- Email: [email protected]
Clinical and MRI correlates of autoreactive antibodies in multiple sclerosis patients
Journal of Neuroimmunology 187 (2007) 159 – 165 www.elsevier.com/locate/jneuroim
Clinical and MRI correlates of autoreactive antibodies in multiple sclerosis patients Neeta Garg a,b,⁎, Robert Zivadinov b,c , Murali Ramanathan b,d , Irene Vasiliu e , Jaclyn Locke b , Kelly Watts c , Jordan Lema c , Jyotsna Rajeswary b , Frederick E. Munschauer b , Julian Ambrus Jr. e , Bianca Weinstock-Guttman b a Multiple Sclerosis Center, UMass Memorial Medical Center, Worcester, MA 01605, United States Baird Multiple Sclerosis Center, Jacobs Neurological Institute, Buffalo General Hospital, Buffalo, NY 14203, United States Buffalo Neuroimaging Analysis Center, Jacobs Neurological Institute, State University of New York, Buffalo, NY 14203, United States d Department of Pharmaceutical Sciences, State University of New York, Buffalo, NY 14260, United States e Department of Rheumatology and Immunology, State University of New York Buffalo, NY 14203, United States b
c
Received 26 November 2006; received in revised form 1 April 2007; accepted 10 April 2007
Abstract Background: Autoreactive antibodies (ARAB) occur more frequently in patients with multiple sclerosis (MS) than in general population and the presence of these antibodies often causes uncertainty regarding the disease course, response to therapy and the diagnosis of MS. Methods: Retrospective analyses of the ARAB, clinical and MRI data of a consecutive patient cohort of MS and clinically isolated syndrome (CIS) patients were conducted. The patients were evaluated for an extensive panel that included various subtypes of antiphospholipid antibody (APLA) including anti-phosphatidylethanolamine (APE), anti-phosphatidylserine (APS), anti-beta-2-glycoprotein-1 (ABGP), anti-cardiolipin (ACA), and several other ARAB such as antinuclear antibody (ANA), anti-neutrophilic cytoplasmic antibodies (ANCA), anti-thyroid peroxidase antibodies (ATA), anti-SS-A, and anti-SS-B antibodies. Quantitative MRI analysis was performed in a subgroup of MS patients measuring T2lesion volume (LV), T1 black hole LV and brain parenchymal fraction (BPF). Results: A total of 137 patients (mean age 44.7, 84% female) with either MS (n = 111; age: mean 46.5 ± S.D. 10.3 years; disease duration: mean 13.0 ± S.D. 10.4 years; EDSS: mean 3.2 ± S.D. 1.9) or CIS (n = 26; age: mean 37.7 ± S.D. 7.8 years; disease duration: mean 1.3 ± S.D. 1.1 years; EDSS: mean 1.0 ± S.D. 0.7) were enrolled. Among MS patients, 82 were RRMS, 26 SPMS, and 3 had PPMS. Seventy-seven (69%) of MS patients showed presence of one or more ARAB. The proportion of MS patients with APLA was 55% (61 patients); IgM subtype was most frequent. Co-occurrence of ACA and APE was more frequent in SPMS as compared to RRMS (15.4% vs. 1.2%, p = 0.012). The proportion of CIS patients with ARAB was 75% with IgM subtype being the most frequent. However, the ARAB in majority of CIS patients (9 out of 14, 64%) were transient on repeated testing. In a subgroup of 62 MS patients, quantitative MRI analysis showed significantly higher T2-LV in patients with positive APLA (15.1 ml for APLA positive vs. 6.75 ml for APLA negative) after correcting for the disease duration (p = 0.048). The patients with ATA also had significantly higher T2-LV after correction for disease duration (19.0 ml vs.8.5, p = 0.044). Conclusions: ARAB were present in more than two thirds of MS and CIS patients although most of APLA in CIS were transient. The presence of APLA in MS patients was associated with higher T2-LV. © 2007 Elsevier B.V. All rights reserved. Keywords: Autoreactive antibodies; Multiple sclerosis; Antiphospholipid antibodies
1. Introduction ⁎ Corresponding author. Multiple Sclerosis Center, UMass Memorial Medical Center, 119 Belmont Street, Worcester, MA 01605, United States. Tel.: +1 508 793 6566; fax: +1 508 793 6554. E-mail address: [email protected] (N. Garg). 0165-5728/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2007.04.008
Autoreactive antibodies (ARAB) occur more frequently in multiple sclerosis (MS) patients (Barned et al., 1995; De Keyser, 1988; Karussis et al., 1998; Tourbah et al., 1998). A
160
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165
substantial number of MS patients have elevated levels of anticardiolipin antibodies (ACA) with or without elevated antinuclear antibody (ANA) suggesting dysregulation of humoral immunity at multiple levels. Previous reports have shown an increased frequency of immunological abnormalities including presence of ANA, anti-thyroid peroxidase antibody (ATA), and anti-phospholipid antibodies (APLA) in patients with MS (Barned et al., 1995; De Keyser, 1988; Seyfert et al., 1990). The role of ARAB in MS pathophysiology and their effects on the clinical and paraclinical characteristics of the disease are still a matter of debate (Collard et al., 1997; JW et al., 1999; Roussel et al., 2000). Although a number of studies have reported atypical disease course or a predominantly optico-spinal presentation in MS patients with ARAB especially ACA (Karussis et al., 1998; JW et al., 1999), others failed to demonstrate any distinct clinical syndrome leading to the conclusion that APLA may be an epiphenomenon and a non-specific marker of central nervous system injury in these patients (Sastre-Garriga et al., 2001; Tourbah et al., 1998). The aim of this study was: 1) to characterize the clinical and MRI features of MS and clinically isolated syndrome (CIS) patients evaluated with a wide panel of ARAB including antiphosphatidylethanolamine (APE), anti-beta-2-glycoprotein-1 (ABGP), ACA, anti-phosphatidylserine antibody (APS), ANA, anti-neutrophilic cytoplasmic antibodies (ANCA), and ATA and 2) to attempt to better explain the pathogenic role of these antibodies in MS. To our knowledge, this is the first study evaluating quantitative MRI measures of disease severity in MS patients with ARAB. The sequential levels of ARAB were compared in CIS and MS patients for correlation between persistency of antibodies and the disease course.
phenomenon, and history of miscarriages were also collected. The type and duration of disease modifying therapy (DMT) was recorded. Laboratory data on subtypes and titers of APLA were collected by retrospective chart review. The brain MRI data were collected on all available patients and the quantitative analysis was performed. Qualitative spinal cord MRI data (i.e., presence of lesions and atrophy) were also collected through review of the report. 2.2. Antiphospholipid antibody assay APLA and other ARAB were tested using commercially available laboratory assay. The majority were obtained at the same local laboratory which used a standard method to measure these antibodies. An ELISA method was used to measure ANA, ANCA, SS-A, SS-B, and various subtypes of antiphospholipid antibodies. Thyroid peroxidase antibodies were measured using a chemiluminometric assay. The positivity for each subtype of APLA was defined as per the laboratory standards. ANA and ATA were also categorized as positive or negative as per the testing laboratory protocol; the titers of ≥ 1:40 were defined as positive for ANA and the titers of ≥ 2 were considered positive for ATA. The MS patients were divided into two groups based on presence or absence of any one of ARAB. The group with the positive ARAB was compared to the antibody negative group to look for specific patterns of clinical presentation at disease onset, atypical or protracted disease course, and the difference in frequency of associated autoimmune diseases in patients and family members. The distribution and persistency of ARAB was compared between MS and CIS patients. 2.3. MRI protocol and analysis
2. Materials and methods 2.1. Study population Consecutive patients with the diagnosis of MS based on McDonald criteria (McDonald et al., 2001) or CIS evaluated at the Baird MS Center/Jacobs Neurological Institute over 24 months (2003–2005) were enrolled. The study was approved by local Institutional Review Board at Buffalo General Hospital, Baird MS Center. All patients were screened (as per our center's routine protocol) for a broad panel of ARAB. The panel comprised of various subtypes of APLA including IgM, IgA, and IgG subtypes of APE, ABGP, and ACA, IgG and IgM subtype of APS, and other ARAB including ANA, ANCA, ATA, anti-SS-A, and anti-SS-B antibodies (AB). The clinical data were collected by retrospective chart review. Only patients with full data set available were included. The age, age at onset of first symptoms, clinical presentation at onset, disease course and duration, EDSS, presence of associated autoimmune diseases (AID) in patients and their family members were recorded. Additional clinical data including co-morbid medical conditions, frequency of symptoms usually attributed to antiphospholipid antibody syndrome (APLS) such as migraine, livedo reticularis, Raynaud's
From the cohort of 111 MS patients, a subgroup of 62 patients who had complete MRI quantitative analysis data available was identified. The quantitative MRI (QMRI) analysis is routinely obtained prospectively for majority of MS patients at the Baird MS Center. The QMRI measurements included: T2lesion volume (LV), T1 black hole LV and brain parenchymal fraction (BPF). These measures were compared between ARAB positive and negative MS patients. Patients underwent brain MRI using a 1.5-T General Electric Signa 4x/Lx, scanner: T2 weighted-imaging (WI), 3D-SPGR T1-WI, conventional spin-echo T1-WI and FLAIR images were obtained. Analysis was performed by individuals blinded to the clinical characteristics and ARAB test results. T2- and T1-LVs were calculated using a reliable semi-automated contouringthresholding technique for lesion segmentation, as previously described (Zivadinov et al., 2001). For brain extraction and tissue segmentation, we utilized a modified version of SIENAX cross-sectional brain atrophy analysis tool called Hybrid SIENAX (Zivadinov et al., 2004). Briefly, Hybrid SIENAX removes all non-brain, non-CSF tissue from the image volume using JIM Brain Finder Tool (Horsfield et al., 2003) on axial 3D-SPGR T1-WI scans. A three-dimensional edge-finding technique is then applied to the
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165 Table 1 Demographic characteristics MS n = 111 Sex Female Male Race Caucasian African American Age, years Age at onset years Disease duration, years EDSS Duration of DMT, years
obtained at least 2 years after initiation of DMT therapy. Among CIS patients, 15 (58%) were not on any immunotherapy, 8 (31%) on IFN-β, 3 (11%) were on GA therapy. The antibody levels in the CIS group were drawn at baseline, prior to initiation of DMT in all except one patient.
