Article Text

Fatigue in SLE: diagnostic and pathogenic impact of anti-N-methyl-D-aspartate receptor (NMDAR) autoantibodies
Free
1. Andreas Schwarting1,2,
2. Tamara Möckel1,
3. Freya Lütgendorf2,
4. Konstantinos Triantafyllias2,
5. Sophia Grella1,
6. Simone Boedecker1,
7. Arndt Weinmann1,
8. Myriam Meineck1,
9. Clemens Sommer3,
10. Ingrid Schermuly4,
11. Andreas Fellgiebel4,
12. Felix Luessi5,
13. Julia Weinmann-Menke1
1. 1 Division of Rheumatology and Clinical Immunology, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany
2. 2 Acura Rheumatology Center Rhineland Palatinate, Bad Kreuznach, Germany
3. 3 Institute of Neuropathology, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany
4. 4 Department of Psychiatry and Psychotherapy, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany
5. 5 Department of Neurology, University Medical Center of the Johannes Gutenberg University Mainz, Mainz, Germany
1. Correspondence to Andreas Schwarting, Division of Rheumatology and Clinical Immunology, University Medical Center of the Johannes-Gutenberg University Mainz, Mainz 55131, Germany; schwarting{at}uni-mainz.de

## Abstract

Objectives We explored the impact of circulating anti-N-methyl-D-aspartate receptor (NMDAR) antibodies on the severity of fatigue in patients with systemic lupus erythematosus (SLE).

Methods Serum samples of 426 patients with SLE were analysed for the presence of antibodies to the NR2 subunit of the NMDAR. In parallel, the severity of fatigue was determined according to the Fatigue Scale for Motor and Cognitive functions questionnaire. In a subgroup of patients with SLE, the hippocampal volume was correlated with the levels of anti-NR2 antibodies. Isolated immunoglobulin G from patients with anti-NR2 antibodies were used for murine immunohistochemical experiments and functional assays on neuronal cell lines. Treatment effects were studied in 86 patients with lupus under belimumab therapy.

Results We found a close correlation between the titre of anti-NR2 antibodies, the severity of fatigue, the clinical disease activity index (Systemic Lupus Erythematosus Disease Activity Index 2000) and anti-double stranded DNA antibodies—independently of the presence of neuropsychiatric lupus manifestations. Pathogenic effects could be demonstrated by (1) detection of anti-NR2 antibodies in the cerebrospinal fluid, (2) in situ binding of anti-NR2 antibodies to NMDAR of the hippocampus area and (3) distinct functional effects in vitro: downregulating the energy metabolism of neuronal cells without enhanced cytotoxicity. Treatment with belimumab for at least 6 months affected both the severity of fatigue and the levels of anti-NR2 antibodies.

Conclusion The presence of anti-NR2 antibodies in patients with SLE with fatigue is a helpful diagnostic tool and may offer a major approach in the therapeutic management of this important disabling symptom in patients with SLE.

• fatigue
• systemic lupus erythematosus
• anti-nmda receptor antibodies
• belimumab
View Full Text

## Statistics from Altmetric.com

### Key messages

• Fatigue is known as a predominant symptom of patients with sytemic lupus erythematosus (SLE) across different ethnicities resulting in substantially reduced quality of life and work disability. However, fatigue is not part of clinical activity scores (eg, Systemic Lupus Erythematosus Disease Activity Index 2000 or Systemic Lupus Erythematosus International Collaborating Clinics).

• It is well established that autoantibodies reacting with the N-methyl-D-aspartate receptor are closely associated with neurocognitive impairment in neuropsychiatric SLE.

#### What does this study add ?

• A link between one of the most challenging symptoms patients with lupus suffer from and the presence of brain-reacting autoantibodies. The study offers an objective measurement for patients with SLE complaining about fatigue in routine clinical practice.

#### How might this impact on clinical practice or future developments ?

• The measurement of anti-NR2 antibodies in patients with SLE with fatigue might be introduced into clinical disease activity scores to better address the patient-related quality of life in future clinical studies and support the development of an individualised therapy of patients with SLE.

