Pharmacogenomics of interferon-β in multiple sclerosis: What has been accomplished and how can we ensure future progress?

Cytokine & Growth Factor Reviews 26 (2015) 249–261

Contents lists available at ScienceDirect

Cytokine & Growth Factor Reviews journal homepage: www.elsevier.com/locate/cytogfr

Survey

Pharmacogenomics of interferon-b in multiple sclerosis: What has been accomplished and how can we ensure future progress? Rebecca J. Carlson a,d, J. Ronald Doucette b,d,e, Katherine Knox c,e, Adil J. Nazarali a,d,e,* a

Laboratory of Molecular Cell Biology, College of Pharmacy and Nutrition, University of Saskatchewan, Saskatoon, SK S7N 5E5, Canada Department of Anatomy and Cell Biology, College of Medicine, University of Saskatchewan, Saskatoon, SK S7N 5E5, Canada c Department of Physical Medicine and Rehabilitation, College of Medicine, University of Saskatchewan, Saskatoon, SK S7N 5E5, Canada d Neuroscience Research Cluster, Health Sciences Building, University of Saskatchewan, Saskatoon, SK S7K 5E5, Canada e Cameco Multiple Sclerosis Neuroscience Research Center, City Hospital, Saskatoon, SK S7K 0M7, Canada b

A R T I C L E I N F O

A B S T R A C T

Article history: Available online 31 October 2014

Multiple sclerosis (MS) is a progressive disorder of the central nervous system, often resulting in significant disability in early adulthood. The field of pharmacogenomics holds promise in distinguishing responders from non-responders to drug treatment. Most studies on genetic polymorphisms in MS have addressed treatment with interferon-b, yet few findings have been replicated. This review outlines the barriers that currently hinder the validity, reproducibility, and inter-study comparison of pharmacogenomics research as it relates to the use of interferon-b. Notably, statistical power, varying definitions of responder status, varying assay and genotyping methodologies, and anti-interferon-b neutralizing antibodies significantly confound existing data. Future work should focus on addressing these factors in order to optimize interferon-b treatment outcomes in MS. ß 2014 Elsevier Ltd. All rights reserved.

Keywords: Multiple sclerosis Pharmacogenomics Interferon-b Neutralizing antibodies Responders

Contents 1. 2. 3. 4. 5. 6. 7. 8. 9. 10.

Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Pharmacogenomics and MS. . . . . . . . . . . . . . . . . . . Methodology . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Strategies in genetic testing . . . . . . . . . . . . . . . . . . Results of GWAS – are they reliable? . . . . . . . . . . . Interferon regulatory factor 5 . . . . . . . . . . . . . . . . . Confounding factors in pharmacogenomic studies. Power and effect size/odds ratio . . . . . . . . . . . . . . . Definition of responder status. . . . . . . . . . . . . . . . . Natural history. . . . . . . . . . . . . . . . . . . . . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

. . . . . . . . . .

250 250 251 251 251 252 252 252 253 254

Abbreviations: AAN, American Academy of Neurology; ADAR, RNA-specific adenosine deaminase; BAb, binding antibodies; CAST, calpastatin; CGAS, candidate gene association study; CHMP, Committee for Medicinal Products for Human Use; CIT, citron (rho-interacting, serine–threonine kinase 21); CIS, clinically isolated syndrome; CNS, central nervous system; COL25A1, Collagen type XXV a1; CPE, cytopathic effect; CXCL-10, C-X-C motif chemokine 10; EDSS, Expanded Disability Status Scale; ELISA, enzymelinked immunosorbent assay; EMA, European Medicines Agency; GA, glatiramer acetate; GPC5, glypican 5; GRIA3, AMPA 3 ionotropic glutamate receptor; GWAS, genomewide association study; HAPLN1, hyaluronan proteoglycan link protein; HERC5, HECT and RLD domain containing protein ligase 5; HLA, human leukocyte antigen; IFNAR, interferon a/b receptor; IFN-b, interferon-b; IL-6, interleukin 6; IL-8, interleukin 8; IL28B, interleukin 28B; IRF5, interferon regulatory factor 5; IRF8, interferon regulatory factor 8; MRI, magnetic resonance imaging; MS, multiple sclerosis; MxA, myxovirus resistance protein A; NAb, neutralizing antibody; NABINMS, Neutralizing Antibodies on Interferon-b in Multiple Sclerosis; NPAS3, neuronal PAS domain protein 3; OR, odds ratio; PCR, polymerase chain reaction; PPMS, primary-progressive multiple sclerosis; PRMS, progressive-relapsing multiple sclerosis; RRMS, relapsing-remitting multiple sclerosis; SNP, single nucleotide polymorphism; SPMS, secondary-progressive multiple sclerosis; STARD13, steroidogenic acute regulatory gene-related lipid transfer domain-containing 13; TLR6, toll-like receptor 6; TNF-a, tumor necrosis factor a; UEPHA*MS, United Europeans for the Development of Pharmacogenomics in Multiple Sclerosis; USP18, ubiquitin specific peptidase 18; ZFAT, zinc finger and autoimmune thyroid hook domain-containing; ZFHX4, zinc finger homeobox 4. * Corresponding author at: Laboratory of Molecular Cell Biology, College of Pharmacy and Nutrition, and Neuroscience Research Cluster, University of Saskatchewan, Saskatoon, SK S7N 5E5, Canada. Tel.: +1 306 966 6334; fax: +1 306 966 6377. E-mail address: [email protected] (A.J. Nazarali). http://dx.doi.org/10.1016/j.cytogfr.2014.10.008 1359-6101/ß 2014 Elsevier Ltd. All rights reserved.

250

11. 12.

13. 14.

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

Study design. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . Neutralizing antibodies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.1. Binding and neutralizing antibodies . . . . . . . . . . . . . . . . . . 12.2. Differences in immunogenicity between formulations . . . . 12.3. Biological and clinical effects . . . . . . . . . . . . . . . . . . . . . . . . 12.4. Detection and measurement of NAbs . . . . . . . . . . . . . . . . . 12.5. Clinical management . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12.6. Pharmacogenomics of neutralizing antibody development Ramifications for pharmacogenomics . . . . . . . . . . . . . . . . . . . . . . . Summary and considerations for future research . . . . . . . . . . . . . Acknowledgements . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

1. Introduction Multiple sclerosis (MS) is a chronic inflammatory disease that involves the demyelination of axons in the central nervous system (CNS). It is unclear what triggers the development of MS, although its pathogenesis has been linked to an abnormal immune cell response toward myelin and oligodendrocytes. It is believed that Th1 T-helper cells permeate the blood–brain barrier, where they activate cytokines that recruit other immune cells that in turn contribute to axonal damage [1]. These mechanisms may result in an increased number and volume of brain lesions on magnetic resonance imaging (MRI), which correlate to some degree with disease progression [2]. Myelin loss and regeneration, axonal degeneration and brain atrophy all contribute to the disease process. In order to meet MS diagnostic criteria, one must have a combination of clinical and/or MRI findings to support the dissemination of lesions within the CNS over time [3]. The rates of disability progression may be highly variable both at the individual level and within larger cohorts [4]. However, life expectancy is reduced by approximately only a decade [5,6]. Individuals may live a significant portion of their adult life with advanced disability and therefore early prevention of disability remains a major challenge and priority. While measuring and tracking disability outcomes may be challenging in MS, historically the most standard measure of disability has been the Expanded Disability Status Scale (EDSS). The EDSS scale is an ordinal scale ranging from zero (no disability) to 10 (death) [7]. MS disease course may be divided broadly into four categories: relapsing-remitting, secondary-progressive, primary-progressive, and progressive-relapsing. Relapsing-remitting (RRMS) is the most common presenting course and is characterized by recurrent relapses that are interspersed with periods of partial or complete symptom resolution. The term clinically isolated syndrome (CIS) has been coined to describe persons who have experienced only one clinical demyelinating relapse. The majority of persons with CIS will eventually meet diagnostic criteria for RRMS and the majority of individuals with RRMS transition to a secondaryprogressive (SPMS) course of the disease over time. In SPMS, relapses may cease altogether or become less frequent, recovery is usually incomplete if relapses do occur and disability accumulation occurs independent of relapses. Approximately 20% of patients present with a primary-progressive (PPMS) course accumulating permanent disability from their first symptom onset in the absence of relapses [8]. An even smaller percentage of people (5%) experience disability progression from MS onset as well as superimposed relapses and are categorized as a progressive-relapsing disease (PRMS) course. In higher-risk areas such as Scandinavia, Iceland, and North America, the incidence of MS is significant, where it occurs in approximately 1–2 in 1000 [9]. It is recognized that treatment optimization in MS is challenged not only by pharmacogenomics factors, but also by other factors including practice patterns,

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

. . . . . . . . . . . .

254 254 254 254 254 254 255 255 256 257 258 258

funding for treatment, access to care and limitations imposed by treatment side effects [10]. While all these factors are relevant, pharmacogenomics may offer a partial solution to some of these challenges by identifying those patients who are most likely to be non-responders to a specific treatment early in the disease course. Although regional differences may exist, available disease modifying therapies approved for the treatment of MS include injectable interferons (IFN-b1a, IFN-b1b), glatiramer acetate, intravenous natalizumab, and more recently oral fingolimod, teriflunomide and dimethyl fumerate. All of these treatment options are indicated for use in patients with the relapsing form of MS, since to-date there is insufficient evidence these treatments are effective for progressive forms of the disease. However, even among those on treatment with relapsing MS, up to 49% of patients being treated with IFN-b fail to respond to therapy [11]. 2. Pharmacogenomics and MS Due to the relatively high incidence of MS in certain countries and the risk of unnecessarily exposing patients to serious side effects, genetic determinants of treatment success with diseasemodifying therapies could have a significant clinical impact [12,13]. IFN-b was the first disease-modifying therapy approved for the treatment of MS and hence has been widely utilized. The discovery of a biomarker or genetic polymorphism could potentially predict those in whom IFN-b is less likely to be effective. A number of genes inside and outside the human leukocyte antigen (HLA) locus have been investigated in smallscale linkage studies, and polymorphisms in one gene in particular have been independently verified based on large-scale genomewide association studies (GWAS) [14,15]. These studies have brought researchers closer to identifying genetic polymorphisms that may be tested in practice to predict patient response. However, important confounding factors, especially the influence of neutralizing antibodies (NAb), have not been consistently controlled for. Such confounding factors should be accounted for when interpreting patient outcomes related to pharmacogenomic data in patients being treated with disease-modifying therapies such as IFN-b (Fig. 1). The objective of this review is to determine what can be learned from existing pharmacogenomics studies in order to guide future research direction so as to enhance the value of pharmacogenomics in predicting treatment responses in MS. Several studies have been undertaken to identify variances in gene polymorphisms and gene expression related to IFN-b response. These have explored the baseline expression of cytokines, particularly the inflammatory mediator interleukin 8 (IL-8). Type 1 IFN pathway activation has also been associated with IFN-b response; those that overexpress type 1 IFN-induced genes were associated with an increase in relapses while on IFN-b, which may suggest that the efficacy of IFN-b is reduced in such patients ([16,17]; reviewed in Verweij and Vosslamber [18]). Studies focusing on gene expression profiles and their influence on IFN-b