CIS n = 26 89 (80) 22 (20)
105 (95) 6 (5) 46.5 ± 10.3 (17–78) 33.6 ± 9.5 (10–57) 13.0 ± 10.4 (0.7–44) 3.2 ± 1.9 (0–9) 3.9 ± 2.4 (0–11)
161
24 (92) 2 (8) 22 (85) 4 (15) 37.7 ± 7.8 (21–52) 36.1 ± 7.7 (18–51) 1.3 ± 1.1 (0.2–6) 1.0 ± 0.7 (0–2.5) 0.6 ± 0.6 (0.1–2.2)
The frequencies are n (%) and the remaining variables are mean ± S.D. (range).
original image, using information from the “deskulled” image, to generate an estimate of the skull boundary (Smith et al., 2002). Subsequently, the deskulled brain and the skull images are registered to a standard brain map, using the skull as a scaling constraint, in order to determine a subject-specific normalization factor. The deskulled brain image is then processed with the FAST (Zhang et al., 2001) automated image segmentation tool using a three-dimensional tissue class model with K-means segmentation for estimation of initial tissue-class intensity parameters and individual tissue volumes. Brain parenchymal fraction (BPF) is calculated as follows: BPF ¼ ðGM þ WMÞ=ðGM þ WM þ CSFÞ Hybrid SIENAX is fully automated. In previous work, using the same MRI dataset, the scan–rescan variability was 0.1% for BPF. 2.4. Data analysis The SPSS (SPSS Inc., Chicago, IL) statistical analysis program was used. Appropriate non-parametric tests for independent (Mann–Whitney test) or Chi-square or Fisher tests were used for continuous and frequency variables. MRI measures were analyzed using the general linear model (GLM) with disease duration and age as a covariate. 3. Results
3.2. Frequency and distribution of ARAB in MS The distribution and subtypes of ARAB in MS patients is shown in Table 2. Seventy-seven (69%) patients showed presence of one or more ARAB. APLA was present in isolation or in combination with other antibodies in 61 (55%) patients. ANA was positive in four patients only, ANCA in one patient. None of the patients had anti-SS-A or SS-B antibody. Lupus anticoagulant was available in 78 patients and was positive in only one case, associated with elevated partial thromboplastin time (PTT) and APE and ABGP levels. PTT values were available in 65 patients: five patients had elevated PTT, 3 had associated APLA; none of these patients had clinical symptoms suggestive of APLS. None of the patients with positive APLA had evidence of thrombocytopenia. Among 61 patients positive for APLA, 13 had associated ATA and 48 had isolated APLA. Among various subtypes of APLA, IgM was most prevalent, seen in 44 (72%) patients. Antithyroid antibodies were present in 25 (22%) patients, 12 isolated and 13 were associated with APLA; of these, only 5 (20%) had associated thyroid hormone abnormalities, mostly elevated TSH. A history of associated thyroiditis (based on presence of anti-thyroid antibody and past or present thyroid hormone abnormalities) was present in 9 and autoimmune hepatitis in one in ATA positive group; the majority of these patients had significant elevation of thyroid peroxidase antibody. Sequential levels of APLA and ATA (most repeated within 6–18 months) were available in 24 patients; 8 (33%) were normal, 16 (77%) showed persistent elevation. The majority of persistent titers (12, 75%) were APE, three were ACA (PTT elevated in one), and one showed ATA. 3.3. Frequency and distribution of ARAB in CIS The presence of ARAB in CIS is shown in Table 2. The ARAB titers in majority of CIS patients were measured within Table 2 Frequency (%) and distribution of autoreactive antibodies in MS and CIS
3.1. Demographics
Autoreactive antibodies
CDMS n = 111
CIS n = 27
The demographic characteristics of patients are shown in Table 1. The study cohort consisted of 137 (mean age 44.7, 84% female) MS and CIS patients. The 111 MS (82 RRMS, 26 SPMS, and 3 PPMS) had mean disease duration of 13 years (range 0.7–44) and a median EDSS of 3.0 (range 0–9). The 26 CIS patients had a median EDSS of 1.0 (range 0–2.5). The majority of MS patients (84, 76%) were on betainterferon (IFN-β) therapy, 9 (8%) glatiramer acetate (GA), 5 (4%) immunosuppressive agents, and 13 (12%) were not on any immunotherapy. The majority of ARAB in MS patients were
ARAB positive APLA APTE ACA APS ABGP Multiple subtypes Mixed (APLA + ATA) Anti-thyroid antibodies (isolated) Anti-nuclear antibodies ANA + ANCA
77 (69%) 61 (55%) 18 (16%) 7 (6%) 7(6%) 2 (2%) 14 (13%) 13 (12%) 12 (10%) 3 (3%) 1 (1%)
20 (74%) 16 (59%) 6 (22%) 0 (0%) 2 (7%) 1 (4%) 4 (15%) 3 (11%) 4 (15%) 0 (0%) 0 (0%)
162
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165
3 months of symptom onset and prior to initiation of DMT. Twenty out of 26 (77%) were positive for one or more ARAB, mostly APLA. Overall, APLA was present, isolated (13) or in combination with ATA (3), in 16 (61%) patients. The majority of APLA were IgM subtype, present in 13 (81%). ATA was found in 7 (27%) patients, only two of these showed thyroid hormone abnormalities. No correlation was found between ARAB status and specific clinical presentation at onset in CIS patients. The sequential levels of ARAB were available in 14 patients (repeated within six months); repeat APLA were absent in 9 (64%). All patients with repeat ATA (4) showed persistent high elevation (2 with abnormal thyroid hormone levels). APLA was persistently elevated in 4, most were ACA (3). The persistent subtype of APLA was IgM in 3, IgG in one. 3.4. Clinical correlations The comparison between the ARAB positive and ARAB negative MS patients is shown in Table 3. The ARAB positive group was older than the ARAB negative group (48.0 vs. 43.0 years, p = 0.024 in a t-test). The gender ratio and age at onset were similar in the two groups. No difference was found in age of onset, clinical presentation at onset, disease course, and EDSS. However, MS patients with ARAB had longer disease duration (14.5 vs. 9.7 years, p = 0.011 in a t-test). The group with APLA were older (48.2 vs. 44.2 years, p = 0.046 in a t-test), had longer disease duration (15.0 vs. 10.6 years, p = 0.027 in a t-test), and higher EDSS compared to APLA negative patients (3.6 vs. 2.7, p = 0.024 in a Mann– Whitney test). A trend towards EDSS differences between the APLA positive and APLA negative groups was observed after correcting for disease duration in a GLM analysis (p = 0.10). Table 3 Comparison of clinical features in MS patients based on autoreactive antibody status
% Female Age, years Age of onset, years Disease course: RR-MS SP-MS PP-MS EDSS Disease duration Clinical presentation Optic neuritis Brain stem syndrome Myelopathy Other Associated autoimmune disease Family history of autoimmune disease Cerebrospinal fluid abnormalities: Beta-interferon therapy
3.5. Clinical features suggestive of APLS The clinical signs and symptoms to suggest a co-existent APLS were compared between APLA positive and negative group. In APLA positive group, 7 of 61 (11%) patients had ALPS related symptoms, 4 had migraine, 2 miscarriages, and one Raynaud's phenomenon. In APLA negative group, 11 of 50 (22%) had associated APLS symptoms: 9 had migraine, one miscarriage, and one Raynaud's phenomenon. Overall, the difference was not significant between APLA positive and negative groups. 3.6. Association with autoimmune diseases Overall, 16 (14%) patients had associated AID, thyroiditis (9) was the most common followed by psoriasis (3), ulcerative colitis (2), rheumatoid arthritis (1), and hepatitis (1). The group with positive ARAB had a higher percentage of patients with other autoimmune diseases, but the difference did not reach statistical significance (18% vs. 6%, p = 0.07). The group with positive ATA had a higher frequency of associated AID (35% vs. 8%, p = 0.003), all but one were autoimmune thyroiditis. There was no difference in the age of onset, clinical presentation, and disease course in patients with associated AID. Thirty-eight (34%) patients had positive family history of autoimmune disease; presence of ARAB was not predictive of family history of AID. The family history of AID and MS was not significantly different in ARAB positive and negative patients.
ARAB positive n = 77
ARAB negative n = 34
Pvalue
3.7. Correlation with MRI findings
82% 48.0 ± 10.1 33.9 ± 9.3
76% 43.0 ± 10.5 33.0 ± 10.0
NS 0.024 NS NS
54 (70%) 20 (26%) 3 (4%) 3.4 ± 1.9 14.5 ± 11.3
28 (82%) 6 (18%) 0 2.8 ± 1.7 9.7 ± 7.5
21 (27%) 26 (34%) 13 (17%) 17 (22%) 14 (18%) 27 (35%)
11 (32%) 8 (24%) 4 (12%) 11 (32%) 2 (6%) 11 (32%)
0.07 NS
18 of 23 (78%)
3 of 4 (75%)
NS
60 (78%)
24 (70%)
NS
Quantitative MRI measures were available for 62 MS patients. The relationship of MRI measures with ARAB, APLA and ATA status in the MS group is summarized in Table 4. There was a significant difference in T2-LV in patients with ARAB (13.5 ml in the ARAB positive group vs. 6.5 ml in the ARAB negative group, p = 0.032) in a t-test. However, the difference was not significant after correcting for the years of disease duration in GLM analysis. There was significant difference in T2-LV in patients with APLA (15.1 ml in the APLA positive group vs. 6.7 ml in the APLA negative group) that remained significant (p = 0.048) after correcting for the years of disease duration in GLM analysis. Likewise, there was significant difference in T2-LV in patients with ATA (19.0 ml in the ATA positive group vs. 8.5 ml in the ATA negative group) that remained significant (p = 0.044) after correcting for the years of disease duration in GLM analysis. Similar GLM analyses correcting for both age and disease duration were also conducted and the association between T2-LV and APLA and T2-LV and ATA remained
NS 0.011 NS
The χ test was used for analysis of frequency data, the t-test was used for continuous variables and the Mann–Whitney test was used for EDSS. 2
There was also a trend towards higher frequency of APLA in SPMS (69%) compared to RRMS patients (49%, p = 0.076, Fisher exact test), no significant difference was found in ATA between the two disease types. Co-occurrence of ACA and APE was more frequent in SPMS as compared to RRMS patients (15.4% vs. 1.2%, p = 0.012 in a Fisher exact test).
