Preserved Antigen-Specific Immune Response in Patients with Multiple Sclerosis Responding to IFNβ-Therapy

Background Interferon-beta (IFNβ) regulates the expression of a complex set of pro- as well as anti-inflammatory genes. In cohorts of MS patients unstratified for therapeutic response to IFNβ, normal vaccine-specific immune responses have been observed. Data capturing antigen-specific immune responses in cohorts of subjects defined by response to IFNβ-therapy are not available. Objective To assess antigen-specific immune responses in a cohort of MS patients responding clinically and radiologically to IFNβ. Methods In 26 MS patients, clinical and MRI disease activity were assessed before and under treatment with IFNβ. Humoral and cellular immune response to influenza vaccine was prospectively characterized in these individuals, and 33 healthy controls by influenza-specific Enzyme-Linked Immunosorbent Assay (ELISA) and Enzyme Linked Immuno Spot Technique (ELISPOT). Results Related to pre-treatment disease activity, IFNβ reduced clinical and radiological MS disease-activity. Following influenza vaccination, frequencies of influenza-specific T cells and concentrations of anti-influenza A and B IgM and IgG increased comparably in MS-patients and in healthy controls. Conclusions By showing in a cohort of MS-patients responding to IFNβ vaccine-specific immune responses comparable to controls, this study indicates that antigen-specific immune responses can be preserved under successful IFNβ-therapy.


Introduction
IFNb, as all type I interferons (IFNa, IFNb, IFNe, IFNk, IFNx, and IFNv), binds to the IFNa receptor (IFNAR) [1], resulting in phosphorylation of signal transducer and activator of transcription (STAT) complexes that regulate the expression of a complex set of pro-as well as anti-inflammatory genes [2]. In patients with relapsing MS, IFNb suppresses in a portion of patients clinical and subclinical inflammatory autoimmunity via a variety of (postulated) mechanisms (reduced T cell mediated inflammation, altered function of antigen-presenting and other immune cells, stabilization of the blood-brain barrier) [3][4][5][6][7][8][9][10], while no signs of a general immunosuppressive effect have been noted. Also, non-suppressed vaccine-induced inhibition of hemagglutination suggested some degree of selectivity of IFNb in suppressing autoimmune inflammation [11,12]. However, these studies were done in cohorts of patients that were not defined with regard to their response to IFNb-therapy. Therefore, potential subclinical immuno-inhibitory effects of IFNb in subjects responding to IFNbtherapy may have been concealed. In search of a potential (subclinical) immuno-inhibitory effect of IFNb we here prospectively monitored humoral and cellular vaccine-specific immunity in a cohort of patients with MS defined by clinical and radiological response to IFNb-treatment as well as in healthy controls.

Study subjects and procedures
An open-label, observational, combined retrospective and prospective study was performed aiming (i) to assess in patients with MS the clinical and MRI response to initiation of IFNbtreatment (retrospective part) and (ii) to compare the adaptive immune response induced by influenza-vaccination in the same cohort of patients with MS under established IFNb-therapy, and in healthy controls (HC) (prospective part). The institutional review board of Basel approved the study. After written informed consent, blood samples from study subjects were obtained before and 7, 14 and 28 days after seasonal influenza-vaccination with MutagripH (Sanofi Pasteur SA, Lyon). The prospective part of the trial was conducted during the influenza-vaccination periods 2008/2009 and 2009/2010. Inclusion criteria for patients at the time of recruitment into the prospective part of the study were definite relapsing MS, treatment with IFNb, and age $18 and #65 years. Inclusion criteria for healthy controls (prospective part of the study) were absence of chronic disease, and age $18 and #65 years. Exclusion criteria for patients and controls were known hypersensitivity to the vaccine under investigation, fever at time of planned vaccination, influenza vaccination ,180 days before recruitment into the study, treatment with immunoglobulins or exogenous blood products within 90 days before recruitment into the study, simultaneous medication with steroids or immunetherapy other than IFNb and pregnancy. The institutional review board of both cantons of Basel approved the study. Retrospectively, the annualized relapse rate and the number of new T2lesions/year in MRI were assessed in the study participants with MS before and after initiation of IFNb-treatment, excluding relapses and new T2 lesions 3 months before and after initiation of IFNb-treatment. MRI data were analysed by a single neuroradiologist -which was blinded for the immunologic outcomes of our study-to reduce inter-rater variability. For the prospective assessment of the adaptive immune response induced by influenza-vaccination, blood samples from study subjects were obtained before and 7, 14 and 28 days after seasonal influenza-vaccination with MutagripH (Sanofi Pasteur SA, Lyon). Study participants were interviewed and examined before and 28 days after influenzavaccination. In patients with MS, the expanded disability status scale (EDSS) score was assessed before and under treatment with IFNb, including prospective assessments on day 0 and day 28 post vaccination. All study participants received a symptoms diary to document side effects of the vaccination, and flu-like symptoms.