## Introduction

Systemic lupus erythematosus (SLE) is a chronic autoimmune disease affecting multiple organ systems in a variety of clinical manifestations. A subscale analysis of quality of life (QoL) in patients with SLE revealed that fatigue, a rather subjective symptom with partly still unknown pathophysiology, may have greater effects on the physical QoL than disease activity or damage index of SLE.1 This underlines the importance of fatigue as an SLE-associated complaint occurring in 67%–90% of patients regardless of their ethnicity.2 3 Often described as a constant feeling of exhaustion, fatigue impairs physical as well as mental aspects of life and is associated with abnormal illness behaviours and work disability.2

Recently, the FATILUP study (FATigue in LUPus), which included 570 patients from the Lupus BioBank of the upper Rhein (LBBR), was conducted to identify the determinants associated with fatigue in a large cohort of european lupus patients. managed to identify determinants associated with fatigue. Definite depression and anxiety were found to be strong contributors, whereas the use of glucocorticoids, disease activity (Systemic Lupus Erythematosus Disease Activity Index 2000 (SLEDAI-2K)) and patient age were more weakly associated.4

Since fatigue itself is a complex and subjective symptom, it is so far not amenable to objective measurements; thus diagnosis is still based on questionnaires. The Fatigue Scale for Motor and Cognitive Functions (FSMC) was initially developed and validated for patients with multiple sclerosis in 2009. Consisting of 20 items, it allows a semiquantitative description of severity and a discrimination between cognitive and motor focused fatigue with overall good reliability, sensitivity and specificity values.5

One major obstacle in the process of establishing objective diagnostic tools is the lack of a complete understanding of fatigue’s pathogenesis.

With regard to SLE and its neuropsychiatric manifestations (neuropsychiatric SLE (NPSLE)), anti-double stranded (ds) DNA antibodies cross-reacting with a single epitope present in GluN2A/B subunits of the N-methyl-D-aspartate receptor (NMDAR) have been identified as pathogenic factors closely associated with NPSLE.6 7 A dose-dependent interaction between anti-NR2 antibodies and self-antigens located in the brain as well as the distribution pattern of NMDAR antibodies linking either impaired memory and hippocampal atrophy or emotional disturbances and atrophy of the amygdala underline the importance of anti-NR2 antibodies in SLE.8 9 Taking into regard the confirmed association of anti-NR2 antibodies and neuropsychiatric manifestations in patients with SLE, changes in antibody titres might possibly predict the clinical condition in individual patients as Ando et al postulated for patients with psychiatric symptoms in general.10

Based on these considerations, we sought to determine the relevance of circulating anti-NR2 antibodies in patients with SLE with fatigue.

## Methods

### Serum samples

We analysed serum samples from patients who fulfilled at least four of the American College of Rheumatology criteria (ACR) for the classification of SLE, after informed consent. Healthy volunteers (age range 18–70 years) were screened for any prior kidney diseases, diabetes, hypertension, apoplex, infection and autoimmune diseases. Freshly drawn blood samples were collected, centrifuged and sera aliquots stored at −80°C. In the subgroup of belimumab-treated patients with SLE, we have used serum samples before and after belimumab treatment. The use of these specimens was approved by the Standing Committee for Clinical Studies of the Johannes-Gutenberg University in adherence to the Declaration of Helsinki. All samples were analysed retrospectively in Mainz.

### Disease activity

Disease activity was evaluated using the SLEDAI-2K and standard serologic activity measures (C3c, C4, anti-dsDNA antibodies, creatinine, C-reactive protein (CRP), erythrocyte sedimentation rate (ESR) and urine measures (proteinuria (24 hours collection) and active sediment). The following standard values were determined: C3 (0.9–1.8 g/L), dsDNA (200–1000 IU/mL) by ELISA and proteinuria (<150 mg/24 hours) by immunoturbidimetric assay. Fatigue was assessed by the FSMC questionnaire.5

### Subjects MRI cohort

Forty outpatients with SLE were admitted at the Department of Internal Medicine of the University Medical Centre Mainz. All patients fulfilled the revised criteria of the ACR.11 Patients underwent a physical examination and laboratory testing (C3, C4). SLE status was assessed using the SLEDAI-2K.12 Sociodemographic data, SLE characteristics (ie, age at diagnosis, disease duration, cortisone dosage, SLEDAI-2K score), psychiatric, neuropsychological and structural imaging parameters of MRI scans were collected. As seven MRI sequences were incomplete, the data of 33 patients only could be analysed. None of these patients were part of the belimumab group.