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165 Table 4 Comparison of MRI features according to autoreactive antibody status MRI measure
ARAB positive
ARAB negative
P-value
Number of patients T2 lesion volume, ml Black hole lesion volume, ml Brain parenchymal fraction
38 13.5 ± 18.4 6.2 ± 15.2 0.822 ± 0.019
24 6.5 ± 5.6 4.6 ± 11.0 0.827 ± 0.027
– 0.13 0.71 0.92
MRI measure
APLA positive
APLA negative
P-value
Number of patients T2 lesion volume, ml Black hole lesion volume, ml Brain parenchymal fraction
30 15.1 ± 19.9 6.9 ± 16.8 0.819 ± 0.018
32 6.7 ± 6.6 4.3 ± 9.9 0.829 ± 0.025
– 0.048 0.49 0.21
MRI measure
ATA positive
ATA negative
P-value
Number of patients T2 lesion volume, ml Black hole lesion volume, ml Brain parenchymal fraction
12 19.0 ± 27.1 5.1 ± 7.7 0.821 ± 0.018
49 8.5 ± 9.9 5.6 ± 14.8 0.825 ± 0.023
– 0.044 0.88 0.87
All data are mean ± S.D. The p-values are from a general linear model after correcting for disease duration.
significant (p = 0.049). No significant differences were found for BPF and in T1 black hole LV in patients with ARAB, APLA or ATA compared to the group without these antibodies. Spinal MRI was available in 75 patients; 46 (61%) of these showed one or more lesions. The frequency and distribution of spinal cord lesions was not different in patients with positive ARAB, APLA, or ATA as compared to ARAB negative patients. 4. Discussion Higher prevalence of various ARAB including antithyroid, parietal cell, and mitochondrial antibodies and ANA has previously been reported in MS patients as compared to other neurological disorders (Colosimo et al., 1997; De Keyser, 1988; Durelli et al., 2001; Karussis et al., 1998; Tourbah et al., 1998; Verdun et al., 2002). Our study of a consecutive group of MS patients at different disease stages showed high prevalence of APLA and ATA. An increased frequency of ATA and inverse correlation between antibody titers and EDSS score was observed in RRMS patients (Annunziata et al., 1999). A high frequency of ATA (22%) and an association with T2-LV in MS patients was observed in our study. High persistent ATA were seen in our CIS patients and possibly reflected coexistent thyroid disease as suggested by thyroid hormone abnormalities in these patients. It is unlikely that IFN-β therapy affected ATA positivity in CIS group as the majority of patients were not on disease modifying therapy at the time of evaluation. APLA positivity in MS patients was reported ranging between 5% and 20% (Fukazawa et al., 1993; Heinzlef et al., 2002; Karussis et al., 1998; Tourbah et al., 1998). In healthy controls, the frequency of APLA was reported to be approximately 6–9% (McIntyre et al., 2003; Vila et al., 1994) and in a recent retrospective study, APLA was found in about 15% of patients with various neurological disorders (Miesbach et al., 2006). Although a major limitation of our study is the lack of a healthy control group, the proportion of APLA positive
163
patients in our MS group was approximately 7-fold higher than reported in the general population. A broader panel to include various subtypes of APLA antibodies used in current study can explain the higher frequency as most of the previous studies have used ACA to define APLA positivity. More than three fourth of MS patients in this study were on beta-interferon therapy and that could have had some influence of ARAB positivity although previous reports have suggested insignificant and transient increase in autoantibodies with betainterferon therapy (Colosimo et al., 1997; Durelli et al., 2001). The presence of APLA in our MS patients probably reflects an ongoing increased B cell activity. The absence of associated clinical features suggestive of APLS, lupus anticoagulant, and PTT elevation in majority of MS patients with positive APLA suggests that these patients are distinct from primary or secondary APLS. The sequential levels of APLA were only available for a small number of MS patients in our study, therefore it is difficult to comment whether this is a persistent or transient response. However, the presence of APLA in this cohort of MS patients with average disease duration greater than 5 years probably reflects a persistent humoral immune response. The trend toward higher disability in APLA positive patients could support the hypothesis that APLA is possibly a marker of B-cell mediated CNS injury in a subset of progressive MS patients. The mechanisms by which APLA can induce CNS injury include molecular mimicry with myelin or other CNS antigens due to their cross reactivity to myelin, myelin-related proteins, and brain phospholipids (Karussis et al., 1998; Marullo et al., 1993), prothrombotic mechanism (Cuadrado and Khamashta, 2000) or by inducing vasospasm (Atsumi et al., 1998). Thus far, there is no convincing evidence of demyelinating potential of APLA based on clinical or animal studies. No specific morphological characteristics of MRI lesions have been described in MS patients with ARAB (Roussel et al., 2000; Tourbah et al., 1998). The MRI lesions in APLS and MS can appear similar (JW et al., 1999; Karussis et al., 1998; Molad et al., 1992). Studies comparing the MRI lesions among MS, APLS, and other systemic immune mediated diseases showed that MS patients have significantly higher burden of disease than the primary APLS patients (Paran et al., 2006; Rovaris et al., 2000). Our study found an association between APLA and ATA and T2-LV. These results suggest an adverse association between presence of these autoantibodies and T2-LV in MS patients. However, association does not represent causation and furthermore, T2 lesions can represent a wide spectrum of brain pathology ranging from reversible demyelination and edema to irreversible Wallerian degeneration. The exact mechanism involved is not clear. A weak but significant correlation was found between cerebral atrophy and lupus anticoagulant (Hachulla et al., 1998) although other prospective studies failed to demonstrate any significant difference on brain atrophy measures in patients with MS as compared to APLS (Cuadrado and Khamashta, 2000; Paran et al., 2006). Our study did not find a significant relation between the T1-LV or brain atrophy and the presence of ARAB suggesting that the association found between ARAB and T2 lesion load may underscore the role for ARAB within the inflammatory aspect of the disease (as
164
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165
measured by T2 lesion load), less so within the degenerative aspect of the disease as measured by brain atrophy and T1 hypointense LV. Although our data supports high incidence of ARAB in MS patients, their pathogenic significance is not well established as in other disorders such as Idiopathic thrombocytopenic purpura (ITP). High APLA were demonstrated in patients with ITP; both the frequency and titers of APLA were significantly higher during exacerbation (mostly IgG) than remission (Bidot et al., 2005). An inverse correlation was found between platelet count and APLA. These findings suggest that APLA may play a role in the exacerbation and remission of ITP. Also treatment with rituximab resulted in clinical remission that was associated with reduction of APLA antibodies supporting an underlying B cell pathogenesis targeted by this specific therapy (Ahn et al., 2005). Our cross-sectional data obtained at different time points during disease course including shortly after an initial acute event, as in CIS, during remission in RRMS, as well as during progressive phase in SPMS did not demonstrate any specific subtype or trends in levels of ARAB. The limitation is possibly due to the cross-sectional design of the study and the small number of patients with repeated testing. Nevertheless, the majority of autoantibodies were IgM and in CIS patients, most became undetectable on follow up suggesting a non-sustained immune response. The majority of antibodies in RRMS patients (performed at different points although more frequently during remission than during a clinical acute relapses) and SPMS were also predominantly IgM and had the tendency to persist in SPMS patients. The predominance of IgM APLA in chronic MS patients indicates a chronic immune response by B cells. While some IgM antibodies are produced by B2 cells as part of the primary humoral immune response, the majority of IgM in the body especially in the serum is produced by B1 cells and marginal zone B cells (MZB)(Anderson et al., 2006; Gunn and Brewer, 2006). In the B1 and MZB subpopulations, IgM antibody is generally produced from memory B cells, and is therefore part of a chronic immune response (Martin et al., 2001; Weller et al., 2004) It is possible that the humoral response in MS is coming predominantly from B1 and MZB cells. From a clinical point of view, IgM anti-phospholipid antibodies have been shown to induce the same clinical manifestations as IgG and IgA antibodies, and should therefore be taken seriously (Godeau and Piette, 2004; Schulze-Lohoff et al., 1989). Further prospective study will be necessary to better define the role of APLA in MS, currently our data is supportive of a more destructive disease process as evidenced by increased T2 lesion load in autoantibody positive patients. 4.1. Conclusions ARAB were present in more than two thirds of MS and CIS patients although most of APLA in CIS were transient. APLA in MS patients were associated with higher T2-LV, trend to an increased disability, and a broader antibody profile (ACA and APE) in secondary progressive patients suggesting a more severe and continuous CNS injury probably related to B cell
