Multiple Sclerosis: genetics, biomarkers, treatments

Abstract

Purpose of review

We discuss new paradigms for understanding the immunopathology of MS through the recent development of high throughput genetic analysis, emergence of numerous candidate biomarkers and the broadening of the treatment arsenal.

Recent findings

The recent use of genome wide association studies provide new tools for a better understanding of Multiple Sclerosis etiology. GWASs have identified many genes implicated in immune regulation and the next step will be to elucidate how those genetic variations influence immune cells function to drive disease development and progression. Furthermore, patient care has seen the emergence of new biomarkers for monitoring disease progression and response to treatment. Finally, the introduction of numerous immunomodulatory treatments will likely improve clinical outcome of MS patients in the future.

Summary

breakthroughs in the field of Multiple Sclerosis have led to a better understanding of the physiopathology of the disease, follow up and treatment of the patients that develop relapsing remitting MS. The next challenge for MS will be to press forward to model and decipher MS progression, which will help both to develop therapeutics and generate knowledge about mechanisms of neurodegeneration.

Introduction

MS is a multifocal demyelinating disease with progressive neurodegeneration caused by an autoimmune response to self-antigens. Clinical symptoms vary based on the site of neurologic lesions and often correlate with invasion of inflammatory cells across the blood-brain barrier with resulting demyelination and edema [1] (figure 1). Development of MS is the result of both genetic predisposition and environmental triggers. In this review we discuss new paradigms for understanding the immunopathology of MS through the recent development of high throughput genetic analysis, emergence of numerous candidate biomarkers and the broadening of the treatment arsenal.

Figure 1.

Defects in peripheral immune regulation lower the activation barrier for autoreactive T cells. (A) In normal homeostasis, APCs digest microbial antigens or self proteins and present them to naive T cells in the context of co-stimulatory molecules. An appropriate cytokine milieu can drive differentiation of these naive autoreactive T cells to a Th1 or Th17 cell phenotype; however, these potentially pathogenic T cells are not activated due to the actions of peripheral regulatory immune cell populations, such as FoxP3+ Tregs and Tr1 cells. Via the actions of co-inhibitory molecules and cytokines such as IL-10 and TGF-β, autoreactive T cells become anergic and autoimmune disease is prevented. Other mechanisms, such as thymic deletion and lack of co-stimulatory molecules on APCs, are also involved in controlling autoreactive T cells. (B) MS patients have defects in peripheral immune regulation, including higher expression of co-stimulatory molecules on APCs, lower CTLA-4 levels, and lower IL-10 production. Additionally, MS patients have an increased frequency of IFN-γ–secreting Tregs relative to healthy controls. Thus, the barrier for activation of autoreactive T cells is lowered for MS patients. Activated myelin-reactive T cells can then adhere to and extravasate across the choroid plexus and BBB, where they can initiate an inflammatory milieu that gives license to further waves of inflammation and eventual epitope spreading. Reproduced with permission from [1]

Genetics

1. The GWAS era

Genome-wide association studies (GWASs) using SNPs from the HapMap project allowed the use of an unbiased approach in scanning the whole genome and identifying SNPs associated with disease [2–5]. The MHC region was quickly identified as associated to MS susceptibility due to larger odd ratios [6]. Nine GWASs and a meta-analysis were performed and, in total, identified approximately 14 regions with genome-wide significance [7–14], including several previously identified associations [7, 15]. A GWAS with over 9,000 cases of MS replicated many of these previously suggested associations and identified an additional 29 novel susceptibility loci [16]. GWAS have now defined 194 genetic variants that are associated with MS, with that number likely to rise to over 400 (IMSGC, unpublished data). To date, approximately 1000 patients with progressive MS have been GWAS, and those subjects cannot be differentiated from patients with relapsing remitting MS. However, studies are planned to investigate larger cohorts of patients with PPMS.

2. Refinement of the GWASs discoveries

An inherent difficulty in GWAS studies is that linkage disequilibrium (LD) in the risk loci makes it difficult to pinpoint the most-likely causative variant. Currently, multiple fine-mapping efforts are underway to determine the most-likely causative SNP in many of these regions, but these have required novel tools to both fine-map and replicate these loci.

Having variants spread across the entire genome improves the breadth of the screen, but makes identifying individual variants within a region difficult. The Immunochip was designed to have dense clusters of variants around the SNPs called for by the original GWAS studies, thereby allowing close interrogation of all autoimmunity associated loci to date [17]. Fine-mapping the variants that were originally identified by GWAS identified 48 novel loci that had not reached the GWAS significance cutoff, but did so with the more targeted immunochip [18]. In addition to the Immunochip, a second targeted Illumina chip, the “MS chip” was developed to replicate recent GWAS hits. The MS chip also allows for identification of rare variants associated with MS that might have a high odds ratio and impact on disease.