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

251

Fig. 1. Sources of confounding factors in current IFN-b pharmacogenomics studies. Numerous factors may mask the association between a genetic polymorphism and response to treatment: differing methods of statistical analysis affect the strength of associations and poor statistical power may result in type II errors; inconsistencies in assay and study designs, including genotyping methods and number of patient cohorts interfere with cross-study comparability; neglecting to control for neutralizing antibodies may lead to the discovery of a false association between a polymorphism and non-response; non-standardized criteria of response and non-response lead to differing proportions of patients categorized as responders or non-responders and hinders cross-study comparison; and patients with naturally more benign disease courses may be falsely labeled as treatment responders and their polymorphisms may be inappropriately associated with IFN-b response.

response are reviewed in more detail elsewhere [19]. While the potential for screening transcriptional expression profiles prior to IFN-b therapy may be promising, this review focuses on genetic polymorphisms associated with IFN-b therapy. 3. Methodology A systematic literature search of available information published in English up to August 2014 was performed using Medline, Embase, and Scopus databases. Key words used in the search were multiple sclerosis, pharmacogenomics, pharmacogenetics, personalized medicine, interferon-beta, Rebif, Avonex, Betaseron, Betaferon, Extavia, neutralizing antibodies, and immunogenicity. In addition, we reviewed relevant sources referenced in various articles’ bibliographies. 4. Strategies in genetic testing Past trials have used two general strategies to identify polymorphisms that may be associated with treatment success. Many researchers have relied on testing genes that were hypothesized to play a role in IFN-b’s mechanism of action. These smaller, more focused candidate gene association studies (CGAS) are limited by the technology’s restricted coverage of the genome, as well as by what is currently known about IFN-b’s effects and its downstream signaling genes [20]. As gene chip technology has advanced, researchers were able to expand the scope of their investigations to the entire genome using gene chip microarrays, which has facilitated the discovery of new and unexpected gene polymorphisms. To date, two GWAS have been performed, using 100 K and 500 K single nucleotide polymorphism (SNP) arrays,

respectively [14,21]. The results of these studies have identified several new SNPs, but also failed to detect many of those implicated in previous CGAS. The lack of reproducibility between previous studies challenges the validity of these findings, and highlights the need for standardization in future protocols in order to detect true polymorphisms that may predict response to IFN-b. 5. Results of GWAS – are they reliable? The first GWAS revealed new potential associations between genes encoding for cell surface receptors and ion channels that are highly concentrated in the CNS [14]. Byun and colleagues trialed a cohort of 206 patients between four centers in southern Europe, who were treated with any one of three IFN-b formulations: IFNb1a (Rebif, Pfizer Inc.; or Avonex, Biogen Idec), or IFN-b1b (Betaseron, Bayer Inc.). They recorded subjects’ relapse rate using clinical and radiological data over at least two years, and applied patients’ pooled DNA samples to an Affymetrix 100 K gene chip. Top-scoring SNPs were further validated using a TaqMan1 assay and analyzed for gene ontology. The strongest associations were found in genes encoding for hyaluronan proteoglycan link protein (HAPLN1), collagen type XXV a1 (COL25A1), calpastatin (CAST), TAFA1, neuronal PAS domain protein 3 (NPAS3), LOC442331, and glypican 5 (GPC5). Interestingly, the authors performed additional tests to verify previous CGAS results. Only fifteen of 112 polymorphisms were confirmed, although they note that their results were not directly comparable as a result of using different microarrays across studies. The association with GPC5 is of note, as it has been implicated in MS susceptibility polymorphism studies as well as in the pharmacogenomic GWAS [22]. The GPC5 gene codes for glypican, a type of heparan sulfate proteoglycan that fulfills

252

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

signaling functions in the extracellular matrix. Although its exact mechanisms of action are still unknown, glypicans have been shown to contribute to neuronal development and function, and are reviewed more thoroughly in Vandenbroeck et al. [19]. A follow-up study successfully reproduced the connection between GPC5’s SNPs and IFN-b responsiveness, which establishes it as currently one of the most promising candidates for future clinical testing [15]. The results of the second GWAS similarly discovered a relationship between IFN-b response in MS patients and ion receptor channel proteins [21]. Like the previous GWAS, European RRMS patients treated with IFN-b were followed for at least two years and classified into responder or non-responder status. One hundred six DNA samples were genotyped with a 500 K Affymetrix array, in contrast to the 100 K array used in the Byun et al. [14] study and the top 2% of identified SNPs underwent gene ontology analysis. The top-scoring SNPs were further tested in an independent cohort of 94 participants. No SNPs reached significance at the corrected p-values. Eighteen SNPs were identified at uncorrected p-values, seven of which were situated within genes for AMPA 3 ionotropic glutamate receptor (GRIA3, in women only), citron (rho-interacting, serine–threonine kinase 21; CIT), RNAspecific adenosine deaminase (ADAR), interferon receptor 2 (IFNAR2), zinc finger and autoimmune thyroid hook domaincontaining (ZFAT), zinc finger homeobox 4 (ZFHX4) and steroidogenic acute regulatory gene-related lipid transfer domain-containing 13 (STARD13). The GRIA3 gene encodes for an AMPA-type glutamate receptor, a key component in neuronal excitation, while the CIT gene encodes for a tyrosine kinase that also functions in gaminobutyric acid and glutamatergic signaling pathways. While the first GWAS also identified SNPs within genes encoding for ion channels and signal transduction pathways, the second GWAS did not find any specific SNPs in common with the first. The role of IFNAR2, which encodes a type 1 interferon receptor subunit, also remains unclear. As IFN-b interacts with the IFNAR receptors as part of its biological mechanism, it has been targeted by several CGAS [23–26]. These CGAS did not find a correlation between IFNAR2 and IFN-b response in MS patients, which is in disagreement with Comabella et al.’s [21] findings. The fact that a significant number of independent studies have failed to show consistent evidence regarding a single SNP further indicates that underlying factors are likely confounding these findings.

protein, these patients may experience increased baseline activity of the IFN pathway and decreased sensitivity to IFN-b treatment [28]. This hypothesis is consistent with previous transcriptional expression profiling studies [16,17]. Vandenbroeck et al. [29] examined SNPs in IRF5 as a result of their association with MS susceptibility. Their primary objectives were to validate the susceptibility evidence for MS as well as viral infections, and to determine if SNPs were also involved in IFN-b response. Two SNPs, including rs4728142 from the previous study, were tested for IFN-b response in RRMS patients treated over the course of two years. The definition of response differed from the previous study, where patients were classified as responders as long as they did not experience relapse or show an increase in their EDSS score over the study period. Both SNPs showed consistent changes toward an association with responder status, but neither reached statistical significance [29]. It is interesting to note that the same SNP, rs4728142, was associated with positive responder status in the work of Vandenbroeck et al. [29], but with nonresponders in the study of Vosslamber et al. [28]. Also, in a recent study, the risk of relapse or disease progression were not associated with the gene variants in IRF5 [30]. Clearly, it is difficult to draw a meaningful conclusion from studies that have differing study designs. However, it seems that a weak association between IRF5 gene variants and response to IFN-b may exist and there is need for further study to elucidate the nature of IRF5’s effect.

6. Interferon regulatory factor 5

Two easily identifiable factors determining the statistical significance of a study’s findings are statistical power and effect size. In the context of the present review, statistical power is the long-term probability of detecting true differences in IFN-b responder status between study participants who possess single or clusters of nucleotide polymorphisms. The power of a study is affected primarily by the size of the effect, the a level of significance, and the sample size. Cohen [31] recommends using a power of no less than 0.80, a value with an acceptably low risk of making a Type II error but also not so large that it would require a prohibitively large sample size to achieve a desired effect. Effect size reflects the magnitude of the difference between experimental and control dependent variables. Whereas the p-value shows the statistical significance of the effect, the effect size tells you the magnitude of the effect the treatment has had on the dependent variable. Knowing the effect size facilitates the determination of the substantive significance of the findings and, more importantly, allows these findings to be quantitatively compared across different study designs. The odds ratio (OR), as used in the context of studies on SNPs, is an example of an effect size index. Simply defined, the OR represents the odds that the patient responds to IFN-b treatment given the presence of a SNP, compared to the odds of the patient being a responder in the absence of the same SNP. An

Two recent independent CGAS have investigated polymorphisms in the interferon regulatory factor 5 (IRF5) gene. This gene codes for a member of the IRF transcription factor family, which regulates the type 1 interferon pathway. The IRFs are linked to cell growth and carcinogenesis as well as various immune functions. The IRF5 factor in particular is a key regulator of inflammatory cytokine production such as interleukin 6 (IL-6) and tumor necrosis factor a (TNFa), and its polymorphisms have also been attributed to the development of systemic lupus erythematosus [27]. As mentioned earlier, the type 1 IFN pathway has been the focus of several CGAS, however Vosslamber et al. [28] and Vandenbroeck et al. [29] were the first to independently find polymorphisms within the same gene. Vosslamber et al. [28] analyzed four gene variants of IRF5 by comparing independent cohorts of 30–261 RRMS patients’ in vivo induction of 11 IFN-response genes, incidence of new lesions on MRI imaging scans, and time-to-first-relapse while receiving IFNb. The SNP rs2004640 surpassed statistical significance in all cohorts, while rs4728142 only met significance in the pharmacological expression test. The authors speculate that since rs2004640 polymorphisms may lead to increased translation of the IRF

7. Confounding factors in pharmacogenomic studies It may be reasonable to assume that many of the key polymorphisms predicting response to IFN-b would have been uncovered given the number of pharmacogenomic studies that have been conducted. However, many SNPs that showed promise in certain studies failed to be replicated in others. To date, potentially only two genes, GPC5 and to some extent, IRF5, have been reliably identified. The disagreement between studies trivializes previous efforts and hinders the direction of future research. For this reason, it is now critical for a standardized method to be adopted in order to rely upon the findings of any further investigations. Considerations for a method such as this are described below. 8. Power and effect size/odds ratio

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

OR >1 can be interpreted as there being an increased risk of disease or increased responsiveness to treatment associated with SNP. If the OR is <1, then the opposite would be true. In clinical trials, differences in efficacy are often established by recruiting hundreds, and sometimes thousands of participants. Most pharmacogenomics studies, however, have been limited to including only up to a few hundred participants, including validation cohorts. For example, the GWAS by Comabella et al. [21] used 106 and 94 participants in the screening and validation stages, respectively. The GWAS by Byun et al. [14] included 206 subjects in the screening stage, 204 in the validation stage, and 285 subjects in a joint analysis that combined old and new participants. In the screening stage, the authors note that they were able to detect an OR of effect size 2 at a minor allele frequency of 0.1 with 44% power, and thus well below Cohen’s [31] recommended power of 80%, which increased to only 60% in the joint analysis. Although, with larger minor allele frequencies, for example 0.2 and 0.3, power increased to 85% and 99% in the original population and 95% and 99% in the joint population. Unfortunately, many studies fail to mention power used in their studies. Others have acknowledged that their studies likely lacked sufficient statistical power to detect treatment response genes, but did not calculate or report what statistical power their studies had [15,26,32]. It would be appropriate to recommend that studies report actual power in the results and these can be calculated prospectively and retrospectively. Hence, it is difficult to determine if these GWAS failed to replicate previous CGAS because the earlier studies were flawed, or if the GWAS merely lacked statistical power. It is recognized that resources are limited in studies of this nature. However, efforts to recruit as many patients as possible may be achieved through the pooling of clinical trial data and collaborative registries for the purpose of research. The formation of a European collaboration of ten research teams, United Europeans for the Development of Pharmacogenomics in Multiple Sclerosis (UEPHA*MS), demonstrated progress in this regard. One of UEPHA*MS’s objectives was to increase the statistical power of pharmacogenomics studies, which they accomplished by promoting research across multiple countries [33]. The team performed a more highly powered study of 588 MS patients, where they confirmed that SNPs in the interleukin 28B (IL28B) gene were not associated with IFN-b response [34]. More collaborations of this kind may improve the robustness of future data. The use of national registries and clinical trial data allowed Gross and colleagues to retrospectively examine 1119 MS subjects [35], while others have recruited up to 812 patients using CGAS [36]. A recent retrospective analysis of several IFN-b and glatiramer acetate (GA) pharmacogenetic markers showed that individuals possessing a CCR5*w/w genotype or CCR5*w allele with CTLA4*G allele were more likely to exhibit IFN-b non-response, whereas those carrying combinations of CCR5*d, IFNAR1*G, DRB1*15, and TGFB1*T SNPs were more likely to respond to IFN-b than GA [37]. Similar comparative trials between the pharmacogenomics of disease-modifying therapies has the potential to further assist clinicians in choosing between two potential drug regimens for a particular patient. Moving forward, enhanced statistical power will be instrumental in detailing treatment outcomes related to pharmacogenomics. 9. Definition of responder status Another objective of UEPHA*MS was to standardize the definition of response and non-response for the purpose of future studies [33], since a lack of standardized definition has been acknowledged as a major challenge limiting cross-study comparisons [19,26,32,38–40]. MS follows a highly variable disease