dysfunction. A prospective study for better defining the association of ARAB with CIS and MS disease course and their role as a possible marker for disease activity/tissue destruction is needed. References Ahn, E.R., Lander, G., Bidot, C.J., Jy, W., Ahn, Y.S., 2005. Long-term remission from life-threatening hypercoagulable state associated with lupus anticoagulant (LA) following rituximab therapy. Am. J. Hematol. 78, 127–129. Anderson, S.M., Hannum, L.G., Shlomchik, M.J., 2006. Memory B cell survival and function in the absence of secreted antibody and immune complexes on follicular dendritic cells. J. Immunol. 176, 4515–4519. Annunziata, P., Lore, F., Venturini, E., Morana, P., Guarino, E., Borghi, S., Guazzi, G.C., 1999. Early synthesis and correlation of serum anti-thyroid antibodies with clinical parameters in multiple sclerosis. J. Neurol. Sci. 168, 32–36. Atsumi, T., Khamashta, M.A., Haworth, R.S., Brooks, G., Amengual, O., Ichikawa, K., Koike, T., Hughes, G.R., 1998. Arterial disease and thrombosis in the antiphospholipid syndrome: a pathogenic role for endothelin 1. Arthritis Rheum. 41, 800–807. Barned, S., Goodman, A.D., Mattson, D.H., 1995. Frequency of anti-nuclear antibodies in multiple sclerosis. Neurology 45, 384–385. Bidot, C.J., Jy, W., Horstman, L.L., Ahn, E.R., Jimenez, J.J., Yaniz, M., Lander, G., Ahn, Y.S., 2005. Antiphospholipid antibodies in immune thrombocytopenic purpura tend to emerge in exacerbation and decline in remission. Br. J. Haematol. 128, 366–372. Collard, R.C., Koehler, R.P., Mattson, D.H., 1997. Frequency and significance of antinuclear antibodies in multiple sclerosis. Neurology 49, 857–861. Colosimo, C., Pozzilli, C., Frontoni, M., Farina, D., Koudriavtseva, T., Gasperini, C., Salvetti, M., Valesini, G., 1997. No increase of serum autoantibodies during therapy with recombinant human interferon-beta1a in relapsing–remitting multiple sclerosis. Acta Neurol. Scand. 96, 372–374. Cuadrado, M.J., Khamashta, M.A., 2000. The anti-phospholipid antibody syndrome (Hughes syndrome): therapeutic aspects. Bailliere's Best Pract. Res. Clin. Rheumatol. 14, 151–163. De Keyser, J., 1988. Autoimmunity in multiple sclerosis. Neurology 38, 371–374. Durelli, L., Ferrero, B., Oggero, A., Verdun, E., Ghezzi, A., Montanari, E., Zaffaroni, M., 2001. Liver and thyroid function and autoimmunity during interferon-beta 1b treatment for MS. Neurology 57, 1363–1370. Fukazawa, T., Moriwaka, F., Mukai, M., Hamada, T., Koike, T., Tashiro, K., 1993. Anticardiolipin antibodies in Japanese patients with multiple sclerosis. Acta Neurol. Scand. 88, 184–189. Godeau, B., Piette, J.C., 2004. The significance and treatment of antiphospholipid antibodies. Presse Med. 33, 944–952. Gunn, K.E., Brewer, J.W., 2006. Evidence that marginal zone B cells possess an enhanced secretory apparatus and exhibit superior secretory activity. J. Immunol. 177, 3791–3798. Hachulla, E., Michon-Pasturel, U., Leys, D., Pruvo, J.P., Queyrel, V., Masy, E., Arvieux, J., Caron, C., Brevet-Coupe, F., Hatron, P.Y., Devulder, B., 1998. Cerebral magnetic resonance imaging in patients with or without antiphospholipid antibodies. Lupus 7, 124–131. Heinzlef, O., Weill, B., Johanet, C., Sazdovitch, V., Caillat-Zucman, S., Tournier-Lasserve, E., Roullet, E., 2002. Anticardiolipin antibodies in patients with multiple sclerosis do not represent a subgroup of patients according to clinical, familial, and biological characteristics. J. Neurol. Neurosurg. Psychiatry 72, 647–649. Horsfield, M.A., Rovaris, M., Rocca, M.A., Rossi, P., Benedict, R.H., Filippi, M., Bakshi, R., 2003. Whole-brain atrophy in multiple sclerosis measured by two segmentation processes from various MRI sequences. J. Neurol. Sci. 216, 169–177. JW, I.J., Conti-Kelly, A.M., Greco, P., Abedi, M., Amos, M., Provenzale, J.M., Greco, T.P., 1999. Anti-phospholipid antibodies in patients with multiple sclerosis and MS-like illnesses: MS or APS? Lupus 8, 109–115. Karussis, D., Leker, R.R., Ashkenazi, A., Abramsky, O., 1998. A subgroup of multiple sclerosis patients with anticardiolipin antibodies and unusual
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165 clinical manifestations: do they represent a new nosological entity? Ann. Neurol. 44, 629–634. Martin, F., Oliver, A.M., Kearney, J.F., 2001. Marginal zone and B1 B cells unite in the early response against T-independent blood-borne particulate antigens. Immunity 14, 617–629. Marullo, S., Clauvel, J.P., Intrator, L., Danon, F., Brouet, J.C., Oksenhendler, E., 1993. Lupoid sclerosis with antiphospholipid and antimyelin antibodies. J. Rheumatol. 20, 747–749. McDonald, W.I., Compston, A., Edan, G., Goodkin, D., Hartung, H.P., Lublin, F.D., McFarland, H.F., Paty, D.W., Polman, C.H., Reingold, S.C., SandbergWollheim, M., Sibley, W., Thompson, A., van den Noort, S., Weinshenker, B.Y., Wolinsky, J.S., 2001. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann. Neurol. 50, 121–127. McIntyre, J.A., Wagenknech, D.R., Waxman, D.W., 2003. Frequency and specificities of antiphospholipid antibodies (aPL) in volunteer blood donors. Immunobiology 207, 59–63. Miesbach, W., Gilzinger, A., Gokpinar, B., Claus, D., Scharrer, I., 2006. Prevalence of antiphospholipid antibodies in patients with neurological symptoms. Clin. Neurol. Neurosurg. 108, 135–142. Molad, Y., Sidi, Y., Gornish, M., Lerner, M., Pinkhas, J., Weinberger, A., 1992. Lupus anticoagulant: correlation with magnetic resonance imaging of brain lesions. J. Rheumatol. 19, 556–561. Paran, D., Chapman, J., Korczyn, A.D., Elkayam, O., Hilkevich, O., Groozman, G.B., Levartovsky, D., Litinsky, I., Caspi, D., Segev, Y., Drory, V.E., 2006. Evoked potential studies in the antiphospholipid syndrome: differential diagnosis from multiple sclerosis. Ann. Rheum. Dis. 65, 525–528. Roussel, V., Yi, F., Jauberteau, M.O., Couderq, C., Lacombe, C., Michelet, V., Gil, R., Couratier, P., Vallat, J.M., Preud'homme, J.L., 2000. Prevalence and clinical significance of anti-phospholipid antibodies in multiple sclerosis: a study of 89 patients. J. Autoimmun. 14, 259–265. Rovaris, M., Viti, B., Ciboddo, G., Gerevini, S., Capra, R., Iannucci, G., Comi, G., Filippi, M., 2000. Brain involvement in systemic immune mediated diseases: magnetic resonance and magnetisation transfer imaging study. J. Neurol. Neurosurg. Psychiatry 68, 170–177. Sastre-Garriga, J., Reverter, J.C., Font, J., Tintore, M., Espinosa, G., Montalban, X., 2001. Anticardiolipin antibodies are not a useful screening tool in a nonselected large group of patients with multiple sclerosis. Ann. Neurol. 49, 408–411.
165
Schulze-Lohoff, E., Krapf, F., Bleil, L., Burmester, G.R., Kolble, K., Nusslein, H.G., Rodl, W., Kalden, J.R., 1989. IGM-containing immune complexes and antiphospholipid antibodies in patients with Sneddon's syndrome. Rheumatol. Int. 9, 43–48. Seyfert, S., Klapps, P., Meisel, C., Fischer, T., Junghan, U., 1990. Multiple sclerosis and other immunologic diseases. Acta Neurol. Scand. 81, 37–42. Smith, S.M., Zhang, Y., Jenkinson, M., Chen, J., Matthews, P.M., Federico, A., De Stefano, N., 2002. Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. NeuroImage 17, 479–489. Tourbah, A., Clapin, A., Gout, O., Fontaine, B., Liblau, R., Batteux, F., Stievenart, J.L., Weill, B., Lubetzki, C., Lyon-Caen, O., 1998. Systemic autoimmune features and multiple sclerosis: a 5-year follow-up study. Arch. Neurol. 55, 517–521. Verdun, E., Isoardo, G., Oggero, A., Ferrero, B., Ghezzi, A., Montanari, E., Zaffaroni, M., Durelli, L., 2002. Autoantibodies in multiple sclerosis patients before and during IFN-beta 1b treatment: are they correlated with the occurrence of autoimmune diseases? J. Interferon Cytokine Res. 22, 245–255. Vila, P., Hernandez, M.C., Lopez-Fernandez, M.F., Batlle, J., 1994. Prevalence, follow-up and clinical significance of the anticardiolipin antibodies in normal subjects. Thromb. Haemost. 72, 209–213. Weller, S., Braun, M.C., Tan, B.K., Rosenwald, A., Cordier, C., Conley, M.E., Plebani, A., Kumararatne, D.S., Bonnet, D., Tournilhac, O., Tchernia, G., Steiniger, B., Staudt, L.M., Casanova, J.L., Reynaud, C.A., Weill, J.C., 2004. Human blood IgM “memory” B cells are circulating splenic marginal zone B cells harboring a prediversified immunoglobulin repertoire. Blood 104, 3647–3654. Zhang, Y., Brady, M., Smith, S., 2001. Segmentation of brain MR images through a hidden Markov random field model and the expectationmaximization algorithm. IEEE Trans. Med. Imag. 20, 45–57. Zivadinov, R., Rudick, R.A., De Masi, R., Nasuelli, D., Ukmar, M., PozziMucelli, R.S., Grop, A., Cazzato, G., Zorzon, M., 2001. Effects of IV methylprednisolone on brain atrophy in relapsing–remitting MS. Neurology 57, 1239–1247. Zivadinov, R., Dwyer, M., Watt, K., 2004. Measurement of cerebral grey and white matter atrophy measurement from various MRI pulse sequences using different segmentation algorithms. J. Neurol. Sci. 251 (Suppl. 3), S89.
