmed in a separate study. Both GWAS also suggested a role for glutamate receptors andADAR.
  • The field of pharmacogenomics in MS is likely to expand over the coming years, with the increasing number of treatment options available and variable efficacy/toxicity profiles, but larger studies are required to identify true associations.

The following is the supplementary data related to this article.

Download file

Supplementary Table A.1 Studies examining genetic polymorphisms as determinants of response to interferon beta or glatiramer acetate treatment in multiple sclerosis patients.

Disclosure statement

The authors report no conflicts of interest.


  • [1] R. Milo, E. Kahana. Multiple sclerosis: geoepidemiology, genetics and the environment. Autoimmun Rev. 2010;9:A387-A394
  • [2] A.D. Sadovnick. Genetic background of multiple sclerosis. Autoimmun Rev. 2012;11:163-166
  • [3] The International Multiple Sclerosis Genetics Consortium (IMSGC) analysis of immune-related loci identifies 48 new susceptibility variants for multiple sclerosis. Nat Genet. 2013;10.1038/ng.2770
  • [4] C. O'Gorman, R. Lucas, B. Taylor. Environmental risk factors for multiple sclerosis: a review with a focus on molecular mechanisms. Int J Mol Sci. 2012;13:11718-11752
  • [5] K.A. Buzzard, S.A. Broadley, H. Butzkueven. What do effective treatments for multiple sclerosis tell us about the molecular mechanisms involved in pathogenesis?. Int J Mol Sci. 2012;13:12665-12709
  • [6] Z. Rotstein, R. Hazan, Y. Barak, A. Achiron. Perspectives in multiple sclerosis health care: special focus on the costs of multiple sclerosis. Autoimmun Rev. 2006;5:511-516
  • [7] D. Ge, J. Fellay, A.J. Thompson, J.S. Simon, K.V. Shianna, T.J. Urban, et al. Genetic variation in IL28B predicts hepatitis C treatment-induced viral clearance. Nature. 2009;461:399-401
  • [8] V. Suppiah, M. Moldovan, G. Ahlenstiel, T. Berg, M. Weltman, M.L. Abate, et al. IL28B is associated with response to chronic hepatitis C interferon-alpha and ribavirin therapy. Nat Genet. 2009;41:1100-1104
  • [9] Y. Tanaka, N. Nishida, M. Sugiyama, M. Kurosaki, K. Matsuura, N. Sakamoto, et al. Genome-wide association of IL28B with response to pegylated interferon-alpha and ribavirin therapy for chronic hepatitis C. Nat Genet. 2009;41:1105-1109
  • [10] S. Dhib-Jalbut, S. Marks. Interferon-beta mechanisms of action in multiple sclerosis. Neurology. 2010;74:S17-S24
  • [11] B.C. Kieseier. The mechanism of action of interferon-β in relapsing multiple sclerosis. CNS Drugs. 2011;25:491-502
  • [12] D.W. Paty, D.K. Li. Interferon beta-1b is effective in relapsing-remitting multiple sclerosis. II. MRI analysis results of a multicenter, randomized, double-blind, placebo-controlled trial. UBC MS/MRI Study Group and the IFNB Multiple Sclerosis Study Group. Neurology. 1993;43:662-667
  • [13] L.D. Jacobs, D.L. Cookfair, R.A. Rudick, R.M. Herndon, J.R. Richert, A.M. Salazar, et al. Intramuscular interferon beta-1a for disease progression in relapsing multiple sclerosis. The Multiple Sclerosis Collaborative Research Group (MSCRG). Ann Neurol. 1996;39:285-294
  • [14] A. Bertolotto. Implications of neutralising antibodies on therapeutic efficacy. J Neurol Sci. 2009;277:S29-S32
  • [15] E. Sbardella, V. Tomassini, C. Gasperini, F. Bellomi, L.A. Cefaro, V.B. Morra, et al. Neutralizing antibodies explain the poor clinical response to interferon beta in a small proportion of patients with multiple sclerosis: a retrospective study. BMC Neurol. 2009;9:54
  • [16] S. Hoffmann, S. Cepok, V. Grummel, K. Lehmann-Horn, J. Hackermüller, P.F. Stadler, et al. HLA-DRB1*0401 and HLA-DRB1*0408 are strongly associated with the development of antibodies against interferon-beta therapy in multiple sclerosis. Am J Hum Genet. 2008;83:219-227