Anti-influenza A and anti-influenza B enzyme-linked immunosorbent assay
Concentrations (given as virotech [VE] units/mL) of IgM and IgG anti-influenza A and anti-influenza B were determined in quadruplicates using a quantitative enzyme-linked immunosorbent assay (ELISA) according to the manufacturer (Genzyme Virotech, Ruesselsheim, Germany). As recommended by the manufacturer, seroprotection was defined as an anti-influenza A/B IgGconcentration of $10 VE/mL.

Statistical analyses
Data were tested for normality with the Shapiro-Wilk test and Levene's test was used to assess the equality of variances. Mann-Whitney test was performed in case of non-normality and/or differing variance among study-groups. Data with normal distribution were assessed by paired Student's two-sided t-test. Fisher's exact test was used for categorical analysis. Values of p, 0.05 were considered to be statistically significant.

Results
Study individuals, effects of IFNb-therapy on MS, and tolerability of influenza vaccination 26 patients with MS and 33 healthy controls were recruited into the study. Characteristics of the study population are summarized in Table 1 (upper part). In patients with MS, the annualized relapse rate decreased after initiation of treatment with IFNb from 1.28 to 0.59 (p = 0.002). Likewise, the number of new T2-lesions/ year decreased after initiation of IFNb-therapy (before IFNbtherapy: 1.8, under IFNb-therapy: 0.6; p = 0.002) ( Table 1, middle part). Importantly, all patients of our cohort hade experienced a reduction of the annualized relapse rate and in all patients in which MRI data were available had a reduction of the number of new T2-lesions/year. Following influenza-vaccination, rates of local injection site reactions were comparable in IFNbtreated patients and in HC, while general symptoms occurred significantly more frequent in IFNb-treated patients with MS (p = 0.0058) (perhaps more adequate p = 0.006 ?) ( Table 1, lower part). Patients with MS and healthy controls did not differ in the frequency of influenza vaccination in the previous years.

Humoral vaccine-specific immune response
Pre-vaccination levels of IgM directed against influenza A and B were comparably low in IFNb-treated patients and in HC. Following vaccination, concentrations of influenza A-and Bspecific IgM increased significantly and comparably in both study groups, and remained increased at comparable levels on day 28 post vaccination ( Fig. 1A/B).
Baseline IgG-levels specific for influenza A and B also were comparable in IFNb-treated patients and in HC. Influenza A-  specific IgG increased significantly and comparably in both study groups at day 14, and remained increased at comparable levels on day 28 post vaccination ( Fig. 2A). Levels of anti-influenza Bspecific IgG also increased significantly by day 7, and remained increased on day 14 and 28 post vaccination in both study groups. IFNb-treated patients mounted a more pronounced response, resulting in significantly higher levels of anti-influenza B IgG on both day 14 and day 28 post vaccination (Fig. 2B).
Before vaccination, a respective 54% and 64% of IFNb-treated patients and HC fulfilled predefined seroprotection criteria (IgG $10 VE/mL) for influenza A (p = 0.89), 50% and 46% for influenza B (p = 0.89) -indicating previous contact with antigen from these viruses in a substantial proportion of study participants (Fig. 2C/D).
At day 7 after vaccination, the proportion of individuals fulfilling seroprotection criteria towards influenza A was increased comparably in both IFNb-treated patients and in HC. By contrast, more IFNb-treated individuals fulfilled seroprotection criteria for anti-influenza B at this time point (p = 0.02). At days 14 and 28, 100% of the IFNb-treated patients fulfilled seroprotection criteria for both anti-influenza A and anti-influenza B. In HC, by contrast, on day 14 and 28 seroprotection criteria for anti-influenza A were only met by 90% and 91%, for anti-influenza B by 78% and 82% (p = 0.01 for both comparisons), respectively. Also, only a respective 75% and 67% of the HC with undetectable levels of pre-vaccination IgG against influenza A and B converted to protective antibody levels -compared to 100% in IFNb-treated patients (p = 0.03 for anti-influenza B IgG) (Fig. 2E/F). In patients with MS, no differences in vaccine-induced humoral immune responses were noted between the used IFNb-preparations.