Clinical and MRI correlates of autoreactive antibodies in multiple sclerosis patients Neeta Garg a,b,⁎, Robert Zivadinov b,c , Murali Ramanathan b,d , Irene Vasiliu e , Jaclyn Locke b , Kelly Watts c , Jordan Lema c , Jyotsna Rajeswary b , Frederick E. Munschauer b , Julian Ambrus Jr. e , Bianca Weinstock-Guttman b a Multiple Sclerosis Center, UMass Memorial Medical Center, Worcester, MA 01605, United States Baird Multiple Sclerosis Center, Jacobs Neurological Institute, Buffalo General Hospital, Buffalo, NY 14203, United States Buffalo Neuroimaging Analysis Center, Jacobs Neurological Institute, State University of New York, Buffalo, NY 14203, United States d Department of Pharmaceutical Sciences, State University of New York, Buffalo, NY 14260, United States e Department of Rheumatology and Immunology, State University of New York Buffalo, NY 14203, United States b
c
Received 26 November 2006; received in revised form 1 April 2007; accepted 10 April 2007
Abstract Background: Autoreactive antibodies (ARAB) occur more frequently in patients with multiple sclerosis (MS) than in general population and the presence of these antibodies often causes uncertainty regarding the disease course, response to therapy and the diagnosis of MS. Methods: Retrospective analyses of the ARAB, clinical and MRI data of a consecutive patient cohort of MS and clinically isolated syndrome (CIS) patients were conducted. The patients were evaluated for an extensive panel that included various subtypes of antiphospholipid antibody (APLA) including anti-phosphatidylethanolamine (APE), anti-phosphatidylserine (APS), anti-beta-2-glycoprotein-1 (ABGP), anti-cardiolipin (ACA), and several other ARAB such as antinuclear antibody (ANA), anti-neutrophilic cytoplasmic antibodies (ANCA), anti-thyroid peroxidase antibodies (ATA), anti-SS-A, and anti-SS-B antibodies. Quantitative MRI analysis was performed in a subgroup of MS patients measuring T2lesion volume (LV), T1 black hole LV and brain parenchymal fraction (BPF). Results: A total of 137 patients (mean age 44.7, 84% female) with either MS (n = 111; age: mean 46.5 ± S.D. 10.3 years; disease duration: mean 13.0 ± S.D. 10.4 years; EDSS: mean 3.2 ± S.D. 1.9) or CIS (n = 26; age: mean 37.7 ± S.D. 7.8 years; disease duration: mean 1.3 ± S.D. 1.1 years; EDSS: mean 1.0 ± S.D. 0.7) were enrolled. Among MS patients, 82 were RRMS, 26 SPMS, and 3 had PPMS. Seventy-seven (69%) of MS patients showed presence of one or more ARAB. The proportion of MS patients with APLA was 55% (61 patients); IgM subtype was most frequent. Co-occurrence of ACA and APE was more frequent in SPMS as compared to RRMS (15.4% vs. 1.2%, p = 0.012). The proportion of CIS patients with ARAB was 75% with IgM subtype being the most frequent. However, the ARAB in majority of CIS patients (9 out of 14, 64%) were transient on repeated testing. In a subgroup of 62 MS patients, quantitative MRI analysis showed significantly higher T2-LV in patients with positive APLA (15.1 ml for APLA positive vs. 6.75 ml for APLA negative) after correcting for the disease duration (p = 0.048). The patients with ATA also had significantly higher T2-LV after correction for disease duration (19.0 ml vs.8.5, p = 0.044). Conclusions: ARAB were present in more than two thirds of MS and CIS patients although most of APLA in CIS were transient. The presence of APLA in MS patients was associated with higher T2-LV. © 2007 Elsevier B.V. All rights reserved. Keywords: Autoreactive antibodies; Multiple sclerosis; Antiphospholipid antibodies
1. Introduction ⁎ Corresponding author. Multiple Sclerosis Center, UMass Memorial Medical Center, 119 Belmont Street, Worcester, MA 01605, United States. Tel.: +1 508 793 6566; fax: +1 508 793 6554. E-mail address: [email protected] (N. Garg). 0165-5728/$ - see front matter © 2007 Elsevier B.V. All rights reserved. doi:10.1016/j.jneuroim.2007.04.008
Autoreactive antibodies (ARAB) occur more frequently in multiple sclerosis (MS) patients (Barned et al., 1995; De Keyser, 1988; Karussis et al., 1998; Tourbah et al., 1998). A
160
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165
substantial number of MS patients have elevated levels of anticardiolipin antibodies (ACA) with or without elevated antinuclear antibody (ANA) suggesting dysregulation of humoral immunity at multiple levels. Previous reports have shown an increased frequency of immunological abnormalities including presence of ANA, anti-thyroid peroxidase antibody (ATA), and anti-phospholipid antibodies (APLA) in patients with MS (Barned et al., 1995; De Keyser, 1988; Seyfert et al., 1990). The role of ARAB in MS pathophysiology and their effects on the clinical and paraclinical characteristics of the disease are still a matter of debate (Collard et al., 1997; JW et al., 1999; Roussel et al., 2000). Although a number of studies have reported atypical disease course or a predominantly optico-spinal presentation in MS patients with ARAB especially ACA (Karussis et al., 1998; JW et al., 1999), others failed to demonstrate any distinct clinical syndrome leading to the conclusion that APLA may be an epiphenomenon and a non-specific marker of central nervous system injury in these patients (Sastre-Garriga et al., 2001; Tourbah et al., 1998). The aim of this study was: 1) to characterize the clinical and MRI features of MS and clinically isolated syndrome (CIS) patients evaluated with a wide panel of ARAB including antiphosphatidylethanolamine (APE), anti-beta-2-glycoprotein-1 (ABGP), ACA, anti-phosphatidylserine antibody (APS), ANA, anti-neutrophilic cytoplasmic antibodies (ANCA), and ATA and 2) to attempt to better explain the pathogenic role of these antibodies in MS. To our knowledge, this is the first study evaluating quantitative MRI measures of disease severity in MS patients with ARAB. The sequential levels of ARAB were compared in CIS and MS patients for correlation between persistency of antibodies and the disease course.
phenomenon, and history of miscarriages were also collected. The type and duration of disease modifying therapy (DMT) was recorded. Laboratory data on subtypes and titers of APLA were collected by retrospective chart review. The brain MRI data were collected on all available patients and the quantitative analysis was performed. Qualitative spinal cord MRI data (i.e., presence of lesions and atrophy) were also collected through review of the report. 2.2. Antiphospholipid antibody assay APLA and other ARAB were tested using commercially available laboratory assay. The majority were obtained at the same local laboratory which used a standard method to measure these antibodies. An ELISA method was used to measure ANA, ANCA, SS-A, SS-B, and various subtypes of antiphospholipid antibodies. Thyroid peroxidase antibodies were measured using a chemiluminometric assay. The positivity for each subtype of APLA was defined as per the laboratory standards. ANA and ATA were also categorized as positive or negative as per the testing laboratory protocol; the titers of ≥ 1:40 were defined as positive for ANA and the titers of ≥ 2 were considered positive for ATA. The MS patients were divided into two groups based on presence or absence of any one of ARAB. The group with the positive ARAB was compared to the antibody negative group to look for specific patterns of clinical presentation at disease onset, atypical or protracted disease course, and the difference in frequency of associated autoimmune diseases in patients and family members. The distribution and persistency of ARAB was compared between MS and CIS patients. 2.3. MRI protocol and analysis
2. Materials and methods 2.1. Study population Consecutive patients with the diagnosis of MS based on McDonald criteria (McDonald et al., 2001) or CIS evaluated at the Baird MS Center/Jacobs Neurological Institute over 24 months (2003–2005) were enrolled. The study was approved by local Institutional Review Board at Buffalo General Hospital, Baird MS Center. All patients were screened (as per our center's routine protocol) for a broad panel of ARAB. The panel comprised of various subtypes of APLA including IgM, IgA, and IgG subtypes of APE, ABGP, and ACA, IgG and IgM subtype of APS, and other ARAB including ANA, ANCA, ATA, anti-SS-A, and anti-SS-B antibodies (AB). The clinical data were collected by retrospective chart review. Only patients with full data set available were included. The age, age at onset of first symptoms, clinical presentation at onset, disease course and duration, EDSS, presence of associated autoimmune diseases (AID) in patients and their family members were recorded. Additional clinical data including co-morbid medical conditions, frequency of symptoms usually attributed to antiphospholipid antibody syndrome (APLS) such as migraine, livedo reticularis, Raynaud's
From the cohort of 111 MS patients, a subgroup of 62 patients who had complete MRI quantitative analysis data available was identified. The quantitative MRI (QMRI) analysis is routinely obtained prospectively for majority of MS patients at the Baird MS Center. The QMRI measurements included: T2lesion volume (LV), T1 black hole LV and brain parenchymal fraction (BPF). These measures were compared between ARAB positive and negative MS patients. Patients underwent brain MRI using a 1.5-T General Electric Signa 4x/Lx, scanner: T2 weighted-imaging (WI), 3D-SPGR T1-WI, conventional spin-echo T1-WI and FLAIR images were obtained. Analysis was performed by individuals blinded to the clinical characteristics and ARAB test results. T2- and T1-LVs were calculated using a reliable semi-automated contouringthresholding technique for lesion segmentation, as previously described (Zivadinov et al., 2001). For brain extraction and tissue segmentation, we utilized a modified version of SIENAX cross-sectional brain atrophy analysis tool called Hybrid SIENAX (Zivadinov et al., 2004). Briefly, Hybrid SIENAX removes all non-brain, non-CSF tissue from the image volume using JIM Brain Finder Tool (Horsfield et al., 2003) on axial 3D-SPGR T1-WI scans. A three-dimensional edge-finding technique is then applied to the
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165 Table 1 Demographic characteristics MS n = 111 Sex Female Male Race Caucasian African American Age, years Age at onset years Disease duration, years EDSS Duration of DMT, years
obtained at least 2 years after initiation of DMT therapy. Among CIS patients, 15 (58%) were not on any immunotherapy, 8 (31%) on IFN-β, 3 (11%) were on GA therapy. The antibody levels in the CIS group were drawn at baseline, prior to initiation of DMT in all except one patient.