  • [17] F. Weber, S. Cepok, C. Wolf, A. Berthele, M. Uhr, T. Bettecken, et al. Single-nucleotide polymorphisms in HLA- and non-HLA genes associated with the development of antibodies to interferon-β therapy in multiple sclerosis patients. Pharmacogenomics J. 2012;12:238-245
  • [18] D. Buck, S. Cepok, S. Hoffmann, V. Grummel, A. Jochim, A. Berthele, et al. Influence of the HLA-DRB1 genotype on antibody development to interferon beta in multiple sclerosis. Arch Neurol. 2011;68:480-487
  • [19] C. Enevold, A.B.. Oturai, P.S. Sørensen, L.P. Ryder, N. Koch-Henriksen, K. Bendtzen. Polymorphisms of innate pattern recognition receptors, response to interferon-beta and development of neutralizing antibodies in multiple sclerosis patients. Mult Scler. 2010;16:942-949
  • [20] U. Shankar kumar. The human leukocyte antigen (HLA) system. Int J Hum Genet. 2004;4:91-103
  • [21] D.J. Penn, P. Ilmonen. Major histocompatibility complex (MHC). (eLS. John Wiley & Sons Ltd., Chichester, 2005) 10.1038/npg.els.0003986
  • [22] J.R. Oksenberg, L.F. Barcellos. Multiple sclerosis genetics: leaving no stone unturned. Genes Immun. 2005;6:375-387
  • [23] C. Fusco, V. Andreone, G. Coppola, V. Luongo, F. Guerini, E. Pace, et al. HLA-DRB1*1501 and response to copolymer-1 therapy in relapsing-remitting multiple sclerosis. Neurology. 2001;57:1976-1979
  • [24] P. Villoslada, L.F. Barcellos, J. Rio, A.B.. Begovich, M. Tintore, J. Sastre-Garriga, et al. The HLA locus and multiple sclerosis in Spain. Role in disease susceptibility, clinical course and response to interferon-beta. J Neuroimmunol. 2002;130:194-201
  • [25] O. Fernández, V. Fernández, C. Mayorga, M. Guerrero, A. León, J.A. Tamayo, et al. HLA class II and response to interferon-beta in multiple sclerosis. Acta Neurol Scand. 2005;112:391-394
  • [26] M. Comabella, M. Fernández-Arquero, J. Río, A. Guinea, M. Fernández, M.C. Cenit, et al. HLA class I and II alleles and response to treatment with interferon-beta in relapsing-remitting multiple sclerosis. J Neuroimmunol. 2009;210:116-119
  • [27] S. Cunningham, C. Graham, M. Hutchinson, A. Droogan, K. O'Rourke, C. Patterson, et al. Pharmacogenomics of responsiveness to interferon IFN-beta treatment in multiple sclerosis: a genetic screen of 100 type I interferon-inducible genes. Clin Pharmacol Ther. 2005;78:635-646
  • [28] O.G. Kulakova, E.Y. Tsareva, A.N. Boyko, S.G. Shchur, E.I. Gusev, D. Lvovs, et al. Allelic combinations of immune-response genes as possible composite markers of IFN-β efficacy in multiple sclerosis patients. Pharmacogenomics. 2012;13:1689-1700
  • [29] S. Gay, R.E. Gay, W.J. Koopman. Molecular and cellular mechanisms of joint destruction in rheumatoid arthritis: two cellular mechanisms explain joint destruction?. Ann Rheum Dis. 1993;52:S39-S47
  • [30] H. Yang, M. Kala, B.G. Scott, E. Goluszko, H.A. Chapman, P. Christadoss. Cathepsin S is required for murine autoimmune myasthenia gravis pathogenesis. J Immunol. 2005;174:1729-1737
  • [31] T. Burster, A. Beck, E. Tolosa, P. Schnorrer, R. Weissert, M. Reich, et al. Differential processing of autoantigens in lysosomes from human monocyte-derived and peripheral blood dendritic cells. J Immunol. 2005;175:5940-5949
  • [32] D. Haves-Zburof, T. Paperna, A. Gour-Lavie, I. Mandel, L. Glass-Marmor, A. Miller. Cathepsins and their endogenous inhibitors cystatins: expression and modulation in multiple sclerosis. J Cell Mol Med. 2011;15:2421-2429
  • [33] S. Ni Choileain, A.L. Astier. CD46 processing: a means of expression. Immunobiology. 2012;217:169-175