Cellular vaccine-specific immune response
The frequency of T cells producing IFN-gamma in response to influenza-antigen was assessed by ELISpot. Before vaccination, frequencies of influenza-specific IFN-gamma secreting T cells were comparable in IFNb-treated patients and in HC, as was the number of individuals with no detectable influenza-specific cellular response. By day 7 post-vaccination, frequencies significantly increased in both groups and reached similar levels (HD: p = 0.0093; MS-IFNb: p = 0.025) (Fig. 3). Numbers of influenzaspecific T cells remained similarly increased until day 14 postvaccination in both study groups. By day 28 post-vaccination, frequencies of IFN-gamma-secreting cells contracted to prevaccination levels in both groups. The proportion of patients with a strong vaccine-specific cellular immune response (predefined cutoff: .250 SFC/10 6 PBMC) was also comparable in both groups at all post-vaccination time-points (data not shown). Of note, at all time points a tendency towards a higher frequency of vaccinespecific T cells in IFNb-treated patients was evident when in IFNb-treated patients with MS (IFNb) and healthy controls (HC) (red lines indicate the median 6 IQR). The percentage of patients fulfilling IgG seroprotection criteria for influenza A (panel C) and influenza B (panel D) is shown before (day 0) and at day 7, 14 and 28 after influenza vaccination in IFNb-treated patients with MS (IFNb) and healthy controls (HC). The percentage of IFNb-treated patients with MS (IFNb), and healthy controls (HC), converting from sero-negative pre-vaccination to seroprotection following vaccination is shown for influenza A (panel E) and influenza B (panel F) (day 7-28). * indicates p, 0.05; ** indicates p, 0.001 doi:10.1371/journal.pone.0078532.g002 compared to controls. Again, in patients with MS no differences in the vaccine-induced cellular immune response were noted between the used IFNb-preparations.

Discussion
The key observation of our study was that in individuals responding clinically and radiologically to treatment with IFNb vaccine-specific humoral and cellular immune responses are preserved.
Also, previous studies assessing vaccine-specific immune responses in patients treated with IFNb have shown no differences when compared to healthy controls or untreated patients with MS [11,12,14]. However, these studies have not been stratified with regard to therapeutic response to IFNb. Depending on the criteria for therapeutic response being used, up to 47% of the patients have been reported not to respond to treatment with IFNb [15,16]. Therefore, the above-mentioned vaccination-studies might have missed immunological effects in patients responding to IFNb-therapy. However, in comparison to controls we found in our cohort with documented reduction of the relapse rate and the number of new T2-lesions in MRI, preserved vaccine specific immune responses. This finding does not support subclinical immune-inhibitory effects of IFNb. By contrast, our data indicate that antigen-specific immune responses in IFNb-treated patients with MS are at least comparable to controls.
Besides possibly uniform immunological activity of IFNb, also pleiotropic effects of the cytokine have been discussed [17][18][19][20][21]. In our study the proportion of patients with general symptoms following vaccination, vaccination-induced influenza B seroprotection, and the rate of conversion from undetectable to protective anti-influenza B IgG levels was higher in IFNb-treated individuals. However, neither was vaccine-induced humoral immune response consistently increased, nor was the vaccine-specific cellular immune response enhanced. Therefore, a general pleiotropic effect cannot be derived from our data.
Limitations of our study are (i) the lack of a control group of MS-patients that do not respond to treatment with IFNb, missing information on correlations of the vaccine-response with (ii) a potential IFNb-induced lymphopenia and (iii) the MHC haplotype status of our study subjects. However, the broadened therapeutic options for MS patients that do not respond to first-line therapies prevented us from recruiting patients with continuous inflammatory disease activity under therapy with IFNb. Additional limitations of our study are its insufficient power to evaluate clinical endpoints (such as protection from influenza infection), the retrospective nature of the MS-disease activity assessment, and that an -albeit unlikely-MS-intrinsic effect, that has been indicated in yet small studies [17,22], cannot be ultimately excluded. However, comprehensively investigating for the first time in a cohort of patients with MS the clinical and radiologic course of disease as well as both humoral and cellular vaccinespecific immune responses, our data indicate preserved antigenspecific immunity in IFNb-treated individuals. For clinicians, knowledge of this can be informative when discussing with IFNbtreated patients questions related to vaccinations.