CIS n = 26 89 (80) 22 (20)
105 (95) 6 (5) 46.5 ± 10.3 (17–78) 33.6 ± 9.5 (10–57) 13.0 ± 10.4 (0.7–44) 3.2 ± 1.9 (0–9) 3.9 ± 2.4 (0–11)
161
24 (92) 2 (8) 22 (85) 4 (15) 37.7 ± 7.8 (21–52) 36.1 ± 7.7 (18–51) 1.3 ± 1.1 (0.2–6) 1.0 ± 0.7 (0–2.5) 0.6 ± 0.6 (0.1–2.2)
The frequencies are n (%) and the remaining variables are mean ± S.D. (range).
original image, using information from the “deskulled” image, to generate an estimate of the skull boundary (Smith et al., 2002). Subsequently, the deskulled brain and the skull images are registered to a standard brain map, using the skull as a scaling constraint, in order to determine a subject-specific normalization factor. The deskulled brain image is then processed with the FAST (Zhang et al., 2001) automated image segmentation tool using a three-dimensional tissue class model with K-means segmentation for estimation of initial tissue-class intensity parameters and individual tissue volumes. Brain parenchymal fraction (BPF) is calculated as follows: BPF ¼ ðGM þ WMÞ=ðGM þ WM þ CSFÞ Hybrid SIENAX is fully automated. In previous work, using the same MRI dataset, the scan–rescan variability was 0.1% for BPF. 2.4. Data analysis The SPSS (SPSS Inc., Chicago, IL) statistical analysis program was used. Appropriate non-parametric tests for independent (Mann–Whitney test) or Chi-square or Fisher tests were used for continuous and frequency variables. MRI measures were analyzed using the general linear model (GLM) with disease duration and age as a covariate. 3. Results
3.2. Frequency and distribution of ARAB in MS The distribution and subtypes of ARAB in MS patients is shown in Table 2. Seventy-seven (69%) patients showed presence of one or more ARAB. APLA was present in isolation or in combination with other antibodies in 61 (55%) patients. ANA was positive in four patients only, ANCA in one patient. None of the patients had anti-SS-A or SS-B antibody. Lupus anticoagulant was available in 78 patients and was positive in only one case, associated with elevated partial thromboplastin time (PTT) and APE and ABGP levels. PTT values were available in 65 patients: five patients had elevated PTT, 3 had associated APLA; none of these patients had clinical symptoms suggestive of APLS. None of the patients with positive APLA had evidence of thrombocytopenia. Among 61 patients positive for APLA, 13 had associated ATA and 48 had isolated APLA. Among various subtypes of APLA, IgM was most prevalent, seen in 44 (72%) patients. Antithyroid antibodies were present in 25 (22%) patients, 12 isolated and 13 were associated with APLA; of these, only 5 (20%) had associated thyroid hormone abnormalities, mostly elevated TSH. A history of associated thyroiditis (based on presence of anti-thyroid antibody and past or present thyroid hormone abnormalities) was present in 9 and autoimmune hepatitis in one in ATA positive group; the majority of these patients had significant elevation of thyroid peroxidase antibody. Sequential levels of APLA and ATA (most repeated within 6–18 months) were available in 24 patients; 8 (33%) were normal, 16 (77%) showed persistent elevation. The majority of persistent titers (12, 75%) were APE, three were ACA (PTT elevated in one), and one showed ATA. 3.3. Frequency and distribution of ARAB in CIS The presence of ARAB in CIS is shown in Table 2. The ARAB titers in majority of CIS patients were measured within Table 2 Frequency (%) and distribution of autoreactive antibodies in MS and CIS
3.1. Demographics
Autoreactive antibodies
CDMS n = 111
CIS n = 27
The demographic characteristics of patients are shown in Table 1. The study cohort consisted of 137 (mean age 44.7, 84% female) MS and CIS patients. The 111 MS (82 RRMS, 26 SPMS, and 3 PPMS) had mean disease duration of 13 years (range 0.7–44) and a median EDSS of 3.0 (range 0–9). The 26 CIS patients had a median EDSS of 1.0 (range 0–2.5). The majority of MS patients (84, 76%) were on betainterferon (IFN-β) therapy, 9 (8%) glatiramer acetate (GA), 5 (4%) immunosuppressive agents, and 13 (12%) were not on any immunotherapy. The majority of ARAB in MS patients were
ARAB positive APLA APTE ACA APS ABGP Multiple subtypes Mixed (APLA + ATA) Anti-thyroid antibodies (isolated) Anti-nuclear antibodies ANA + ANCA
77 (69%) 61 (55%) 18 (16%) 7 (6%) 7(6%) 2 (2%) 14 (13%) 13 (12%) 12 (10%) 3 (3%) 1 (1%)
20 (74%) 16 (59%) 6 (22%) 0 (0%) 2 (7%) 1 (4%) 4 (15%) 3 (11%) 4 (15%) 0 (0%) 0 (0%)
162
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165
3 months of symptom onset and prior to initiation of DMT. Twenty out of 26 (77%) were positive for one or more ARAB, mostly APLA. Overall, APLA was present, isolated (13) or in combination with ATA (3), in 16 (61%) patients. The majority of APLA were IgM subtype, present in 13 (81%). ATA was found in 7 (27%) patients, only two of these showed thyroid hormone abnormalities. No correlation was found between ARAB status and specific clinical presentation at onset in CIS patients. The sequential levels of ARAB were available in 14 patients (repeated within six months); repeat APLA were absent in 9 (64%). All patients with repeat ATA (4) showed persistent high elevation (2 with abnormal thyroid hormone levels). APLA was persistently elevated in 4, most were ACA (3). The persistent subtype of APLA was IgM in 3, IgG in one. 3.4. Clinical correlations The comparison between the ARAB positive and ARAB negative MS patients is shown in Table 3. The ARAB positive group was older than the ARAB negative group (48.0 vs. 43.0 years, p = 0.024 in a t-test). The gender ratio and age at onset were similar in the two groups. No difference was found in age of onset, clinical presentation at onset, disease course, and EDSS. However, MS patients with ARAB had longer disease duration (14.5 vs. 9.7 years, p = 0.011 in a t-test). The group with APLA were older (48.2 vs. 44.2 years, p = 0.046 in a t-test), had longer disease duration (15.0 vs. 10.6 years, p = 0.027 in a t-test), and higher EDSS compared to APLA negative patients (3.6 vs. 2.7, p = 0.024 in a Mann– Whitney test). A trend towards EDSS differences between the APLA positive and APLA negative groups was observed after correcting for disease duration in a GLM analysis (p = 0.10). Table 3 Comparison of clinical features in MS patients based on autoreactive antibody status
% Female Age, years Age of onset, years Disease course: RR-MS SP-MS PP-MS EDSS Disease duration Clinical presentation Optic neuritis Brain stem syndrome Myelopathy Other Associated autoimmune disease Family history of autoimmune disease Cerebrospinal fluid abnormalities: Beta-interferon therapy
3.5. Clinical features suggestive of APLS The clinical signs and symptoms to suggest a co-existent APLS were compared between APLA positive and negative group. In APLA positive group, 7 of 61 (11%) patients had ALPS related symptoms, 4 had migraine, 2 miscarriages, and one Raynaud's phenomenon. In APLA negative group, 11 of 50 (22%) had associated APLS symptoms: 9 had migraine, one miscarriage, and one Raynaud's phenomenon. Overall, the difference was not significant between APLA positive and negative groups. 3.6. Association with autoimmune diseases Overall, 16 (14%) patients had associated AID, thyroiditis (9) was the most common followed by psoriasis (3), ulcerative colitis (2), rheumatoid arthritis (1), and hepatitis (1). The group with positive ARAB had a higher percentage of patients with other autoimmune diseases, but the difference did not reach statistical significance (18% vs. 6%, p = 0.07). The group with positive ATA had a higher frequency of associated AID (35% vs. 8%, p = 0.003), all but one were autoimmune thyroiditis. There was no difference in the age of onset, clinical presentation, and disease course in patients with associated AID. Thirty-eight (34%) patients had positive family history of autoimmune disease; presence of ARAB was not predictive of family history of AID. The family history of AID and MS was not significantly different in ARAB positive and negative patients.
ARAB positive n = 77
ARAB negative n = 34
Pvalue
3.7. Correlation with MRI findings
82% 48.0 ± 10.1 33.9 ± 9.3
76% 43.0 ± 10.5 33.0 ± 10.0
NS 0.024 NS NS
54 (70%) 20 (26%) 3 (4%) 3.4 ± 1.9 14.5 ± 11.3
28 (82%) 6 (18%) 0 2.8 ± 1.7 9.7 ± 7.5
21 (27%) 26 (34%) 13 (17%) 17 (22%) 14 (18%) 27 (35%)
11 (32%) 8 (24%) 4 (12%) 11 (32%) 2 (6%) 11 (32%)
0.07 NS
18 of 23 (78%)
3 of 4 (75%)
NS
60 (78%)
24 (70%)
NS
Quantitative MRI measures were available for 62 MS patients. The relationship of MRI measures with ARAB, APLA and ATA status in the MS group is summarized in Table 4. There was a significant difference in T2-LV in patients with ARAB (13.5 ml in the ARAB positive group vs. 6.5 ml in the ARAB negative group, p = 0.032) in a t-test. However, the difference was not significant after correcting for the years of disease duration in GLM analysis. There was significant difference in T2-LV in patients with APLA (15.1 ml in the APLA positive group vs. 6.7 ml in the APLA negative group) that remained significant (p = 0.048) after correcting for the years of disease duration in GLM analysis. Likewise, there was significant difference in T2-LV in patients with ATA (19.0 ml in the ATA positive group vs. 8.5 ml in the ATA negative group) that remained significant (p = 0.044) after correcting for the years of disease duration in GLM analysis. Similar GLM analyses correcting for both age and disease duration were also conducted and the association between T2-LV and APLA and T2-LV and ATA remained
NS 0.011 NS
The χ test was used for analysis of frequency data, the t-test was used for continuous variables and the Mann–Whitney test was used for EDSS. 2
There was also a trend towards higher frequency of APLA in SPMS (69%) compared to RRMS patients (49%, p = 0.076, Fisher exact test), no significant difference was found in ATA between the two disease types. Co-occurrence of ACA and APE was more frequent in SPMS as compared to RRMS patients (15.4% vs. 1.2%, p = 0.012 in a Fisher exact test).