  • [34] M.L. Opsahl, P.G. Kennedy. Early and late HHV-6 gene transcripts in multiple sclerosis lesions and normal appearing white matter. Brain. 2005;128:516-527
  • [35] J.M. Reynaud, B. Horvat. Human herpes virus 6 and neuroinflammation. ISRN Virol. 2013;1110.5402/2013/834890
  • [36] R. Alvarez-Lafuente, F. Blanco-Kelly, M. Garcia-Montojo, A. Martínez, Heras V. De Las, M.I. Dominguez-Mozo, et al. CD46 in a Spanish cohort of multiple sclerosis patients: genetics, mRNA expression and response to interferon-beta treatment. Mult Scler. 2011;17:513-520
  • [37] E. Ersoy, C.N. Kuş, U. Sener, I. Coker, Y. Zorlu. The effects of interferon-beta on interleukin-10 in multiple sclerosis patients. Eur J Neurol. 2005;12:208-211
  • [38] S. Wergeland, A. Beiske, H. Nyland, H. Hovdal, D. Jensen, J.P. Larsen, et al. IL-10 promoter haplotype influence on interferon treatment response in multiple sclerosis. Eur J Neurol. 2005;12:171-175
  • [39] C. O'Doherty, A. Favorov, S. Heggarty, C. Graham, O. Favorova, M. Ochs, et al. Genetic polymorphisms, their allele combinations and IFN-beta treatment response in Irish multiple sclerosis patients. Pharmacogenomics. 2009;10:1177-1186
  • [40] J.Y. Lee, D. Goldman, L.M. Piliero, M. Petri, K.E. Sullivan. Interferon-gamma polymorphisms in systemic lupus erythematosus. Genes Immun. 2001;2:254-257
  • [41] O.H. Kantarci, A. Goris, D.D. Hebrink, S. Heggarty, S. Cunningham, I. Alloza, et al. IFNG polymorphisms are associated with gender differences in susceptibility to multiple sclerosis. Genes Immun. 2005;6:153-161
  • [42] A. Martínez. de las Heras V, Mas Fontao A, Bartolomé M, de la Concha EG, Urcelay E, et al. An IFNG polymorphism is associated with interferon-beta response in Spanish MS patients. J Neuroimmunol. 2006;173:196-199
  • [43] F. Sellebjerg, H.O. Madsen, C.V. Jensen, J. Jensen, P. Garred. CCR5 delta32, matrix metalloproteinase-9 and disease activity in multiple sclerosis. J Neuroimmunol. 2000;102:98-106
  • [44] M.K. Arababadi, G. Hassanshahi, H. Azin, V.A. Salehabad, M. Araste, R. Pourali, et al. No association between CCR5-Δ32 mutation and multiple sclerosis in patients of southeastern Iran. LabMedicine. 2010;41:31-33 10.1309/LM9TU9ID1CGZVLXL
  • [45] L.M. Burdick, N. Somani, A.K. Somani. Type I IFNs and their role in the development of autoimmune diseases. Expert Opin Drug Saf. 2009;8:459-472
  • [46] M.J. de Veer, M. Holko, M. Frevel, E. Walker, S. Der, J.M. Paranjape, et al. Functional classification of interferon-stimulated genes identified using microarrays. J Leukoc Biol. 2001;69:912-920
  • [47] M.M. Brierley, E.N. Fish. Review: IFN-alpha/beta receptor interactions to biologic outcomes: understanding the circuitry. J Interferon Cytokine Res. 2002;22:835-845
  • [48] L.C. Platanias. Mechanisms of type-I- and type-II-interferon-mediated signalling. Nat Rev Immunol. 2005;5:375-386
  • [49] J.S. Rawlings, K.M. Rosler, D.A. Harrison. The JAK/STAT signaling pathway. J Cell Sci. 2004;117:1281-1283
  • [50] K. Rakesh, D.K. Agrawal. Controlling cytokine signaling by constitutive inhibitors. Biochem Pharmacol. 2005;70:649-657
  • [51] U. Sriram, L.F. Barcellos, P. Villoslada, J. Rio, S.E. Baranzini, S. Caillier, et al. Pharmacogenomic analysis of interferon receptor polymorphisms in multiple sclerosis. Genes Immun. 2003;4:147-152
  • [52] L. Leyva, O. Fernández, M. Fedetz, E. Blanco, V.E. Fernández, B. Oliver, et al. IFNAR1 and IFNAR2 polymorphisms confer susceptibility to multiple sclerosis but not to interferon-beta treatment response. J Neuroimmunol. 2005;163:165-171
  • [53] H. Nguyen, J. Hiscott, P.M. Pitha. The growing family of interferon regulatory factors. Cytokine Growth Factor Rev. 1997;8:293-312