253

progression, particularly among those with a relapsing-remitting course. Even small discrepancies between studies on the definition of responder status may significantly impact the proportion of patients classified as responders or non-responders [11], and therefore whether or not a particular SNP is associated with treatment success. Studies have used various definitions and criteria ranging from relapse rate, to sustained change in EDSS scores as well as MRI-related outcomes such as lesion load, number of enhancing lesions, and brain atrophy. Furthermore, thresholds for each of these criteria are inconsistent. For example, the two GWAS studies described above used a combination of relapse number and EDSS to differentiate between responders and nonresponders. However, Byun et al. [14] classified non-responders as those suffering from at least two relapses or an increase in EDSS by one point, while Comabella et al. [21] considered participants as non-responders if they experienced one or more relapses as well as an increase in EDSS by one point that persisted for two visits separated by 6 months. While Comabella and colleagues’ criteria are more stringent, they are more consistent with available data that reliably predicts treatment success to IFN-b [11]. A UEPHA*MS document raises concerns about using MRI criteria because performing scans in larger populations is expensive and may be unduly burdensome for both the researchers and participants. Instead, they recommend studying populations that exhibit extreme phenotypes related to relapse and EDSS measures to ensure specificity in the absence of MRI data [33]. However, data pertaining only to the extreme phenotypes may not be applicable to the general population with MS. A review by Rı´o et al. [41] examined the available evidence of predictors of long-term treatment outcomes, and recommended that both clinical (i.e., relapse rate and EDSS score) and MRI measures should be utilized to define treatment response in a clinical setting. Specifically, Rı´o and colleagues recommend that baseline MRI and clinical assessments should be performed prior to starting therapy, with a follow-up exam 6–12 months later. Patients should be considered potential non-responders when they exhibit 3 new active lesions on MRI and with the occurrence of any relapse or change in the EDSS score. Similarly, the Canadian MS treatment optimization recommendations consider MRI, relapse and EDSS outcomes in assessing response to treatment. However, the recommendations consider multiple definitions for a failed treatment response across a gradient of severity. For example, 3 new lesions even in the absence of a new relapse or EDSS progression may be considered a treatment failure [42]. The definition of treatment failure is likely a moving target as more potent therapies are approved for the treatment of MS and as treatment expectations change over time. Future work evaluating associations between SNPs and treatment response should therefore explore the relationship between SNPs and a wide range of treatment responder definitions. The spectrum might range from so-called disease activity free (no relapses, no new MRI findings and no disability progression) to severe activity in all three of these domains [42]. If the relationship between a SNP and a treatment response holds true across variable definitions for a treatment response, it is more likely to be clinically relevant in its predictive value. Many have used varying combinations of relapse, EDSS and MRI outcome measures already, but only one focused on interferon treatment response and considered all three of these outcomes [43]. This trial, however, only provided short-term outcomes as follow-up was limited to 6 months. Since MS is a lifelong disease and in clinical practice a person may remain on treatment for many years, it is imperative to capture the long-term treatment effects of IFN-b. Two GWAS and three CGAS mentioned above have performed follow-up for at least 2 years, yet there are no data beyond two years assessing the possible predictive value of

254

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

pharmacogenomics. Future pharmacogenomics studies should aim to evaluate a mutually agreed upon range of responder definitions and consider the long-term efficacy over the full course of the disease. Follow-up, even after treatment has been discontinued, should be considered through longitudinal registries since it is possible that SNPs may influence disability outcomes decades later at times when other markers of disease activity, such as MRI and relapses, are known to be less relevant. 10. Natural history Since MS is a heterogeneous disease, an important confounding factor in current pharmacogenomic research is the possibility that patients respond less favorably to IFN-b due to a naturally more aggressive disease course rather than a genetic polymorphism. There is a large reported variability in disease outcomes between individuals, across studies and between populations. For example, mean time to needing a cane has been reported to range between 15 and 32 years [4]. Relapse rates in the placebo arm of clinical trials over time have also varied considerably [44,45]. Unfortunately, implementing a placebo, natural history cohort to mitigate these effects is not feasible due to the ethical implications arising from depriving a patient of effective treatment. Multiple, large treated samples with pharmacogenomic profiles over time may help to control for the factors related to the variable natural history of the disease. 11. Study design The overall study designs of GWAS and CGAS vary greatly between groups, from the number of tested cohorts, to genotyping technology and protocols, to statistical analysis. While more recent research must accommodate more advanced technologies and their protocols, the enormous variability in study design greatly impedes the application of their results to other trials. Independent validation of a group’s results using an identical study protocol may provide a starting point for determining an optimal and easily reproducible design that may be applied to other pharmacogenomics research. 12. Neutralizing antibodies 12.1. Binding and neutralizing antibodies A great deal of evidence is available describing how neutralizing antibodies (NAbs) can influence treatment success or failure. The development of anti-drug antibodies in patients receiving IFN-b therapy was described in the original IFN-b1b clinical trials [46], yet their influence on clinical outcomes has remained controversial. The classification of anti-IFN-b antibodies into two distinct types, binding and neutralizing, further complicates their involvement in disease progression. While up to 80% of patients eventually develop IFN-b binding antibodies (BAbs), they were historically regarded as benign since they did not impair IFN-b biomarker production [47–49]. However, it has become apparent that BAb titer approaches NAb titer as the sensitivity of the assay increases, which indicates that BAbs are simply NAbs with lower binding affinities to IFN-b [50–52]. Recent tests using more sensitive biomarkers suggest that BAbs do in fact inhibit IFN-receptor activity and thus bioavailability of IFN-b [53]. In either case, higher levels of BAbs predict the likelihood of developing higher-affinity IgG1 and IgG4 subtype NAbs that inhibit IFN-b’s ability to interact with its receptor [47,48,53–55]. BAbs are typically quantified using ELISA methodology, while NAbs are characterized by testing for IFN-b-specific activity biomarkers [51].

12.2. Differences in immunogenicity between formulations The three available formulations of IFN-b appear to produce varying amounts of NAbs. Two IFN-b-1a products, Avonex and Rebif, are consistently less immunogenic than the IFN-b1b product Betaseron [49,56–62]. In its clinical trials, Avonex exhibited a NAb incidence of 2–22%, although the maximum incidence decreased to 6% after its formulation and manufacturing process were adjusted. Rebif and Betaseron were associated with 5–28% and 38–47% positivity in their clinical trials, respectively [63]. Some studies show that more patients on IFN-b1b revert back to NAb-negative status vs. IFN-b1a after 2 or more years, but other studies do not support this [64–66]. Regardless, the type of IFN-b does not seem to influence long-term treatment response [67]. Several theories have been proposed to explain the relative increase in immunogenicity between IFN-b1a vs. IFN-b1b. IFN-b1b is not identical in amino acid sequence to endogenous human IFN-b and lacks glycosylation, which may increase its propensity to form aggregates in vivo [68–72]. Pharmaceutical stability has also been implicated to contribute to the relative immunogenicities between products [73]. In addition to product formulations, NAb formation is also affected by dose, frequency, and route of administration, with intramuscular delivery consistently exhibiting fewer antibodies than subcutaneous routes [58,74–80]. 12.3. Biological and clinical effects As previously mentioned, NAbs are believed to abrogate IFN-b therapy as a result of interfering with IFNAR-receptor binding. Studies have repeatedly demonstrated that the presence of NAbs is correlated with decreased measures of IFN-b activity, as quantified by biomarkers including myxovirus resistance protein A (MxA) and corresponding mRNA, viperin, and other markers of IFN-b activity [47,48,53,81–84]. It would therefore seem inevitable that treatment efficacy would suffer as a result of decreased bioavailability. However, the accumulation of data regarding the clinical impact of NAbs is inconclusive. The majority of studies conclude that NAbs significantly impair IFN-b efficacy: many trials, including one of 1309 Danish MS patients, found an increase in relapse rate and/or worsening MRI outcomes in NAb positive groups, particularly when NAbs are present in higher titers, compared to NAb negative groups [49,67,74–76,79,85–89]. However, a single larger trial of 6698 MS participants failed to detect significant clinical differences in relapse rate, EDSS scores, or lesions on MRI between patients with NAbs and those without [90]. This discrepancy in clinical findings may, in part, be due to factors that also plague pharmacogenomics studies, including issues with statistical power and differing definitions of NAb positivity [91]. 12.4. Detection and measurement of NAbs Another likely source of variation may be derived from detection and quantification of NAbs. Measurement of NAbs was performed in many of IFN-b’s pivotal trials using a cytopathic effect (CPE) assay [46,74,79,85], as recommended by the World Health Organization [92]. The CPE assay is an in vitro method, where a patient’s serum blood sample is added to a cell line challenged with IFN-b and virus. The ability of IFN-b to inhibit viral replication reflects the patient’s NAb titer as calculated by the Kawade method [93,94]. Although this method is sensitive and specific, other assays have been developed that are less laborintensive, easier to standardize and automate, and have increased accessibility and safety by eliminating the need for use of viruses [95,96]. Investigators began testing the protein and mRNA expression of myxovirus protein A as a reliable biomarker of IFN-b’s receptor activity, and the European Medicines Agency