As the Immunochip provided dense genotyping of MS associated Loci, this allows for an unbiased screen for most-likely causal SNPs for MS. The Wellcome trust pioneered a Bayesian analysis that was able to determine credible SNPs from three different diseases [19]. We have recently developed a new algorithm for fine mapping causal variants based on genetic evidence [20]. This model, called Probabilistic Identification of Causal SNPs (PICS), is a Bayesian algorithm modeled on dense genotyping data from the recent Immunochip study of MS [18]. Using the PICS algorithm, we were able to predict causal variants for GWAS data even when dense genotyping data was not available based on imputation to the 1000 genomes project [21].

3. From genetics to functional immunology

The next step upon us is characterizing phenotypic differences associated with variants in disease states. Several allelic variants in genes for cytokine receptors and co-stimulatory molecules have been associated with defects in Treg homeostasis. Our group and others have previously described that despite normal frequency of Tregs in MS patients compared to healthy controls, there is a loss of functional suppression by Tregs [22, 23]. Treg functional plasticity may also contribute to the autoimmune disease, as several studies have demonstrated that Tregs can produce inflammatory cytokines under certain conditions [24–27].

While many allelic variants associated with disease have been described, each variant alone carries a small increase in disease risk. As such, it is likely that multiple variants together represent a cumulative burden greater than the individual variants alone, falling within a limited number of signalling cascades primarily associated with immune responses and cytokine signalling [16]. In particular, variants within the NFκB signaling cascade as well as the STAT3/4/5 signalling cascades are highly represented in MS. It has been shown that CD4 cells from RRMS patients exhibit altered STAT3 signalling after IL6 stimulation [28]. In addition, we have demonstrated that naïve CD4 cells from MS patients exhibit constitutive activation of p65 NFκB [29]. This pathway is of special interest for MS as it is involved in both inflammation and neurodegeneration processes, thus possibly contributing to long-term disease progression [30].

4. Genetics, disease progression, and response to treatment

GWAS only identify variants associated with susceptibility to disease. The next question in MS genetics is determining what genetic variants are also associated with progression, severity, and response to treatment. A recent study compares genetic variants to CSF IgG levels and oligoclonal bands (OCBs) in 6950 MS patients from nine countries. Two regions, the MHC locus and the immunoglobulin heavy chain locus (IGHC), showed significant association to both IgG index and OCBs [31]. This confirmed a number of earlier, small cohort studies suggesting association of MHC and IGHC loci in CSF IgG and OCBs [32–38].

Biomarkers

1. Biomarkers of disease conversion

The McDonald criteria for MS diagnosis require two or more clinical episodes with two or more lesions on MRI appearing in separate loci over time [39–41]. Patients with clinically isolated syndrome (CIS) have a variable chance of conversion to relapsing remitting MS (RRMS) depending upon the MRI lesion load [42]. Some patients progress from RRMS to irreversible progressive disability called secondary progressive MS (SPMS). Approximately 10–15% of patients exhibit progressive disease after initial symptoms without relapses termed primary progressive MS (PPMS) [1]. It has been shown that early therapeutic intervention delays long-term disease progression [43, 44]. As such, determining those individuals presenting with optic neuritis or CIS with high risk of developing RRMS, SPMS, or PPMS would allow earlier treatment and improved outcomes. The presence of gadolinium enhancing lesions at CIS and OCBs in the CSF has been shown to be predictive biomarkers of subjects with high risk of progression to MS.

2. Biomarkers currently in clinical use

The presence of OCBs and high IgG levels in the CSF of MS patients was first shown in 1957 [45, 46]. Recently, renewed interest has been sparked in OCBs as the presence of OCBs in CSF is predictive of conversion from CIS to MS [47, 48]. As such, while it is no longer required for diagnosis of MS, testing for OCBs still represents a useful tool for ruling out other possible diagnoses and for prognostication of CIS conversion [49].

While it is clear that white matter lesions on MRI are indicative of progression from CIS to a clinically defined MS, the correlation between clinical disability (as measured by EDSS) and T2-weighted white matter lesion load varies broadly between different studies [50–53]. Presence of Gadolinium-enhancing lesions on magnetic resonance Imaging (MRI) in MS is indicative of active inflammation and lesion burden [54, 55]. Number and size of enhancing MRI lesions are predictive of both onset and severity of relapses [54–57], however there is a weak or absent correlation between Gadolinium-enhancing lesions and cognitive decline in RRMS [58]. More recently, studies have shown a positive correlation between overall grey matter atrophy and cognitive dysfunction suggesting that grey matter atrophy, rather than white matter lesion load, may be a useful biomarker for prediction of clinical severity [59].