  • [54] G. Kristjansdottir, J.K. Sandling, A. Bonetti, I.M. Roos, L. Milani, C. Wang, et al. Interferon regulatory factor 5 (IRF5) gene variants are associated with multiple sclerosis in three distinct populations. J Med Genet. 2008;45:362-369
  • [55] V. Dideberg, G. Kristjansdottir, L. Milani, C. Libioulle, S. Sigurdsson, E. Louis, et al. An insertion-deletion polymorphism in the interferon regulatory factor 5 (IRF5) gene confers risk of inflammatory bowel diseases. Hum Mol Genet. 2007;16:3008-3016
  • [56] S. Sigurdsson, H.H. Göring, G. Kristjansdottir, L. Milani, G. Nordmark, J.K. Sandling, et al. Comprehensive evaluation of the genetic variants of interferon regulatory factor 5 (IRF5) reveals a novel 5 bp length polymorphism as strong risk factor for systemic lupus erythematosus. Hum Mol Genet. 2008;17:872-881
  • [57] K. Vandenbroeck, I. Alloza, B. Swaminathan, A. Antigüedad, D. Otaegui, J. Olascoaga, et al. Validation of IRF5 as multiple sclerosis risk gene: putative role in interferon beta therapy and human herpes virus-6 infection. Genes Immun. 2011;12:40-45
  • [58] S. Vosslamber, L.F. van der Voort, I.J. van den Elskamp, R. Heijmans, C. Aubin, B.M. Uitdehaag, et al. Interferon regulatory factor 5 gene variants and pharmacological and clinical outcome of Interferonβ therapy in multiple sclerosis. Genes Immun. 2011;12:466-472
  • [59] P.L. De Jager, X. Jia, J. Wang, P.I. de Bakker, L. Ottoboni, N.T. Aggarwal, et al. Meta-analysis of genome scans and replication identify CD6, IRF8 and TNFRSF1A as new multiple sclerosis susceptibility loci. Nat Genet. 2009;41:776-782
  • [60] International Multiple Sclerosis Genetics Consortium. The genetic association of variants in CD6, TNFRSF1A and IRF8 to multiple sclerosis: a multicenter case–control study. PLoS One. 2011;6:e18813
  • [61] R. Gross, B.C. Healy, S. Cepok, T. Chitnis, S.J. Khoury, B. Hemmer, et al. Population structure and HLA DRB1 1501 in the response of subjects with multiple sclerosis to first-line treatments. J Neuroimmunol. 2011;233:168-174
  • [62] S. Malhotra, C. Morcillo-Suárez, R. Nurtdinov, J. Rio, E. Sarro, M. Moreno, et al. Roles of the ubiquitin peptidase USP18 in multiple sclerosis and the response to interferon-β treatment. Eur J Neurol. 2013;20:1390-1397
  • [63] F. Gilli, F. Marnetto, M. Caldano, A. Sala, S. Malucchi, A. Di Sapio, et al. Biological responsiveness to first injections of interferon-beta in patients with multiple sclerosis. J Neuroimmunol. 2005;158:195-203
  • [64] B. Weinstock-Guttman, M. Tamaño-Blanco, K. Bhasi, R. Zivadinov, M. Ramanathan. Pharmacogenetics of MXA SNPs in interferon-beta treated multiple sclerosis patients. J Neuroimmunol. 2007;182:236-239
  • [65] C. López-Gómez, A. Pino-Ángeles, T. Órpez-Zafra, M.J. Pinto-Medel, B. Oliver-Martos, J. Ortega-Pinazo, et al. Candidate gene study of TRAIL and TRAIL receptors: association with response to interferon beta therapy in multiple sclerosis patients. PLoS ONE. 2013;:e62540 10.1371/journal.pone.0062540
  • [66] A. Rauch, Z. Kutalik, P. Descombes, T. Cai, J. Di Iulio, T. Mueller, et al. Genetic variation in IL28B is associated with chronic hepatitis C and treatment failure: a genome-wide association study. Gastroenterology. 2010;138:1338-1345
  • [67] L.E. Martínez-Gómez, N.C. Chávez-Tapia, A.I. Burguete-García, N. Aguilar-Olivos, V. Madrid-Marina, M. Román-Bahena, et al. IL28B polymorphisms predict the response to chronic hepatitis C virus infection treatment in a Mexican population. Ann Hepatol. 2012;11:876-881
  • [68] S. Malhotra, C. Morcillo-Suárez, D. Brassat, R. Goertsches, J. Lechner-Scott, E. Urcelay, et al. IL28B polymorphisms are not associated with the response to interferon-β in multiple sclerosis. J Neuroimmunol. 2011;239:101-104