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165 Table 4 Comparison of MRI features according to autoreactive antibody status MRI measure
ARAB positive
ARAB negative
P-value
Number of patients T2 lesion volume, ml Black hole lesion volume, ml Brain parenchymal fraction
38 13.5 ± 18.4 6.2 ± 15.2 0.822 ± 0.019
24 6.5 ± 5.6 4.6 ± 11.0 0.827 ± 0.027
– 0.13 0.71 0.92
MRI measure
APLA positive
APLA negative
P-value
Number of patients T2 lesion volume, ml Black hole lesion volume, ml Brain parenchymal fraction
30 15.1 ± 19.9 6.9 ± 16.8 0.819 ± 0.018
32 6.7 ± 6.6 4.3 ± 9.9 0.829 ± 0.025
– 0.048 0.49 0.21
MRI measure
ATA positive
ATA negative
P-value
Number of patients T2 lesion volume, ml Black hole lesion volume, ml Brain parenchymal fraction
12 19.0 ± 27.1 5.1 ± 7.7 0.821 ± 0.018
49 8.5 ± 9.9 5.6 ± 14.8 0.825 ± 0.023
– 0.044 0.88 0.87
All data are mean ± S.D. The p-values are from a general linear model after correcting for disease duration.
significant (p = 0.049). No significant differences were found for BPF and in T1 black hole LV in patients with ARAB, APLA or ATA compared to the group without these antibodies. Spinal MRI was available in 75 patients; 46 (61%) of these showed one or more lesions. The frequency and distribution of spinal cord lesions was not different in patients with positive ARAB, APLA, or ATA as compared to ARAB negative patients. 4. Discussion Higher prevalence of various ARAB including antithyroid, parietal cell, and mitochondrial antibodies and ANA has previously been reported in MS patients as compared to other neurological disorders (Colosimo et al., 1997; De Keyser, 1988; Durelli et al., 2001; Karussis et al., 1998; Tourbah et al., 1998; Verdun et al., 2002). Our study of a consecutive group of MS patients at different disease stages showed high prevalence of APLA and ATA. An increased frequency of ATA and inverse correlation between antibody titers and EDSS score was observed in RRMS patients (Annunziata et al., 1999). A high frequency of ATA (22%) and an association with T2-LV in MS patients was observed in our study. High persistent ATA were seen in our CIS patients and possibly reflected coexistent thyroid disease as suggested by thyroid hormone abnormalities in these patients. It is unlikely that IFN-β therapy affected ATA positivity in CIS group as the majority of patients were not on disease modifying therapy at the time of evaluation. APLA positivity in MS patients was reported ranging between 5% and 20% (Fukazawa et al., 1993; Heinzlef et al., 2002; Karussis et al., 1998; Tourbah et al., 1998). In healthy controls, the frequency of APLA was reported to be approximately 6–9% (McIntyre et al., 2003; Vila et al., 1994) and in a recent retrospective study, APLA was found in about 15% of patients with various neurological disorders (Miesbach et al., 2006). Although a major limitation of our study is the lack of a healthy control group, the proportion of APLA positive
163
patients in our MS group was approximately 7-fold higher than reported in the general population. A broader panel to include various subtypes of APLA antibodies used in current study can explain the higher frequency as most of the previous studies have used ACA to define APLA positivity. More than three fourth of MS patients in this study were on beta-interferon therapy and that could have had some influence of ARAB positivity although previous reports have suggested insignificant and transient increase in autoantibodies with betainterferon therapy (Colosimo et al., 1997; Durelli et al., 2001). The presence of APLA in our MS patients probably reflects an ongoing increased B cell activity. The absence of associated clinical features suggestive of APLS, lupus anticoagulant, and PTT elevation in majority of MS patients with positive APLA suggests that these patients are distinct from primary or secondary APLS. The sequential levels of APLA were only available for a small number of MS patients in our study, therefore it is difficult to comment whether this is a persistent or transient response. However, the presence of APLA in this cohort of MS patients with average disease duration greater than 5 years probably reflects a persistent humoral immune response. The trend toward higher disability in APLA positive patients could support the hypothesis that APLA is possibly a marker of B-cell mediated CNS injury in a subset of progressive MS patients. The mechanisms by which APLA can induce CNS injury include molecular mimicry with myelin or other CNS antigens due to their cross reactivity to myelin, myelin-related proteins, and brain phospholipids (Karussis et al., 1998; Marullo et al., 1993), prothrombotic mechanism (Cuadrado and Khamashta, 2000) or by inducing vasospasm (Atsumi et al., 1998). Thus far, there is no convincing evidence of demyelinating potential of APLA based on clinical or animal studies. No specific morphological characteristics of MRI lesions have been described in MS patients with ARAB (Roussel et al., 2000; Tourbah et al., 1998). The MRI lesions in APLS and MS can appear similar (JW et al., 1999; Karussis et al., 1998; Molad et al., 1992). Studies comparing the MRI lesions among MS, APLS, and other systemic immune mediated diseases showed that MS patients have significantly higher burden of disease than the primary APLS patients (Paran et al., 2006; Rovaris et al., 2000). Our study found an association between APLA and ATA and T2-LV. These results suggest an adverse association between presence of these autoantibodies and T2-LV in MS patients. However, association does not represent causation and furthermore, T2 lesions can represent a wide spectrum of brain pathology ranging from reversible demyelination and edema to irreversible Wallerian degeneration. The exact mechanism involved is not clear. A weak but significant correlation was found between cerebral atrophy and lupus anticoagulant (Hachulla et al., 1998) although other prospective studies failed to demonstrate any significant difference on brain atrophy measures in patients with MS as compared to APLS (Cuadrado and Khamashta, 2000; Paran et al., 2006). Our study did not find a significant relation between the T1-LV or brain atrophy and the presence of ARAB suggesting that the association found between ARAB and T2 lesion load may underscore the role for ARAB within the inflammatory aspect of the disease (as
164
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165
measured by T2 lesion load), less so within the degenerative aspect of the disease as measured by brain atrophy and T1 hypointense LV. Although our data supports high incidence of ARAB in MS patients, their pathogenic significance is not well established as in other disorders such as Idiopathic thrombocytopenic purpura (ITP). High APLA were demonstrated in patients with ITP; both the frequency and titers of APLA were significantly higher during exacerbation (mostly IgG) than remission (Bidot et al., 2005). An inverse correlation was found between platelet count and APLA. These findings suggest that APLA may play a role in the exacerbation and remission of ITP. Also treatment with rituximab resulted in clinical remission that was associated with reduction of APLA antibodies supporting an underlying B cell pathogenesis targeted by this specific therapy (Ahn et al., 2005). Our cross-sectional data obtained at different time points during disease course including shortly after an initial acute event, as in CIS, during remission in RRMS, as well as during progressive phase in SPMS did not demonstrate any specific subtype or trends in levels of ARAB. The limitation is possibly due to the cross-sectional design of the study and the small number of patients with repeated testing. Nevertheless, the majority of autoantibodies were IgM and in CIS patients, most became undetectable on follow up suggesting a non-sustained immune response. The majority of antibodies in RRMS patients (performed at different points although more frequently during remission than during a clinical acute relapses) and SPMS were also predominantly IgM and had the tendency to persist in SPMS patients. The predominance of IgM APLA in chronic MS patients indicates a chronic immune response by B cells. While some IgM antibodies are produced by B2 cells as part of the primary humoral immune response, the majority of IgM in the body especially in the serum is produced by B1 cells and marginal zone B cells (MZB)(Anderson et al., 2006; Gunn and Brewer, 2006). In the B1 and MZB subpopulations, IgM antibody is generally produced from memory B cells, and is therefore part of a chronic immune response (Martin et al., 2001; Weller et al., 2004) It is possible that the humoral response in MS is coming predominantly from B1 and MZB cells. From a clinical point of view, IgM anti-phospholipid antibodies have been shown to induce the same clinical manifestations as IgG and IgA antibodies, and should therefore be taken seriously (Godeau and Piette, 2004; Schulze-Lohoff et al., 1989). Further prospective study will be necessary to better define the role of APLA in MS, currently our data is supportive of a more destructive disease process as evidenced by increased T2 lesion load in autoantibody positive patients. 4.1. Conclusions ARAB were present in more than two thirds of MS and CIS patients although most of APLA in CIS were transient. APLA in MS patients were associated with higher T2-LV, trend to an increased disability, and a broader antibody profile (ACA and APE) in secondary progressive patients suggesting a more severe and continuous CNS injury probably related to B cell