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

(EMA) endorsed its use and standardization in future studies [96]. The CPE and MxA assays are still problematic as significant variations in NAb titers and categorization of NAb positivity between laboratories have been found, which are highly dependent on small differences in positive cut-off titer or sensitivity analysis [52,58,97,98]. Despite this, the European Federation of Neurological Societies Task Force recommended clinicians to use either the CPE or MxA assay in their guidelines while following a standardized protocol to mitigate interlaboratory variance [99]. Efforts are underway to develop more rapid and reliable techniques, including quantitative real-time PCR methods to detect MxA gene expression and the use of luciferase-based reporter-gene assays to detect downstream IFNb receptor activity [51,100,101]. Researchers are also pursuing the investigation of new biomarkers such as interferon-inducible 6–16 mRNA, C-X-C motif chemokine 10 (CXCL-10), ubiquitin specific peptidase 18 (USP18) and HECT and RLD domain containing protein ligase 5 (HERC5) gene expression, and others that may be more sensitive, specific, and amenable to testing in clinical practice [53,102,103]. 12.5. Clinical management Due to the ambiguous data surrounding the clinical impact of NAbs on treatment response to IFN-b, disagreement has surfaced regarding their clinical management in practice. Clinical guidelines developed in North America and Europe concur that high titers of NAbs are likely detrimental to treatment outcome, but the strength of their recommendations for patient testing and management differs significantly. The European document recommends mandatory testing for NAbs in all patients on IFNb, and clearly outlines the timing and thresholds at which alternative therapy should be considered: patients’ blood should be collected at 12 and 24 months of therapy, and then every 3–6 months if they test positive for NAbs. Patients with repeated high titers should be discontinued or switched to alternative therapy. Due to significant cross-reactivity between the different IFN-b formulations [56,58,60,104], the alternative must be a non-IFN product [99]. In contrast, the outline released by the Therapeutics and Technology Assessment Subcommittee of the American Academy of Neurology (AAN) declines making clear recommendations on NAb testing on the basis of a perceived absence of Class I level evidence. The AAN claims that the inherent inability of NAb studies to randomize patients into NAb positive and negative status permits the possibility that extraneous patient-specific factors could be confounding the relationship between NAb status and response. The group also echoes concerns by previous researchers; notably, that studies have used varying definitions of NAb positivity status, and not all have accounted for reversion to seronegativity, which may also obscure the magnitude of NAbs’ effect. A lack of statistical power in the evidence was also cited as an important factor in their decision. Based on these observations, the panel concluded that IFN-b1a is likely less immunogenic than IFN-b1b, that persistent high titers of NAbs are associated with decreased treatment effectiveness, and that future research should focus on standardizing future protocols and determining the longterm impact of NAbs from post-marketing studies [90]. It is interesting to note that the two European panel members of the committee withdrew their authorship because in their opinion, the document’s recommendations did not wholly reflect the available evidence. In particular, the members disputed that only randomized trials may be considered Class I evidence, and instead expressed that only clinician blinding is necessary to be considered a reliable study. On that basis, they disagreed that the quality of evidence precluded making clinical recommendations, and

255

suggested testing for NAbs under the same conditions as the European guidelines [105]. A European consortium, the Neutralizing Antibodies on Interferon-b in Multiple Sclerosis (NABINMS), gathered several years later in response to inconsistent application of the European guidelines in practice, and developed a more specific treatment algorithm based on stratified definitions of clinical response and NAb titer. Clinical decisions based on NAb testing vary depending on the patient’s current clinical status: in patients with low disease activity, patients should consider switching to alternate therapy if their NAb titer is persistently high over several tests at 3–6 month intervals, whereas those with intermediate disease status should be switched more readily. Those with high disease activity should be switched regardless of NAb status. The group also reiterated that the alternative therapy should not consist of another IFN-b product [106]. 12.6. Pharmacogenomics of neutralizing antibody development Interestingly, one GWAS and one CGAS have also uncovered four SNPs in HLA and non-HLA regions that predict NAb formation: rs9272105, DRB1*0401, and DRB1*0408 within the HLA region and rs4961252 on chromosome position 8q24.3 [107,108]. In the CGAS, researchers genotyped 268 IFN-b-treated MS patients’ HLA-A, -B, C, -DRB1, and -DQB1 alleles and 242 patients’ HLADRB1*04 alleles. Antibodies against IFN-b were detected by capture ELISA and the biological in vivo activity of IFN-b was measured by Mx1 gene expression using TaqMan1 real-time PCR to assess biologically active antibodies. Antibodies were considered biologically active when Mx1 induction was decreased by >50% compared to newly treated control donors that were antibody-negative. They found a 5- and 14-times increase in the risk of developing anti-drug antibodies in patients with DRB1*0401 and DRB1*0408 polymorphisms respectively. However, they did not predict neutralizing vs. binding capacity [107]. In response to the success of their CGAS, the same group performed a GWAS in 362 IFN-b-treated patients using a chip containing over 300,000 SNPs. The validation cohort used a smaller chip to test a further 818 patients. The resulting statistically significant SNPs in rs4961252 and rs9272105 were not found to interact with the previously discovered polymorphisms in DRB1*0401 and DRB1*0408, and so the authors suggest that their conferred risks may be considered additive. Therefore, it may be possible to calculate a patient’s risk of NAb development prior to starting therapy, thereby allowing those with high risk of immunogenicity to consider potentially more effective alternative treatments [108]. A larger follow-up study of 1093 patients confirmed the roles of HLADRB1*0401 and HLADRB1*0408 on antibody development in addition to another likely risk allele, DRB1*1601. This gene failed adjusted significance testing in the screening cohort but surpassed significance in the validation cohort. Other genes, HLADRB1*0301, HLADRB1*0404, and HLADRB1*1104 were associated with a decreased risk of antibody development, but did not meet significance testing. Again, the relationship between these genes and ELISA could be calculated as additive risk ratios, where homozygous carriers are associated with the most pronounced effects on antibody development and MxA induction. They reported the OR for those carrying two risk alleles as 16.04 and 10.9 for ELISA positivity and impaired MxA induction, respectively. The OR for homozygous carriers of protective alleles was 0.49 for ELISA activity and 0.40 for impaired MxA induction, although again, these genes did not meet significance in the genotyping assays [109]. Lastly, an additional CGAS of 567 MS patients assessed multiple parameters of IFN-b response, including the formation of NAbs in relation to 42 SNPs in innate pattern recognition receptor genes

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

256

[110]. Patients’ DNA was probed using a MicroPlex-xTAG beadset, and NAb testing was performed using a modified CPE assay. A gender-differentiated analysis found that males with a SNP in toll-like receptor 6 (TLR6) (rs5743810) were more likely to develop NAbs after both 12 and 24 months of treatment. There was no association in females, nor any association to clinical response. The authors concede that their finding may be a false positive result and that further testing is warranted. If more conclusive evidence supporting the role of NAbs in IFN-b response is found, predicting patients’ risk of immunogenicity using tests for markers such as these is likely to become useful in the clinic.

13. Ramifications for pharmacogenomics Since the medical community recognizes that NAbs may influence treatment responses, pharmacogenomics researchers should consider that NAbs may also meaningfully influence pharmacogenomic clinical associations. As Table 1 illustrates, only one study fully acknowledged the potential effects of NAbs and incorporated NAb testing into its study design [25]. Cunningham et al. [25] tested all participants for NAbs and categorized them into those with high titer according to the CPE assay and those without. NAb status was reportedly ‘‘considered separately’’

Table 1 Pharmacogenomics research finding statistically significant genetic predictors of response to IFN-b. Pharmacogenomics study

Study type

Genes containing significant SNPs

Controlled for NAbs?

Sample size

Non-responder definition

Odds ratio (OR)

Kulakova et al. [37]

CGAS

CCR5, IFNAR1, DRB1, TGFB1, CTLA4

No

538

13.2, 13.1

Lo´pez-Go´mez et al. [36]

CGAS

TRAILR-1

Yes

509 + 226 in validation cohort; 350 in NAB analysis

Malhotra et al. [116]

CGAS

USP18

No

225

Kulakova et al. [117]

CGAS

No

253

Alvarez-Lafuente et al. [118]

CGAS

No

406

EDSS and relapse data over one year; cutoffs not specified

0.76; 1.21

Gross et al. [35]

CGAS

TGFB1*T; Combination of CCR5*d + IFNAR1*G + IFNB1*T/T; CCR5*d + IFNAR1*G + IFNG*T CD46 (no SNPs reach significance after correction) IRF8 (failed significance testing in lesser-powered replication cohort)

‘‘Clinically non-optimal’’: presence of relapses or sustained progression of EDSS over 2 years of treatment Presence of relapses, or an increase in EDSS score of 1.5 points if baseline score was <1, or increase of 1 point if baseline score was 1 Presence of relapses or increase of 1 EDSS point during first two years of treatment Occurrence of relapses or progression of EDSS score over 2 years as measured every 3 months

No in screening cohort; Yes in replication cohort

756 + 211 in replication cohort

Hazard ratio: 0.45; 0.53

Vosslamber et al. [28]

CGAS

IRF5

No

75 (MRI cohort); 261 (time to first relapse)

Time to first event, defined as: relapse, change in T2 hyperintense lesion burden or presence of any gadoliniumenhancing lesion on MRI, or increase in EDSS score by 1 point sustained over 6 months 1 new T2-weighted lesions on MRI over one year; shorter time to first relapse

Comabella et al. [21]

GWAS

No

106 + 94 in validation cohort

At least 1 relapse and an increase in EDSS score of 1 point over 2 consecutive visits separated by 6 months

1.7 (in men)–22.4

Ce´nit et al. [15]

CGAS

GRIA3, ADAR, IFNAR2, CIT, ZFAT, ZFHX4, STARD13 (uncorrected p-values) GPC5

No

199 + 493 healthy controls

0.1–0.39

O’Doherty et al. [26]

CGAS

No

255

Byun et al. [14]

GWAS

No

206 + 81 for joint analysis

At least 2 relapses or increase in EDSS score of 1 point over 2 years

0.22–3.87

Martı´nez et al. [119]

CGAS

Combination of JAK2- IL10RBGBP1-PIAS1 and JAK2-IL10-CASP3 HAPLN1, GPC5, COL25A1, CAST, TAFA1, NPAS3, LOC442331 IFNG

At least 2 relapses or increase in EDSS score by 1 point over 2 years Relapse rate remaining the same or increasing after 6–9 months of treatment

No

110

0–1.80

Cunningham et al. [25]

CGAS

IFNAR1, LMP7, CTSS, MxA

Yes

230

Wergeland et al. [43]

CGAS

IL-10 (significant for MRI T1-contrast enhancing lesions only)

No

63

Presence of relapses within first 2 years of treatment Relapse rate stays the same or increases after 6–9 months of treatment New occurrence of T1-contrast enhancing lesions or T2hyperintense MRI lesions, new relapses, or increase in EDSS scale at 6 months

0.30 (joint analysis)

0.7; 1.3; 0.8

0.1–14.3

Not reported

0.1–14.3

0.38–6.37

Not reported

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

from response criteria, but further specific details behind their analysis were not provided [25]. Although Cunningham et al. [25] did not find any association between NAbs and treatment response in their population, further studies may not arrive at the same conclusion. Lo´pez-Go´mez et al. [36] took a different approach for controlling for NAb factors, by creating a separate validation cohort in which NAbs were tested at 12 and 24 months in all patients. They found that the significant SNP in the screening phase reproducibly induced differences in response in NAbnegative patients. Gross et al. [35] similarly included a NAb validation cohort, although they only had access to relapse data and were unable to replicate their initial findings [35]. Nevertheless, this may be a useful balanced approach for future studies. As the number of partipants grows to bolster statistical power, the effects of NAbs may be sufficiently controlled by using validation cohorts while avoiding the need to test the entire patient population. Ideally, testing procedures should also follow a common protocol. In 1999, the European Medicines Agency formed a committee to develop a standardized assay to be used in clinical practice. The Committee for Medicinal Products for Human Use (CHMP) agreed to develop a standard protocol for the MxA assay in collaboration with the three marketing authorization holders of IFN-b products, Ares-Serono, Biogen, and Schering AG. The final validated assay requires incubating A549 cells seeded with patients’ serum and challenged with IFN. Experiments found that using IFN-b1a as the challenge IFN produced the least intra- and inter-laboratory variation compared to IFN-b1b products. After lysing the cells, MxA is measured by ELISA, using a rat anti-MxA monoclonal antibody to capture and a biotinylated mouse antiMxA monoclonal antibody for detection. As in other methods, the titer is determined using the Kawade formula [96]. A more detailed description of the protocol is given in Wadhwa et al. [95]. Pharmacogenomics studies should follow validated protocols such as these to ensure cross-study comparability and improve reproducibility. As there is still a certain degree of inter-laboratory variation in the MxA assay [95], it would be practical for researchers to consider standardizing the adoption of more reliable and rapid assays once developed. 14. Summary and considerations for future research Pharmacogenomics promises to tailor patients’ drug therapy to their own unique genetic make-up. The translation of this research will undoubtedly avoid unnecessary side effects, reduce costs, and most importantly, increase the overall effectiveness of drugs for patients. This is especially relevant in the MS population, where not all patients respond to first-line disease-modifying therapies. IFN-b was the first-approved therapy for relapsing-remitting multiple sclerosis and is still in use today, yet some estimates claim that as many as 49% of patients experience early treatment failure [11]. The ostensible goal of pharmacogenomics research is to develop a genetic test that could be put into routine clinical use, one that may predict a patient’s response to treatment based on genetic polymorphisms. While many efforts have been made to identify SNPs that predict response to IFN-b, progress is slow as a result of low reproducibility. Others have vocalized a number of factors that currently contribute to the disagreement between studies [19,26,32,38–40,111]. Small sample sizes increase the likelihood that important SNPs are not detected, while lack of comparable, contemporary natural history control groups and non-random treatment allocations in clinical practice further impair the ability to distinguish the influence of SNPs on treatment outcomes. These latter two limitations are less easily addressed due to ethical and economic concerns.