  • [69] E. Byun, S.J. Caillier, X. Montalban, P. Villoslada, O. Fernández, D. Brassat, et al. Genome-wide pharmacogenomic analysis of the response to interferon beta therapy in multiple sclerosis. Arch Neurol. 2008;65:337-344
  • [70] M.D. Cénit, F. Blanco-Kelly, V. de las Heras, M. Bartolomé, E.G. de la Concha, et al. Glypican 5 is an interferon-beta response gene: a replication study. Mult Scler. 2009;15:913-917
  • [71] M. Comabella, D.W. Craig, C. Morcillo-Suárez, J. Río, A. Navarro, M. Fernández, et al. Genome-wide scan of 500,000 single-nucleotide polymorphisms among responders and nonresponders to interferon beta therapy in multiple sclerosis. Arch Neurol. 2009;66:972-978
  • [72] C.X. George, Z. Gan, Y. Liu, C.E. Samuel. Adenosine deaminases acting on RNA, RNA editing, and interferon action. J Interferon Cytokine Res. 2011;31:99-117
  • [73] Z. Feng, R. Prentice, S. Srivastava. Research issues and strategies for genomic and proteomic biomarker discovery and validation: a statistical perspective. Pharmacogenomics. 2004;5:709-719
  • [74] G. Zou, H. Zhao. The impacts of errors in individual genotyping and DNA pooling on association studies. Genet Epidemiol. 2004;26:1-10
  • [75] J.C. Barrett, L.R. Cardon. Evaluating coverage of genome-wide association studies. Nat Genet. 2006;38:659-662
  • [76] M.B. Bornstein, A. Miller, S. Slagle, M. Weitzman, H. Crystal, E. Drexler, et al. A pilot trial of Cop 1 in exacerbating-remitting multiple sclerosis. N Engl J Med. 1987;317:408-414
  • [77] G. Comi, M. Filippi, J.S. Wolinsky. European/Canadian multicenter, double-blind, randomized, placebo-controlled study of the effects of glatiramer acetate on magnetic resonance imaging—measured disease activity and burden in patients with relapsing multiple sclerosis. European/Canadian Glatiramer Acetate Study Group. Ann Neurol. 2001;49:290-297
  • [78] R. Aharoni. The mechanism of action of glatiramer acetate in multiple sclerosis and beyond. Autoimmun Rev. 2013;12:543-553
  • [79] O. Neuhaus, C. Farina, H. Wekerle, R. Hohlfeld. Mechanisms of action of glatiramer acetate in multiple sclerosis. Neurology. 2001;56:702-708
  • [80] I. Grossman, N. Avidan, C. Singer, D. Goldstaub, L. Hayardeny, E. Eyal, et al. Pharmacogenetics of glatiramer acetate therapy for multiple sclerosis reveals drug-response markers. Pharmacogenet Genomics. 2007;17:657-666
  • [81] E.Y. Tsareva, O.G. Kulakova, A.N. Boyko, S.G. Shchur, D. Lvovs, A.V. Favorov, et al. Allelic combinations of immune-response genes associated with glatiramer acetate treatment response in Russian multiple sclerosis patients. Pharmacogenomics. 2012;13:43-53
  • [82] S. Dhib-Jalbut, R.M. Valenzuela, K. Ito, M. Kaufman, M.A. Picone, S. Buyske, et al. HLA DR and DQ alleles and haplotypes associated with clinical response to glatiramer acetate in multiple sclerosis. Mult Scler Relat Disord. 2013;4:340-348
  • [83] H. Butzkueven, J. Chapman, E. Cristiano, F. Grand'Maison, M. Hoffmann, G. Izquierdo, et al. MSBase: an international, online registry and platform for collaborative outcomes research in multiple sclerosis. Mult Scler. 2006;12:769-774
  • [84] K. Vandenbroeck, M. Comabella, E. Tolosa, R. Goertsches, D. Brassat, R. Hintzen, et al. United Europeans for development of pharmacogenomics in multiple sclerosis network. Pharmacogenomics. 2009;10:885-894


School of Pharmacy and Medical Sciences, University of South Australia, Adelaide, Australia

lowast Corresponding author at: School of Pharmacy and Medical Sciences, University of South Australia, Adelaide 5000, Australia. Tel.: + 61 8 83022489; fax: + 61 8 83022389.