dysfunction. A prospective study for better defining the association of ARAB with CIS and MS disease course and their role as a possible marker for disease activity/tissue destruction is needed. References Ahn, E.R., Lander, G., Bidot, C.J., Jy, W., Ahn, Y.S., 2005. Long-term remission from life-threatening hypercoagulable state associated with lupus anticoagulant (LA) following rituximab therapy. Am. J. Hematol. 78, 127–129. Anderson, S.M., Hannum, L.G., Shlomchik, M.J., 2006. Memory B cell survival and function in the absence of secreted antibody and immune complexes on follicular dendritic cells. J. Immunol. 176, 4515–4519. Annunziata, P., Lore, F., Venturini, E., Morana, P., Guarino, E., Borghi, S., Guazzi, G.C., 1999. Early synthesis and correlation of serum anti-thyroid antibodies with clinical parameters in multiple sclerosis. J. Neurol. Sci. 168, 32–36. Atsumi, T., Khamashta, M.A., Haworth, R.S., Brooks, G., Amengual, O., Ichikawa, K., Koike, T., Hughes, G.R., 1998. Arterial disease and thrombosis in the antiphospholipid syndrome: a pathogenic role for endothelin 1. Arthritis Rheum. 41, 800–807. Barned, S., Goodman, A.D., Mattson, D.H., 1995. Frequency of anti-nuclear antibodies in multiple sclerosis. Neurology 45, 384–385. Bidot, C.J., Jy, W., Horstman, L.L., Ahn, E.R., Jimenez, J.J., Yaniz, M., Lander, G., Ahn, Y.S., 2005. Antiphospholipid antibodies in immune thrombocytopenic purpura tend to emerge in exacerbation and decline in remission. Br. J. Haematol. 128, 366–372. Collard, R.C., Koehler, R.P., Mattson, D.H., 1997. Frequency and significance of antinuclear antibodies in multiple sclerosis. Neurology 49, 857–861. Colosimo, C., Pozzilli, C., Frontoni, M., Farina, D., Koudriavtseva, T., Gasperini, C., Salvetti, M., Valesini, G., 1997. No increase of serum autoantibodies during therapy with recombinant human interferon-beta1a in relapsing–remitting multiple sclerosis. Acta Neurol. Scand. 96, 372–374. Cuadrado, M.J., Khamashta, M.A., 2000. The anti-phospholipid antibody syndrome (Hughes syndrome): therapeutic aspects. Bailliere's Best Pract. Res. Clin. Rheumatol. 14, 151–163. De Keyser, J., 1988. Autoimmunity in multiple sclerosis. Neurology 38, 371–374. Durelli, L., Ferrero, B., Oggero, A., Verdun, E., Ghezzi, A., Montanari, E., Zaffaroni, M., 2001. Liver and thyroid function and autoimmunity during interferon-beta 1b treatment for MS. Neurology 57, 1363–1370. Fukazawa, T., Moriwaka, F., Mukai, M., Hamada, T., Koike, T., Tashiro, K., 1993. Anticardiolipin antibodies in Japanese patients with multiple sclerosis. Acta Neurol. Scand. 88, 184–189. Godeau, B., Piette, J.C., 2004. The significance and treatment of antiphospholipid antibodies. Presse Med. 33, 944–952. Gunn, K.E., Brewer, J.W., 2006. Evidence that marginal zone B cells possess an enhanced secretory apparatus and exhibit superior secretory activity. J. Immunol. 177, 3791–3798. Hachulla, E., Michon-Pasturel, U., Leys, D., Pruvo, J.P., Queyrel, V., Masy, E., Arvieux, J., Caron, C., Brevet-Coupe, F., Hatron, P.Y., Devulder, B., 1998. Cerebral magnetic resonance imaging in patients with or without antiphospholipid antibodies. Lupus 7, 124–131. Heinzlef, O., Weill, B., Johanet, C., Sazdovitch, V., Caillat-Zucman, S., Tournier-Lasserve, E., Roullet, E., 2002. Anticardiolipin antibodies in patients with multiple sclerosis do not represent a subgroup of patients according to clinical, familial, and biological characteristics. J. Neurol. Neurosurg. Psychiatry 72, 647–649. Horsfield, M.A., Rovaris, M., Rocca, M.A., Rossi, P., Benedict, R.H., Filippi, M., Bakshi, R., 2003. Whole-brain atrophy in multiple sclerosis measured by two segmentation processes from various MRI sequences. J. Neurol. Sci. 216, 169–177. JW, I.J., Conti-Kelly, A.M., Greco, P., Abedi, M., Amos, M., Provenzale, J.M., Greco, T.P., 1999. Anti-phospholipid antibodies in patients with multiple sclerosis and MS-like illnesses: MS or APS? Lupus 8, 109–115. Karussis, D., Leker, R.R., Ashkenazi, A., Abramsky, O., 1998. A subgroup of multiple sclerosis patients with anticardiolipin antibodies and unusual
N. Garg et al. / Journal of Neuroimmunology 187 (2007) 159–165 clinical manifestations: do they represent a new nosological entity? Ann. Neurol. 44, 629–634. Martin, F., Oliver, A.M., Kearney, J.F., 2001. Marginal zone and B1 B cells unite in the early response against T-independent blood-borne particulate antigens. Immunity 14, 617–629. Marullo, S., Clauvel, J.P., Intrator, L., Danon, F., Brouet, J.C., Oksenhendler, E., 1993. Lupoid sclerosis with antiphospholipid and antimyelin antibodies. J. Rheumatol. 20, 747–749. McDonald, W.I., Compston, A., Edan, G., Goodkin, D., Hartung, H.P., Lublin, F.D., McFarland, H.F., Paty, D.W., Polman, C.H., Reingold, S.C., SandbergWollheim, M., Sibley, W., Thompson, A., van den Noort, S., Weinshenker, B.Y., Wolinsky, J.S., 2001. Recommended diagnostic criteria for multiple sclerosis: guidelines from the International Panel on the diagnosis of multiple sclerosis. Ann. Neurol. 50, 121–127. McIntyre, J.A., Wagenknech, D.R., Waxman, D.W., 2003. Frequency and specificities of antiphospholipid antibodies (aPL) in volunteer blood donors. Immunobiology 207, 59–63. Miesbach, W., Gilzinger, A., Gokpinar, B., Claus, D., Scharrer, I., 2006. Prevalence of antiphospholipid antibodies in patients with neurological symptoms. Clin. Neurol. Neurosurg. 108, 135–142. Molad, Y., Sidi, Y., Gornish, M., Lerner, M., Pinkhas, J., Weinberger, A., 1992. Lupus anticoagulant: correlation with magnetic resonance imaging of brain lesions. J. Rheumatol. 19, 556–561. Paran, D., Chapman, J., Korczyn, A.D., Elkayam, O., Hilkevich, O., Groozman, G.B., Levartovsky, D., Litinsky, I., Caspi, D., Segev, Y., Drory, V.E., 2006. Evoked potential studies in the antiphospholipid syndrome: differential diagnosis from multiple sclerosis. Ann. Rheum. Dis. 65, 525–528. Roussel, V., Yi, F., Jauberteau, M.O., Couderq, C., Lacombe, C., Michelet, V., Gil, R., Couratier, P., Vallat, J.M., Preud'homme, J.L., 2000. Prevalence and clinical significance of anti-phospholipid antibodies in multiple sclerosis: a study of 89 patients. J. Autoimmun. 14, 259–265. Rovaris, M., Viti, B., Ciboddo, G., Gerevini, S., Capra, R., Iannucci, G., Comi, G., Filippi, M., 2000. Brain involvement in systemic immune mediated diseases: magnetic resonance and magnetisation transfer imaging study. J. Neurol. Neurosurg. Psychiatry 68, 170–177. Sastre-Garriga, J., Reverter, J.C., Font, J., Tintore, M., Espinosa, G., Montalban, X., 2001. Anticardiolipin antibodies are not a useful screening tool in a nonselected large group of patients with multiple sclerosis. Ann. Neurol. 49, 408–411.
165
Schulze-Lohoff, E., Krapf, F., Bleil, L., Burmester, G.R., Kolble, K., Nusslein, H.G., Rodl, W., Kalden, J.R., 1989. IGM-containing immune complexes and antiphospholipid antibodies in patients with Sneddon's syndrome. Rheumatol. Int. 9, 43–48. Seyfert, S., Klapps, P., Meisel, C., Fischer, T., Junghan, U., 1990. Multiple sclerosis and other immunologic diseases. Acta Neurol. Scand. 81, 37–42. Smith, S.M., Zhang, Y., Jenkinson, M., Chen, J., Matthews, P.M., Federico, A., De Stefano, N., 2002. Accurate, robust, and automated longitudinal and cross-sectional brain change analysis. NeuroImage 17, 479–489. Tourbah, A., Clapin, A., Gout, O., Fontaine, B., Liblau, R., Batteux, F., Stievenart, J.L., Weill, B., Lubetzki, C., Lyon-Caen, O., 1998. Systemic autoimmune features and multiple sclerosis: a 5-year follow-up study. Arch. Neurol. 55, 517–521. Verdun, E., Isoardo, G., Oggero, A., Ferrero, B., Ghezzi, A., Montanari, E., Zaffaroni, M., Durelli, L., 2002. Autoantibodies in multiple sclerosis patients before and during IFN-beta 1b treatment: are they correlated with the occurrence of autoimmune diseases? J. Interferon Cytokine Res. 22, 245–255. Vila, P., Hernandez, M.C., Lopez-Fernandez, M.F., Batlle, J., 1994. Prevalence, follow-up and clinical significance of the anticardiolipin antibodies in normal subjects. Thromb. Haemost. 72, 209–213. Weller, S., Braun, M.C., Tan, B.K., Rosenwald, A., Cordier, C., Conley, M.E., Plebani, A., Kumararatne, D.S., Bonnet, D., Tournilhac, O., Tchernia, G., Steiniger, B., Staudt, L.M., Casanova, J.L., Reynaud, C.A., Weill, J.C., 2004. Human blood IgM “memory” B cells are circulating splenic marginal zone B cells harboring a prediversified immunoglobulin repertoire. Blood 104, 3647–3654. Zhang, Y., Brady, M., Smith, S., 2001. Segmentation of brain MR images through a hidden Markov random field model and the expectationmaximization algorithm. IEEE Trans. Med. Imag. 20, 45–57. Zivadinov, R., Rudick, R.A., De Masi, R., Nasuelli, D., Ukmar, M., PozziMucelli, R.S., Grop, A., Cazzato, G., Zorzon, M., 2001. Effects of IV methylprednisolone on brain atrophy in relapsing–remitting MS. Neurology 57, 1239–1247. Zivadinov, R., Dwyer, M., Watt, K., 2004. Measurement of cerebral grey and white matter atrophy measurement from various MRI pulse sequences using different segmentation algorithms. J. Neurol. Sci. 251 (Suppl. 3), S89.