257

Suggestions that could be adopted are a fuller spectrum of possible treatment responses observed in clinical practice, agreedupon definitions of responder status and adherence to standardized methodologies. Responder status criteria have not reflected current evidence that predicts long-term success of IFN-b treatment, which includes MRI, relapse, and EDSS outcomes. Discrepant criteria related to definitions for responder status have inevitably led to differing proportions of MS patients categorized as responders or non-responders and confounded the true relationship between genetic differences and IFN-b response. Moreover, varying study protocols, gene chip technologies, and statistical analyses further complicate current research. Developing standardized methods and more clarity and consistency in defining responder status would greatly improve cross-comparison of results between studies. It is also clear that future research must begin to control for NAbs to IFN-b. Although the extent of their involvement in treatment failure has yet to be determined, the medical community has acknowledged that they most likely influence MS patients’ disease course. Therefore, it is suggested that future pharmacogenomics studies should control for NAbs in order to prevent NAb-induced treatment failure to be attributed to a false SNP. As with other aspects of their methodology, the assays for NAb testing should be standardized and validated. The EMA has developed an appreciably reliable MxA assay that may be used in pharmacogenomics studies to further enhance crossstudy comparability. Whether testing for NAb-predictive SNPs should be adopted as a useful measure remains to be determined. As new investigational SNPs are continually being examined for their putative roles in IFN-b therapy, those uncovered in previous CGAS and GWAS are still awaiting replication. Two SNPs from previous studies, GPC5 and IRF5, have been independently validated in separate laboratories after many others failed replication. Many more validation studies such as these will be needed to establish definitive progress; a list of promising candidate genes for follow-up study are provided by UEPHA*MS [33]. In addition, the two previous GWAS found several SNPs that were only weakly associated with response, indicating that IFN-b outcome is likely governed by multiple genes that contribute to a complex mechanism. Therefore future efforts must balance their focus between discovering new polymorphisms and solidifying known ones. Multiple sclerosis research and treatment have seen significant progress, with the identification of new susceptibility markers and approval of the new disease-modifying therapies fingolimod, teriflunomide, and dimethyl fumarate. As new data emerge concerning their longer-term safety and efficacy, we may expect to see these agents utilized along-side IFN-b and other injectables as 1st line treatment options [112–115]. Unfortunately, in contrast, pharmacogenomics progress as it relates to IFN-b has been slow. Unless many of the challenges identified are consistently and collaboratively addressed, it seems unlikely that the application of pharmacogenomics to tailor MS patients’ drug therapy will occur in the near future. Newer agents in MS present an opportune time to expand the science of pharmacogenomics research without repeating the same methodological inconsistencies which have occurred with IFN-b. As costs and risks of new treatments escalate, it becomes of further importance to be able to predict clinical response patterns associated with treatment. It is unclear how the upcoming PEGylated IFN-b1a and 1b products will affect current research and recommendations; however, researchers must continue to collaborate to ensure that their efforts are useful to the community and build upon established progress. The formation of UEPHA*MS was an important step in this

258

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

direction, and further collective efforts may eventually translate into a clinical tool that will lead to improved outcomes for MS patients.

[21]

Acknowledgements [22]

Funding for this work was provided by an operating grant from the Canadian Institutes of Health Research (CIHR) and the Saskatchewan Regional Partnership Program. We are grateful to Dr. Darren Nickel for reviewing the manuscript. Due to space limitations we extend apologies to colleagues whose work may not have been cited.

[24]

References

[25]

[1] Frohman EM, Racke MK, Raine CS. Multiple sclerosis: the plaque and its pathogenesis. N Engl J Med 2006;354:942–55, http://www.ncbi.nlm.nih.gov/ pubmed/16510748. [2] Truyen L, van Waesberghe JH, van Walderveen MA, van Oosten BW, Polman CH, Hommes OR, et al. Accumulation of hypointense lesions (‘black holes’) on T1 spin-echo MRI correlates with disease progression in multiple sclerosis. Neurology 1996;47:1469–76, http://www.ncbi.nlm.nih.gov/pubmed/8960729. [3] Polman CH, Reingold SC, Banwell B, Clanet M, Cohen JA, Filippi M, et al. Diagnostic criteria for multiple sclerosis: 2010 revisions to the McDonald criteria. Ann Neurol 2011;69:292–302, http://www.ncbi.nlm.nih.gov/ pubmed/21387374. [4] Tremlett H, Zhao Y, Rieckmann P, Hutchinson M. New perspectives in the natural history of multiple sclerosis. Neurology 2010;74:2004–15, http:// www.ncbi.nlm.nih.gov/pubmed/20548045. [5] Hader WJ. Disability and survival of multiple sclerosis in Saskatoon, Saskatchewan. Can J Neurol Sci 2010;37:28–35, http://www.ncbi.nlm.nih.gov/ pubmed/20169770. [6] Ragonese P, Aridon P, Salemi G, D’Amelio M, Savettieri G. Mortality in multiple sclerosis: a review. Eur J Neurol 2008;15:123–7, http:// www.ncbi.nlm.nih.gov/pubmed/18217882. [7] Kurtzke JF. Rating neurologic impairment in multiple sclerosis: an expanded disability status scale (EDSS). Neurology 1983;33:1444–52, http:// www.ncbi.nlm.nih.gov/pubmed/6685237. [8] Compston A, Coles A. Multiple sclerosis. Lancet 2002;359:1221–31, http:// www.ncbi.nlm.nih.gov/pubmed/11955556. [9] Oksenberg JR, Baranzini SE, Sawcer S, Hauser SL. The genetics of multiple sclerosis: SNPs to pathways to pathogenesis. Nat Rev Genet 2008;9:516–26, http://www.ncbi.nlm.nih.gov/pubmed/18542080. [10] Rieckmann P, Boyko A, Centonze D, Coles A, Elovaara I, Havrdova´ E, et al. Future MS care: a consensus statement of the MS in the 21st century steering group. J Neurol 2013;260:462–9, http://www.ncbi.nlm.nih.gov/pubmed/ 22936203. [11] Rı´o J, Nos C, Tintore´ M, Te´llez N, Gala´n I, Pelayo R, et al. Defining the response to interferon-beta in relapsing-remitting multiple sclerosis patients. Ann Neurol 2006;59:344–52, http://www.ncbi.nlm.nih.gov/pubmed/16437558. [12] Carlson RJ, Doucette JR, Nazarali AJ. Current developments in pharmacogenomics of multiple sclerosis. Cell Mol Neurobiol 2014;34(8):1081–5. [13] Mahurkar S, Suppiah V, O‘Doherty C. Pharmacogenomics of interferon beta and glatiramer acetate response: a review of the literature. Autoimmun Rev 2014;13:178–86, http://www.ncbi.nlm.nih.gov/pubmed/24189284. [14] Byun E, Caillier SJ, Montalban X, Villoslada P, Ferna´ndez O, Brassat D, et al. Genome-wide pharmacogenomic analysis of the response to interferon beta therapy in multiple sclerosis. Arch Neurol 2008;65:337–44, http:// www.ncbi.nlm.nih.gov/pubmed/18195134. [15] Ce´nit MDC, Blanco-Kelly F, de las Heras V, Bartolome´ M, de la Concha EG, Urcelay E, et al. Glypican 5 is an interferon-beta response gene: a replication study. Mult Scler 2009;15:913–7, http://www.ncbi.nlm.nih.gov/pubmed/ 19556317. [16] Comabella M, Lu¨nemann JD, Rı´o J, Sa´nchez A, Lo´pez C, Julia` E, et al. A type I interferon signature in monocytes is associated with poor response to interferon-beta in multiple sclerosis. Brain 2009;132:3353–65, http:// www.ncbi.nlm.nih.gov/pubmed/19741051. [17] Van Baarsen LG, Vosslamber S, Tijssen M, Baggen JM, van der Voort LF, Killestein J, et al. Pharmacogenomics of interferon-beta therapy in multiple sclerosis: baseline IFN signature determines pharmacological differences between patients. PLoS ONE 2008;3:e1927. http://dx.doi.org/10.1371/journal.pone.0001927, http://www.ncbi.nlm.nih.gov/pubmed/18382694. [18] Verweij CL, Vosslamber S. Relevance of the type I interferon signature in multiple sclerosis towards a personalized medicine approach for interferonbeta therapy. Discov Med 2013;15:51–60, http://www.ncbi.nlm.nih.gov/ pubmed/23375014. [19] Vandenbroeck K, Urcelay E, Comabella M. IFN-beta pharmacogenomics in multiple sclerosis. Pharmacogenomics 2010;11:1137–48, http:// www.ncbi.nlm.nih.gov/pubmed/20712530. [20] Pravica V, Popadic D, Savic E, Markovic M, Drulovic J, Mostarica-Stojkovic M. Single nucleotide polymorphisms in multiple sclerosis: disease susceptibility

[23]

[26]

[27]

[28]

[29]

[30]