Natalizumab (Tysabri) is a monoclonal antibody targeted to the α4 integrin. Blockade of α4β1 results in diminished T cell trafficking to the CNS and reduces relapse rate by 68% [60]. However, progressive multifocal leukoencephalopathy (PML) emerged as a rare adverse event from natalizumab treatment, generally occurring late (N 24 months) after initiating treatment [61]. PML is caused by reactivation of a latent JC virus in immunocompromised individuals. This leads to a debilitating encephalopathy that is often fatal. Previous infection with the JC virus can now be determined by seropositivity for JC viral antibodies prior to initiating and during natalizumab treatment. Only two cases of PML have been reported that were seronegative for JCV antibody [61, 62], making JCV antibodies an extremely useful clinical biomarker for assessing the risk of PML. However, while 50% of MS patients [62, 63] are JCV Ab seropositive, less than 1% will develop PML [64], obviating the need for more specific predictive markers of PML.

3. Potential biomarkers

CNS neurofilaments (Nfl) are released after axonal damage. Both the heavy chain and light chains are associated with axonal damage in MS and interestingly NF-M has been shown to be targeted in the CNS during EAE, the mouse model for MS [65, 66]. Recent studies have demonstrated that Nfl levels in CSF are increased in both RRMS and progressive MS compared to healthy control subjects [67]. In patients with RRMS, Nfl is increased at all disease stages, but fluctuates consistent with clinical course and the presence of active lesions by MRI [68]. Several studies suggest that Nfl levels could be a prognostic biomarker for an aggressive disease course and high risk for secondary progression [69–72] and correlate with treatment response to Fingolimod, natalizumab, and rituximab [43, 73, 74].

Other markers of neuronal and glial cell damage have been shown to be elevated in MS patients compared to healthy controls. Glial fibrillary acidic protein (GFAP), myelin basic protein (MBP), S100β (an astrocyte proliferation marker), tau, NCAM, NGF, CNTF and ferritin expression in the CSF have also been suggested as potential biomarkers [75–79]. However, they are non-specific and in the modern era with MRI, they are not useful as biomarkers.

CD163 is a monocyte/macrophage specific membrane marker. Upon activation, macrophages cleave CD163 from the surface and shed soluble CD163 (sCD163) that can be detected in the blood and CSF [80, 81]. Two studies demonstrated increased sCD163 levels in the blood [82] and the CSF of MS patients [83]. YKL-40 (Chitinase-3-like 1) is an activation marker for glia, macrophages, vascular smooth muscle cells, airway epithelia, and chondrocytes [84]. Elevated YKL-40 levels have been found in the serum of many inflammatory conditions [85–87]. In MS, YKL-40 levels in the CSF were found to be significantly higher in CIS patients that converted to MS as compared to patients that remained as CIS and has been correlated with shorter time to MS conversion and more rapid progression [72, 84, 88].

Osteopontin (OPN) is an early activation marker on T cells with a role in T cell costimulation and IFNγ expression [89]. OPN is highly expressed within MS lesions [89, 90] and is significantly higher in MS blood and CSF than healthy controls. Over the course of a five years follow up study, OPN correlated to disease severity and relapse rate [91–95]. High levels of OPN in the CSF also correlated to disease severity in PPMS [96]. However, there is disagreement on OPN levels as a prognostic biomarker of disease severity as two studies in MS have not found this association [97, 98].

Both CD4 and CD8 T cells are present in MS lesions and are believed to play a central role in disease development. Identification of a single inciting antigen triggering activation of myelin reactive T cells may not be possible, as the original targeted antigens may be unique to each patient and evolve throughout disease progression due to epitope spreading. Though, myelin-reactive T cells were observed repeatedly in MS patients and determining the inflammatory profiles of those cells may represent a biomarker for disease and disease progression. We and others originally described increased frequencies of myelin-reactive CD4 and CD8 cells in the peripheral blood from MS patients [99–101]. We have identified a phenotypic and transcriptional profile of myelin reactive T cells unique to MS patients compared to healthy donors [102, 103] suggesting the possibility that unique, MS-specific profiles of T cell libraries may identify individuals that will convert from initial diagnosis and CIS to MS.

Treatments

1. The first treatments

In 1993, FDA approved IFN-β1b to treat relapsing forms of MS [104, 105]. These agents reduce disease severity quite well for some patients, and might delay progression and improve survival [106–109]. Long-term follow-up of patients in a pivotal trial of intramuscular IFN-β1a revealed a poor prognosis for patients who exhibited active radiographic disease during the two years on drug [110]. This finding raised the hypothesis that patients with MS might exhibit a similarly mixed response to therapies as do patients with a range of other autoimmune disorders. Observational studies using different IFN-β preparations confirmed that this effect was shared among all members of the IFN-β drug class [111]. Yet, IFN-β therapy in SPMS patients fails to alter disease progression [112].

For glatiramer acetate, distinctions between good responders and poor responders were not as clearcut as with IFN-β [113]. Interferons are endogenous regulatory cytokines that increase or decrease transcriptional initiation for hundreds of genes in a cell-type-dependent fashion [114]. Therefore, in patients with MS who had good or poor responses to treatment, bioinformatic analysis of patterns in interferon-induced gene expression might predict clinical responses to IFN-β. These data could, in turn, yield insights into pathogenic mechanisms [115, 116].

2. Natalizumab and Fingolimod

Natalizumab has undergone a complex, and continuing, process of integration into clinical practice [117, 118]. Administered by monthly intravenous infusion, natalizumab exerts impressive inhibitory effects on the inflammatory aspects of MS, with >65% reduction in relapses during two years of treatment, and >90% suppression of new inflammatory MRI lesions [60, 119], with PML being the major side effect as discussed above [120].

Given the overall rarity of PML, this entirely unexpected complication was clearly caused by natalizumab, provoking immediate voluntary suspension of the drug’s distribution, though eventually no other PML cases were found to be incubating in the study population [121]. The mechanism by which natalizumab causes PML remains unknown [122]. Natalizumab-PML is not associated with generalized immunosuppression, and may be mechanism-driven [123]. The search for host factors that predispose to PML besides JCV infection continues.

The other second-era drug, fingolimod, is a prodrug that is converted in vivo to a sphingosine-1-phosphate (S1P) analogue. Fingolimod downregulates S1P receptor 1 on leukocytes and the endothelium, trapping naive and central memory T lymphocytes in lymph nodes. Treatment with fingolimod thereby suppresses MS disease activity, with 55–60% lower relapse rates and an impressive reduction of MRI-visible activity compared to placebo [124–127]. Fingolimod induces lymphopenia, unassociated with opportunistic infections beyond disseminated herpes zoster. For this reason, patients must have documented varicella zoster virus (VCV) immunity to be considered for fingolimod [128]. Unfortunately, both fingolimod and natalizumab fails to show efficacy in PPMS treatment [129]. These data suggests that while relapsing remitting MS represents primarily an inflammatory process of the adoptive immune system, secondary progressive MS may represent CNS activation of the innate immune system.

3. The future of MS therapy

This period is characterized by the introduction of medications including both small-molecules and biologics. Two oral immunomodulatory medications (teriflunomide and dimethyl-fumarate) have been approved [130–133]. After phase III trials, alemtuzumab—a leukocyte-depleting CD52 antibody —was approved [134–136]. Other agents, including laquinimod [137, 138] and daclizumab (an anti-CD25 antibody)[139–142], ocrelizumab and ofatumumab (both CD20 antibodies targeting B Lymphocytes), are undergoing advanced clinical testing. Recent data suggests the potential usefulness of Ocrelizumab as a first line drug in the treatment of both early RRMS and PPMS [143, 144].

Each agent tested so far in the third era has had a distinct mechanism of action, and yet has shown efficacy in double-blinded controlled trials. The variety of agents shown to reduce relapse frequency and MRI-monitored disease activity is consistent with recent chromatid mapping studies of allelic variants associated with MS. These studies indicated that multiple cell types— including Th17 cells, FOXP3 regulatory T cells, B cells and macrophages—are involved in MS disease pathogenesis [20].

Conclusions

In the past five years, potent insights have been made into the genetic basis of the disease, its environmental associations, its characterization by MRI, and its susceptibility to treatment by immunomodulation. This progress now poises the MS community for advancement to the prized objective of alleviating the burden exerted by progressive MS.

Key points.

• High throughput analysis of the genetic burden in MS have allowed the identification of numerous genes and pathways involved in the regulation of immune functions.

• Beyond MRI, OCBs and JC virus monitoring, new biomarkers are being characterized that will likely become useful in the clinics to follow disease progression and response to treatment.

• Several immunomodulatory molecules with various immune targets have proven successful for MS treatment, broadening the therapeutic arsenal and helping a better understanding of MS physiopathology.