[31] [32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

and treatment response biomarkers. Immunol Res 2012;52:42–52, http:// www.ncbi.nlm.nih.gov/pubmed/22392049. Comabella M, Craig DW, Morcillo-Sua´rez C, Rı´o J, Navarro A, Ferna´ndez M, 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;66:972–8, http://www.ncbi.nlm.nih.gov/pubmed/ 19667218. Baranzini SE, Wang J, Gibson RA, Galwey N, Naegelin Y, Barkhof F, et al. Genome-wide association analysis of susceptibility and clinical phenotype in multiple sclerosis. Hum Mol Genet 2009;18:767–78, http://www.ncbi.nlm.nih.gov/pubmed/19010793. Sriram U, Barcellos LF, Villoslada P, Rio J, Baranzini SE, Caillier S, et al. Pharmacogenomic analysis of interferon receptor polymorphisms in multiple sclerosis. Genes Immun 2003;4:147–52, http://www.ncbi.nlm.nih.gov/ pubmed/12618863. Ferna´ndez O, Ferna´ndez V, Mayorga C, Guerrero M, Leo´n A, Tamayo JA, et al. HLA class II and response to interferon-beta in multiple sclerosis. Acta Neurol Scand 2005;112:391–4, http://www.ncbi.nlm.nih.gov/pubmed/16281922. Cunningham S, Graham C, Hutchinson M, Droogan A, O’Rourke K, Patterson C, 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;78:635–46, http://www.ncbi.nlm.nih.gov/ pubmed/16338279. O’Doherty C, Villoslada P, Vandenbroeck K. Pharmacogenomics of type I interferon therapy: a survey of response-modifying genes. Cytokine Growth Factor Rev 2007;18:211–22, http://www.ncbi.nlm.nih.gov/pubmed/ 17540610. Santana-de Anda K, Go´mez-Martı´n D, Dı´az-Zamudio M, Alcocer-Varela J. Interferon regulatory factors: beyond the antiviral response and their link to the development of autoimmune pathology. Autoimmun Rev 2011;11:98– 103, http://www.ncbi.nlm.nih.gov/pubmed/21872684. Vosslamber S, van der Voort LF, van den Elskamp IJ, Heijmans R, Aubin C, Uitdehaag BMJ, et al. Interferon regulatory factor 5 gene variants and pharmacological and clinical outcome of interferon-b therapy in multiple sclerosis. Genes Immun 2011;12:466–72, http://www.ncbi.nlm.nih.gov/pubmed/ 21471993. Vandenbroeck K, Alloza I, Swaminathan B, Antigu¨edad A, Otaegui D, Olascoaga J, 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;12:40–5, http://www.ncbi.nlm.nih.gov/pubmed/20861862. Sellebjerg F, Søndergaard HB, Koch-Henriksen N, Sørensen PS, Oturai AB. Prediction of response to interferon therapy in multiple sclerosis. Acta Neurol Scand 2014;(June). http://dx.doi.org/10.1111/ane.12269, http://www.ncbi.nlm.nih. gov/pubmed/24943672 [Epub ahead of print]. Cohen J. Quantitative methods in psychology: a power primer. Psychol Bull 1992;112:155–9. Vandenbroeck K, Matute C. Pharmacogenomics of the response to IFN-b in multiple sclerosis: ramifications from the first genome-wide screen. Pharmacogenomics 2008;9:639–45, http://www.ncbi.nlm.nih.gov/pubmed/ 18466107. Vandenbroeck K, Comabella M, Tolosa E, Goertsches R, Brassat D, Hintzen R, et al. United Europeans for development of pharmacogenomics in multiple sclerosis network. Pharmacogenomics 2009;10:885–94, http:// www.ncbi.nlm.nih.gov/pubmed/19450134. Malhotra S, Morcillo-Sua´rez C, Brassat D, Goertsches R, Lechner-Scott J, Urcelay E, et al. IL28B polymorphisms are not associated with the response to interferon-b in multiple sclerosis. J Neuroimmunol 2011;239:101–4, http://www.ncbi.nlm.nih.gov/pubmed/21889215. Gross R, Healy BC, Cepok S, Chitnis T, Khoury SJ, Hemmer B, et al. Population structure and HLA DRB1 1501 in the response of subjects with multiple sclerosis to first-line treatments. J Neuroimmunol 2011;233:168–74, http:// www.ncbi.nlm.nih.gov/pubmed/21115201. ´ rpez-Zafra T, Pinto-Medel MJ, OliverLo´pez-Go´mez C, Pino-A´ngeles A, O Martos B, Ortega-Pinazo J, et al. Candidate gene study of TRAIL and TRAIL receptors: association with response to interferon beta therapy in multiple sclerosis patients. PLOS ONE 2013;8:e62540. http://dx.doi.org/10.1371/journal.pone.0062540, http://www.ncbi.nlm.nih.gov/pubmed/23658636. Kulakova OG, Tsareva EY, Lvovs D, Favarov AV, Boyko AN, Favarova OO. Comparative pharmacogenetics of multiple sclerosis: IFN-b versus glatiramer acetate. Pharmacogenomics 2014;15:679–85, http://www.ncbi.nlm.nih.gov/pubmed/24798724. Killestein J, Polman CH. Determinants of interferon b efficacy in patients with multiple sclerosis. Nat Rev Neurol 2011;7:221–8, http://www.ncbi.nlm.nih.gov/pubmed/21364522. Miller A, Avidan N, Tzunz-Henig N, Glass-Marmor L, Lejbkowicz I, Pinter RY, et al. Translation towards personalized medicine in multiple sclerosis. J Neurol Sci 2008;274:68–75, http://www.ncbi.nlm.nih.gov/pubmed/ 18789804. Vandenbroeck K, Comabella M. Single-nucleotide polymorphisms in response to interferon-beta therapy in multiple sclerosis. J Interferon Cytokine Res 2010;30:727–32, http://www.ncbi.nlm.nih.gov/pubmed/20836710. Rı´o J, Comabella M, Montalban X. Predicting responders to therapies for multiple sclerosis. Nat Rev Neurol 2009;5:553–60, http://www.ncbi.nlm.nih.gov/pubmed/19794514. Freedman MS, Selchen D, Arnold DL, Prat A, Banwell B, Yeung M, et al. Treatment optimization in MS: Canadian MS Working Group updated

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261

[43]

[44]

[45]

[46]

[47]

[48]

[49]

[50] [51]

[52]

[53]

[54]

[55]

[56]

[57]

[58]

[59]

[60]

[61]

[62]

[63]

recommendations. Can J Neurol Sci 2013;40:307–23, http://www.ncbi.nlm.nih.gov/pubmed/23603165. Wergeland S, Beiske A, Nyland H, Hovdal H, Jensen D, Larsen JP, et al. IL-10 promoter haplotype influence on interferon treatment response in multiple sclerosis. Eur J Neurol 2005;12:171–5, http://www.ncbi.nlm.nih.gov/ pubmed/15693804. Klawiter EC, Cross AH, Naismith RT. The present efficacy of multiple sclerosis therapeutics: is the new 66% just the old 33%? Neurology 2009;73:984–90, http://www.ncbi.nlm.nih.gov/pubmed/19770475. Gold R, Kappos L, Arnold DL, Bar-Or A, Giovannoni G, Selmaj K, et al. Placebocontrolled phase 3 study of oral BG-12 for relapsing multiple sclerosis. N Engl J Med 2012;367:1098–107, http://www.ncbi.nlm.nih.gov/pubmed/22992073. The IFNB Multiple Sclerosis Study Group. Interferon Beta-1b is effective in relapsing-remitting multiple sclerosis. I. Clinical results of a multicenter, randomized, double-blind, placebo-controlled trial. Neurology 1993;43:655– 61, http://www.ncbi.nlm.nih.gov/pubmed/8469318. Deisenhammer F, Reindl M, Harvey J, Gasse T, Dilitz E, Berger T. Bioavailability of interferon beta 1b in MS patients with and without neutralizing antibodies. Neurology 1999;52:1239–43, http://www.ncbi.nlm.nih.gov/pubmed/ 10214750. Vallittu A-M, Halminen M, Peltoniemi J, Ilonen J, Julkunen I, Salmi A, et al. Neutralizing antibodies reduce MxA protein induction in interferon-beta-1atreated MS patients. Neurology 2002;58:1786–90, http://www.ncbi.nlm.nih.gov/pubmed/12084878. Rudick RA, Simonian NA, Alam JA, Campion M, Scaramucci JO, Jones W, et al. Incidence and significance of neutralizing antibodies to interferon beta-1a in multiple sclerosis, multiple sclerosis collaborative research group (MSCRG). Neurology 1998;50:1266–72, http://www.ncbi.nlm.nih.gov/pubmed/ 9595973. Bendtzen K. Anti-IFN BAb and NAb antibodies: a minireview. Neurology 2003;61:S6–10, http://www.ncbi.nlm.nih.gov/pubmed/14610102. Bendtzen K. Critical review: assessment of interferon-b immunogenicity in multiple sclerosis. J Interferon Cytokine Res 2010;30:759–66, http:// www.ncbi.nlm.nih.gov/pubmed/20874253. Ross C, Clemmesen KM, Sørensen PS, Koch-Henriksen N, Bendtzen K. Measuring and evaluating interferon b-induced antibodies in patients with multiple sclerosis. Mult Scler 2006;12:39–46, http://www.ncbi.nlm.nih.gov/pubmed/16459718. Hegen H, Millonig A, Bertolotto A, Comabella M, Giovanonni G, Guger M, et al. Early detection of neutralizing antibodies to interferon-beta in multiple sclerosis patients: binding antibodies predict neutralizing antibody development. Mult Scler 2014;20:577–87, http://www.ncbi.nlm.nih.gov/pubmed/ 24009164. Jensen PEH, Sellebjerg F, Søndergaard HB, Sørensen PS. Correlation between anti-interferon-b binding and neutralizing antibodies in interferon-b-treated multiple sclerosis patients. Eur J Neurol 2012;19:1311–7, http:// www.ncbi.nlm.nih.gov/pubmed/22564111. Sethu S, Govindappa K, Quinn P, Wadhwa M, Stebbings R, Boggild M, et al. Immunoglobulin G1 and immunoglobulin G4 antibodies in multiple sclerosis patients treated with IFNb interact with the endogenous cytokine and activate complement. Clin Immunol 2013;148:177–85, http:// www.ncbi.nlm.nih.gov/pubmed/23770627. Ferna´ndez O, Mayorga C, Luque G, Guerrero M, Guerrero R, Leyva L, et al. Study of binding and neutralising antibodies to interferon-beta in two groups of relapsing-remitting multiple sclerosis patients. J Neurol 2001;248:383–8, http://www.ncbi.nlm.nih.gov/pubmed/11437159. Bertolotto A, Malucchi S, Sala A, Orefice G, Carrieri PB, Capobianco M, et al. Differential effects of three interferon betas on neutralising antibodies in patients with multiple sclerosis: a follow up study in an independent laboratory. J Neurol Neurosurg Psychiatry 2002;73:148–53, http:// www.ncbi.nlm.nih.gov/pubmed/12122172. Ross C, Clemmesen KM, Svenson M, Sørensen PS, Koch-Henriksen N, Skovgaard GL, et al. Immunogenicity of interferon-b in multiple sclerosis patients: influence of preparation, dosage, dose frequency, and route of administration. Danish Multiple Sclerosis Study Group. Ann Neurol 2000;48:706–12, http:// www.ncbi.nlm.nih.gov/pubmed/11079533. Scagnolari C, Bellomi F, Turriziani O, Bagnato F, Tomassini V, Lavolpe V, et al. Neutralizing and binding antibodies to IFN-beta: relative frequency in relapsing-remitting multiple sclerosis patients treated with different IFN-beta preparations. J Interferon Cytokine Res 2002;22:207–13, http:// www.ncbi.nlm.nih.gov/pubmed/11911803. Kivisa¨kk P, Alm GV, Fredrikson S, Link H. Neutralizing and binding antiinterferon-beta (IFN-beta) antibodies. A comparison between IFN-beta-1a and IFN-beta-1b treatment in multiple sclerosis. Eur J Neurol 2000;7:27–34, http://www.ncbi.nlm.nih.gov/pubmed/10809912. Durelli L, Verdun E, Barbero P, Bergui M, Versino E, Ghezzi A, et al. Independent comparison of interferon (INCOMIN) trial study group, every-other-day interferon beta-1b versus once-weekly interferon beta-1a for multiple sclerosis: results of a 2-year prospective randomised multicentre study (INCOMIN). Lancet 2002;359:1453–60, http://www.ncbi.nlm.nih.gov/pubmed/11988242. Gneiss C, Tripp P, Reichartseder F, Egg R, Ehling R, Lutterotti A, et al. Differing immunogenic potentials of interferon beta preparations in multiple sclerosis patients. Mult Scler 2006;12:731–7, http://www.ncbi.nlm.nih.gov/pubmed/ 17263000. Hartung H-P, Munschauer 3rd F, Schellekens H. Significance of neutralizing antibodies to interferon beta during treatment of multiple sclerosis: expert

[64]

[65]

[66]

[67]

[68]

[69]

[70]

[71]

[72]

[73]

[74]

[75]

[76]

[77]

[78]

[79]

[80]

[81]

[82]

[83]

[84]

[85]

259

opinions based on the proceedings of an international consensus conference. Eur J Neurol 2005;12:588–601, http://www.ncbi.nlm.nih.gov/pubmed/ 16053466. Sørensen PS, Koch-Henriksen N, Ross C, Clemmesen KM, Bendtzen K, Danish Multiple Sclerosis Study Group. Appearance and disappearance of neutralizing antibodies during interferon-beta therapy. Neurology 2005;65:33–9, http://www.ncbi.nlm.nih.gov/pubmed/15888603. Bellomi F, Scagnolari C, Tomassini V, Gasperini C, Paolillo A, Pozzilli C, et al. Fate of neutralizing and binding antibodies to IFN beta in MS patients treated with IFN beta for 6 years. J Neurol Sci 2003;215:3–8, http://www.ncbi.nlm.nih.gov/pubmed/14568120. Hegen H, Schleiser M, Gneiss C, Di Pauli F, Ehling R, Kuenz B, et al. Persistency of neutralizing antibodies depends on titer and interferon-beta preparation. Mult Scler 2012;18:610–5, http://www.ncbi.nlm.nih.gov/pubmed/22013146. Koch-Henriksen N, Sørensen PS, Bendtzen K, Flachs EM. The clinical effect of neutralizing antibodies against interferon-beta is independent of the type of interferon-beta used for patients with relapsing-remitting multiple sclerosis. Mult Scler 2009;15:601–5, http://www.ncbi.nlm.nih.gov/ pubmed/19299439. Cohen BA, Oger J, Gagnon A, Giovannoni G. The implications of immunogenicity for protein-based multiple sclerosis therapies. J Neurol Sci 2008;275:7–17, http://www.ncbi.nlm.nih.gov/pubmed/18822434. Farrell RA, Marta M, Gaeguta AJ, Souslova V, Giovannoni G, Creeke PI. Development of resistance to biologic therapies with reference to IFN-b. Rheumatology 2012;51:590–9, http://www.ncbi.nlm.nih.gov/pubmed/22258390. Bertolotto A, Deisenhammer F, Gallo P, Sørensen PS. Immunogenicity of interferon beta: differences among products. J Neurol 2004;251:II15–24, http://www.ncbi.nlm.nih.gov/pubmed/15264108. Bozhinov A, Handzhiyski Y, Genov K, Daskalovska V, Niwa T, Ivanov I, et al. Advanced glycation end products contribute to the immunogenicity of IFN-b pharmaceuticals. J Allergy Clin Immunol 2012;129:855–8, http://www. ncbi.nlm.nih.gov/pubmed/22154379. Fox EJ, Vartanian TK, Zamvil SS. The immunogenicity of disease-modifying therapies for multiple sclerosis: clinical implications for neurologists. Neurologist 2007;13:355–62, http://www.ncbi.nlm.nih.gov/pubmed/18090713. Barnard JG, Babcock K, Carpenter JF. Characterization and quantitation of aggregates and particles in interferon-b products: potential links between product quality attributes and immunogenicity. J Pharm Sci 2013;102:915– 98, http://www.ncbi.nlm.nih.gov/pubmed/23233295. Francis GS, Rice GPA, Alsop JC, PRISMS Study Group. Interferon beta-1a in MS: results following development of neutralizing antibodies in PRISMS. Neurology 2005;65:48–55, http://www.ncbi.nlm.nih.gov/pubmed/16009884. Kappos L, Clanet M, Sandberg-Wollheim M, Radue EW, Hartung H-P, Hohlfeld R, et al. Neutralizing antibodies and efficacy of interferon beta-1a: a 4-year controlled study. Neurology 2005;65:40–7, http://www.ncbi.nlm.nih.gov/ pubmed/16009883. Li DK, Zhao GJ, Paty DW, University of British Columbia MS/MRI. Analysis Research Group. The SPECTRIMS Study Group. Randomized controlled trial of interferon-beta-1a in secondary progressive MS: MRI results. Neurology 2001;56:1505–13, http://www.ncbi.nlm.nih.gov/pubmed/11402107. The Once Weekly Interferon for MS Study Group. Evidence of interferon b-1a dose response in relapsing-remitting MS the OWIMS study. Neurology 1999;53:679–86, http://www.ncbi.nlm.nih.gov/pubmed/1048902. Prevention of relapses and disability by interferon beta-1a subcutaneously in Multiple Sclerosis (PRISMS) Study Group. Randomised double-blind placebocontrolled study of interferon b-1a in relapsing/remitting multiple sclerosis. Lancet 1998;352:1498–504, http://www.ncbi.nlm.nih.gov/pubmed/ 9820297. The Prevention of Relapses and Disability by Interferon-beta-1a Subcutaneously in Mulitple Sclerosis (PRISMS) Study Group, The University of British Columbia MS/MRI Analysis Group. PRISMS-4: long-term efficacy of interferon-beta-1a in relapsing MS. Neurology 2001;56:1628–36, http:// www.ncbi.nlm.nih.gov/pubmed/11425926. Schwid SR, Panitch HS. full results of the evidence of interferon dose–response – European North American Comparative Efficacy (EVIDENCE) study: a multicenter, randomized, assessor-blinded comparison of low-dose weekly versus high-dose, high-frequency interferon b-1a for relapsing multiple sclerosis. Clin Ther 2007;29:2031–48, http://www.ncbi.nlm.nih.gov/pubmed/18035202. Hesse D, Sellebjerg F, Sørensen PS. Absence of MxA induction by interferon beta in patients with MS reflects complete loss of bioactivity. Neurology 2009;73:372–7, http://www.ncbi.nlm.nih.gov/pubmed/19652141. Bertolotto A, Gilli F, Sala A, Capobianco M, Malucchi S, Milano E, et al. Persistent neutralizing antibodies abolish the interferon beta bioavailability in MS patients. Neurology 2003;60:634–9, http://www.ncbi.nlm.nih.gov/ pubmed/12601105. Pachner AR, Warth JD, Pace A, Goelz S, INSIGHT Investigators. Effect of neutralizing antibodies on biomarker responses to interferon beta: the INSIGHT study. Neurology 2009;73:1493–500, http://www.ncbi.nlm.nih.gov/pubmed/19884577. Scagnolari C, Duda P, Bagnato F, De Vito G, Alberelli A, Lavolpe V, et al. Pharmacodynamics of interferon beta in multiple sclerosis patients with or without serum neutralizing antibodies. J Neurol 2007;254:597–604, http:// www.ncbi.nlm.nih.gov/pubmed/17420930. The IFNB Multiple Sclerosis Study Group and The University of British Columbia MS/MRI Analysis Group. Neutralizing antibodies during treatment of multiple sclerosis with interferon beta-1b: experience during the first

260

[86]

[87]

[88]

[89]

[90]

[91]

[92]

[93]

[94]

[95]

[96]

[97]

[98]

[99]

[100]

[101]

[102]

[103]

[104]

[105]

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261 three years. Neurology 1996;47:889–94, http://www.ncbi.nlm.nih.gov/ pubmed/8857714. Sørensen PS, Ross C, Clemmesen KM, Bendtzen K, Frederiksen JL, Jensen K, et al. Clinical importance of neutralising antibodies against interferon beta in patients with relapsing-remitting multiple sclerosis. Lancet 2003;362: 1184– 91, http://www.ncbi.nlm.nih.gov/pubmed/14568740. European study group on interferon beta-1b in secondary progressive MS. Placebo-controlled multicentre randomised trial of interferon [beta]-1b in treatment of secondary progressive multiple sclerosis. Lancet 1998;352:1491–7, http://www.ncbi.nlm.nih.gov/pubmed/9820296. Perini P, Calabrese M, Biasi G, Gallo P. The clinical impact of interferon beta antibodies in relapsing-remitting MS. J Neurol 2004;251:305–9, http:// www.ncbi.nlm.nih.gov/pubmed/15015010. Paolicelli D, D’Onghia M, Pellegrini F, Direnzo V, Iaffaldano P, Lavolpe V, et al. The impact of neutralizing antibodies on the risk of disease worsening in interferon b-treated relapsing multiple sclerosis: a 5 year post-marketing study. J Neurol 2013;260:1562–8, http://www.ncbi.nlm.nih.gov/pubmed/ 23417273. Goodin DS, Hurwitz B, Noronha A. Neutralizing antibodies to interferon1b are not associated with disease worsening in multiple sclerosis. J Int Med Res 2007;35:173–87, http://www.ncbi.nlm.nih.gov/pubmed/ 17542405. Hurwitz BJ. Important sources of variability in clinical studies of neutralizing antibodies against interferon beta. J Neurol Sci 2008;272:8–19, http:// www.ncbi.nlm.nih.gov/pubmed/18620708. World Health Organization. WHO Expert Committee on Biological Standardization. Forty-fifth report. World Health Organization Technical Rep Ser 1995;858:1–101, http://www.ncbi.nlm.nih.gov/pubmed/8650946. Grossberg SE, Kawade Y, Kohase M, Yokoyama H, Finter N. The neutralization of interferons by antibody. I. Quantitative and theoretical analyses of the neutralization reaction in different bioassay systems. J Interferon Cytokine Res 2001;21:729–42, http://www.ncbi.nlm.nih.gov/pubmed/ 11576467. Grossberg SE, Kawade Y, Kohase M, Klein JP. The neutralization of interferons by antibody. II. Neutralizing antibody unitage and its relationship to bioassay sensitivity: the tenfold reduction unit. J Interferon Cytokine Res 2001;21:743–55, http://www.ncbi.nlm.nih.gov/pubmed/11576468. Wadhwa M, Subramanyam M, Goelz S, Goyal J, Jethwa V, Jones W, et al. Use of a standardized MxA protein measurement-based assay for validation of assays for the assessment of neutralizing antibodies against interferon-b. J Interferon Cytokine Res 2013;33:660–71, http://www.ncbi.nlm.nih.gov/ pubmed/23848523. European Medicines Agency. Beta-interferons and neutralising antibodies (in multiplesclerosis); 2008, http://www.ema.europa.eu/docs/en_GB/document_library/Report/2009/11/WC500015395.pdf. Bellomi F, Bramanti P, Trojano M, Scagnolari C, Muto A, Sessa E, et al. Neutralizing and binding antibodies to interferon beta in patients with multiple sclerosis: a comparison of assay results from three Italian centres. J Immunoassay Immunochem 2009;30:40–50, http://www.ncbi.nlm.nih.gov/pubmed/19117201. Hartung H-P, Kieseier B, Goodin DS, Arnason BGW, Comi G, Cook S, et al. Variability in detection and quantification of interferon b-1b-induced neutralizing antibodies. J Neuroinflamm 2012;9:129, http://www.ncbi.nlm.nih.gov/pubmed/22703536. Sørensen PS, Deisenhammer F, Duda P, Hohlfeld R, Myhr K-M, Palace J, et al. Guidelines on use of anti-IFN-B antibody measurements in multiple sclerosis: report of an EFNS task force on IFN-B antibodies in multiple sclerosis. Eur J Neurol 2005;12:817–27, http://www.ncbi.nlm.nih.gov/pubmed/ 16241970. Pachner A, Narayan K, Price N, Hurd M, Dail D. MxA gene expression analysis as an interferon-beta bioactivity measurement in patients with multiple sclerosis and the identification of antibody-mediated decreased bioactivity. Mol Diagn 2003;7:17–25, http://www.ncbi.nlm.nih.gov/ pubmed/14529316. Bertolotto A, Sala A, Caldano M, Capobianco M, Malucchi S, Marnetto F, et al. Development and validation of a real time PCR-based bioassay for quantification of neutralizing antibodies against human interferon-beta. J Immunol Methods 2007;321:19–31, http://www.ncbi.nlm.nih.gov/pubmed/ 17335844. Moore M, Meager A, Wadhwa M, Burns C. Measurement of neutralising antibodies to type I interferons by gene expression assays specific for type 1 interferon-inducible 6–16 mRNA. J Pharm Biomed Anal 2009;49:534–9, http://www.ncbi.nlm.nih.gov/pubmed/19097723. Malhotra S, Bustamante MF, Pe´rez-Miralles F, Rio J, Ruiz de Villa MC, Vegas E, et al. Search for specific biomarkers of IFNb bioactivity in patients with multiple sclerosis. PLOS ONE 2011;6:e23634. http://dx.doi.org/10.1371/ journal.pone.0023634, http://www.ncbi.nlm.nih.gov/pubmed/21886806. Gneiss C, Koudouovoh-Tripp P-M, Ropele S, Gotwald T, Ehling R, Lutterotti A, et al. Influence of interferon-beta therapy switching on neutralizing antibody titres: results from the Austrian Switch Study. Mult Scler 2009;15:1481–8, http://www.ncbi.nlm.nih.gov/pubmed/19965519. Sørensen PS, Bertolotto A. Re: neutralizing antibodies to interferon beta: assessment of their clinical and radiographic impact: an evidence report:

[106]

[107]

[108]

[109]

[110]

[111]

[112]

[113]

[114]

[115]

[116]

[117]

[118]

[119]

report of the therapeutics and technology assessment subcommittee of the american academy of neurology. Neurology 2007;69:1552. author reply 1553. http://www.ncbi.nlm.nih.gov/pubmed/17923618. Polman CH, Bertolotto A, Deisenhammer F, Giovannoni G, Hartung H-P, Hemmer B, et al. Recommendations for clinical use of data on neutralising antibodies to interferon-beta therapy in multiple sclerosis. Lancet Neurol 2010;9(7):740–50, http://www.ncbi.nlm.nih.gov/pubmed/20610349. Hoffmann S, Cepok S, Grummel V, Lehmann-Horn K, Hackermu¨ller J, Stadler PF, 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;83:219–27, http://www.ncbi.nlm.nih.gov/ pubmed/18656179. Weber F, Cepok S, Wolf C, Berthele A, Uhr M, Bettecken TH, et al. Singlenucleotide polymorphisms in HLA-and non-hla genes associated with the development of antibodies to interferon-b therapy in multiple sclerosis patients. Pharmacogenomics J 2012;12:238–45, http://www.ncbi.nlm.nih.gov/pubmed/21502966. Buck D, Cepok S, Hoffmann S, Grummel V, Jochim A, Berthele A, et al. Influence of the HLA-DRB1 genotype on antibody development to interferon beta in multiple sclerosis. Arch Neurol 2011;68:480–7, http:// www.ncbi.nlm.nih.gov/pubmed/21482927. Enevold C, Oturai AB, Sørensen PS, Ryder LP, Koch-Henriksen N, Bendtzen K. Polymorphisms of innate pattern recognition receptors, response to interferon-beta and development of neutralizing antibodies in multiple sclerosis patients. Mult Scler 2010;16:942–9, http://www.ncbi.nlm.nih.gov/pubmed/ 20595247. Macciardi F, Martinelli Boneschi F, Cohen D. Pharmacogenetics of autoimmune diseases: research issues in the case of multiple sclerosis and the role of IFN-beta. J Autoimmun 2005;25(Suppl.):1–5, http://www.ncbi.nlm.nih.gov/ pubmed/16311019. Kira J, Itoyama Y, Kikuchi S, Hao Q, Kurosawa T, Ueda K, et al. Long-term efficacy and safety of Fingolimod in Japanese patients with relapsing multiple sclerosis (MS): 3-year results from a phase 2 extension study. Mult Scler J 2014;20(7):9282 [Poster (P-56) presented at the 6th PACTRIMS Congress, 6– 8 Nov. 2013, Kyoto, Japan]. Miller DH, Fox RJ, Gold R, Havrdova E, Kappos K, Yousry T, et al. 3-Year follow-up of oral BG-12 (dimethyl fumerate) for relapsing-remitting multiple sclerosis (RRMS): integrated MRI outcomes from the DEFINE, CONFIRM, and ENDORSE studies. Mult Scler J 2014;20(7):932 [Poster (P66) presented at the 6th PACTRIMS meeting, 6–8 Nov. 2013, Kyoto, Japan]. Gold R, Phillips JT, Bar-Or A, Hutchinson M, Kappos L, Rana J, et al. 3-Year follow-up of oral BG-12 (Dimethyl Fumerate) for relapsing-remitting multiple sclerosis (RRMS): integrated clinical outcomes from the DEFINE, CONFIRM, and ENDORSE studies. Mult Scler J 2014;20(7):932 [Poster (P67) presented at the 6th PACTRIMS meeting, 6–8 Nov. 2013, Kyoto, Japan]. Leist TP, Freedman MS, Kappos L, Olsson TP, Miller AE, Wolinsky JS, et al. Three placebo-controlled teriflunomide studies: pooled safety data. Mult Scler J 2014;20(7):933 [Poster (P-70) presented at the 6th PACTRIMS meeting, 6–8 Nov. 2013, Kyoto, Japan]. Malhotra S, Morcillo-Sua´rez C, Nurtdinov R, Rio J, Sarro E, Moreno M, et al. Roles of the ubiquitin peptidase USP18 in multiple sclerosis and the response to interferon-b treatment. Eur J Neurol 2013;20:1390–7, http:// www.ncbi.nlm.nih.gov/pubmed/23700969. Kulakova OG, Tsareva EY, Boyko AN, Shchur SG, Gusev EI, Lvovs D, et al. Allelic combinations of immune-response genes as possible composite markers of IFN-beta efficacy in multiple sclerosis patients. Pharmacogenomics 2012;13:1689–700, http://www.ncbi.nlm.nih.gov/pubmed/23171334. Alvarez-Lafuente R, Blanco-Kelly F, Garcia-Montojo M, Martı´nez A, De Las Heras V, Dominguez-Mozo MI, et al. CD46 in a Spanish cohort of multiple sclerosis patients: genetics, mRNA expression and response to interferonbeta treatment. Mult Scler 2011;17:513–20, http://www.ncbi.nlm.nih.gov/ pubmed/21177319. Martı´nez A, de las Heras V, Mas Fontao A, Bartolome´ 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;173:196–9, http:// www.ncbi.nlm.nih.gov/pubmed/16430971.

Rebecca J. Carlson received her BSP with Great Distinction from the College of Pharmacy and Nutrition, University of Saskatchewan in 2014. She was on the Dean’s Honor Roll 2012–2013 and has received numerous awards including the University of Saskatchewan Scholarship, Greystone Scholar member, PepsiCo ExCEL Scholarship, Linda Fritz Scholarship, Jack Moore Memorial Scholarship and Merck, Sharp & Dohme Scholarship. She is an Associate of the Royal Conservatory of Toronto (ARCT) diploma in Piano Performance. Ms. Carlson’s area of research interest is in pharmacogenomics of multiple sclerosis.

R.J. Carlson et al. / Cytokine & Growth Factor Reviews 26 (2015) 249–261 Dr. J. Ronald Doucette is a Professor of Anatomy and Cell Biology, University of Saskatchewan. He received his B.Sc. (Psychology, 1976) from Acadia University, M.Sc. (Physiological Psychology, 1977) from the University of Guelph and his PhD (Neuroanatomy, 1981) from the University of Western Ontario. Dr. Doucette did his post-doctoral fellowships at the McMaster University (with Dr. J. Diamond) and the University of Western Ontario (with Dr. M. Ball). Dr. Doucette is a principal investigator with the Cameco Multiple Sclerosis and Neuroscience Research Center and principal investigator with the Neuroscience Research Cluster, University of Saskatchewan. Dr. Doucette’s primary research interest is in studying how the nervous system repairs itself after injury, focusing in particular on the migratory ability of glial cells grafted into or relatively close to a damaged area. The glial cells he studies are the olfactory ensheathing cells, which in vivo provide ensheathment for olfactory axons. These cells have received increased attention in recent years due to their potential for forming part of a therapeutic approach in the repair of brain/spinal cord injury and in the remyelination of axons in diseases such as multiple sclerosis. Dr. Katherine Knox is an assistant professor at the University of Saskatchewan, Department of Physical Medicine and Rehabilitation. Her clinical practice includes the multidisciplinary rehabilitation services for people with multiple sclerosis, spinal cord injury, brain injury and stroke. She obtained her fellowship training in Physical Medicine and Rehabilitation at the University of Saskatchewan (2006), and her Doctor of Medicine and Kinesiology degrees at McMaster University. She is the director of the Saskatoon

261

Multiple Sclerosis Clinic at Saskatoon City Hospital. Dr. Knox is a primary investigator with the Cameco MS Neuroscience Research Center at the University of Saskatchewan. Her clinical research and health monitoring interests include those related to multiple sclerosis, mobility and health related quality of life. She teaches medical students and residents at the University of Saskatchewan and is a member of the management committee for the Western Pacific endMS Regional Research Training Center.

Dr. Adil J. Nazarali is Professor and Director, Laboratory of Molecular Cell Biology, College of Pharmacy and Nutrition University of Saskatchewan. He received his B.Sc. (Pharmacy, 1977) from Portsmouth School of Pharmacy, England, M.Sc. (Biopharmacy, 1981) from the University of London, England and a Ph.D. (Neurochemistry, 1983) from the University of Alberta, Canada. Dr. Nazarali received his postdoctoral training in molecular cell biology at the National Institutes of Health working with Nobel laureate Dr. Marshall Nirenberg. Dr. Nazarali is a principal investigator with the Cameco Multiple Sclerosis and Neuroscience Research Centre, City Hospital, Saskatoon and principal investigator with the Neuroscience Research Cluster, University of Saskatchewan. Dr. Nazarali is the co-editor of a special memorial issue of Cellular and Molecular Neurobiology honoring Dr. Marshall Nirenberg. Dr. Nazarali’s research area is to advance our understanding of how the body is able to repair damage to the brain, especially as a function of age. Dr. Nazarali’s research focus is on the determinants of myelination in multiple sclerosis, pharmacogenomics of multiple sclerosis, epigenetics and the role of master regulatory Hox genes in development.