The ACR criteria for neuropsychiatric lupus syndromes (NPSLE) were used to classify neuropsychiatric manifestations.13 Fourteen out of 33 patients fulfilled the ACR criteria. Demographical and clinical data did not differ between patients with SLE with and without neuropsychiatric syndromes (all p’s >0.05). Fifteen healthy controls were recruited through advertisement in a local newspaper and screened for the absence or presence of psychiatric disorders via the Stem Item Screening Questions (SSQ) from the diagnostic interview DIA-X.14 Determination of hippocampal volumes was performed using Analyse Software 150 (V.8.1; Biomedical Imaging Software System, Mayo Foundation for medical education and research, Rochester, New York, USA). Hippocampal volumetry methods have been described in details.15 Hippocampal volumes are represented as absolute volumes (left+right in mL). All participants gave written informed consent.

### Cell lines

CHP-126 neuroblastoma cells (Cat. No. 300432), CCF-STTG1 astrocytoma cells (Cat. No. 300388; Cell Lines Services, Eppelheim, Germany).

## ELISA

Anti-NR2 antibody levels in human serum were analysed using Gold Dot NR2 antibody test kit (Cat. No. GD1-001, CIS Biotech) according to the manufacturer’s instructions.

### Immunoglobulin (Ig)G purification of serum

IgGs were purified from serum of patients with SLE with high or low NR2 antibody titres by using the NAb Protein G Spin Kit (Cat. No. 89979, Thermo Scientific) according to manufacturer’s protocol. For desalting, the purified IgGs Zeba Spin Desalting Columns (Cat. No. 89889, Thermo Scientific) were used.

### Immunoprecipitation of anti-NR2-Ab

Immunoprecipitation was performed by using Protein A/G PLUS-Agarose (Cat. No. sc-2003, Santa Cruz Biotechnology) according to manufacturer’s instructions.

### Cytotoxicity and ATP quantification

The 20% purified IgGs (as described above) from patients with SLE with high titres of anti-NR2-Ab in culture media were blocked with 50 ng/µL peptide NR2B (Cat. No. crb1200337, Discovery antibodies, Cleveland, UK) overnight at 4°C on a rotating device.

We stimulated cultured cells (Roswell Park Memorial Institute medium, 20% fetal calf serum, 1% glutamin, 1% penstreptomicin) for 24 hours with purified IgGs from patients with SLE with high titres of anti-NR2-Ab, without anti-NR2-Ab or healthy control serum. After incubating for 24 hours at 37°C at 5%CO2, we analysed cytotoxicity using the ToxiLight BioAssay Kit (Cat. No. LT07-217, Lonza) and the ATP quantification using the Fluorometric and Colorimetric ATP Quantification Kit (Cat. No. PK-CA577-K354, PromoKine) following the manufacturer’s instructions.

### Immunohistochemistry

Murine brain of 5-weeks-old male MRL/MpJ-Faslpr lupus mice were fixed in 4% phosphate buffered formalin, embedded in paraffin and coronal sectioned (5 µm) by the Institute for Neuropathology of the University Medical Center, Mainz. Sections were stained for the presence of NMDA-type glutamate receptor subunit 2B (purified rabbit anti-mouse GluN2B: GluRec2C-Rb-Af300, RRID: AB_2571762, Frontier Institute). Sections were either incubated with primary GluN2B antibodies at a 1:50 dilution, purified IgGs from patients with SLE with high titres of anti-NR2-Ab (1:20), healthy control serum (1:20) or NR2 positive SLE serum which was blocked with peptide NR2B at a 1:20 dilution overnight at 4°C (details are given in the online supplementary file).

### Statistical analysis

We used the non-parametric Mann-Whitney U test for comparison between two groups and the Kruskal-Wallis test for comparisons between three or more groups. For correlation analysis, we used the Spearman correlation coefficient.

The data represent the mean±SEM and were prepared using GraphPad Prism V.7.0 (GraphPad, San Diego, California, USA).

## Results

### Fatigue is a predominant symptom in the Mainz lupus cohort

The Mainz lupus cohort comprises 569 patients with SLE. Analysis of this cohort versus 159 healthy controls demonstrated significantly increased anti-NR2 antibodies in sera of patients with SLE (figure 1A). A subgroup of 426 patients ranging from 18 to 76 years completed the FSMC questionnaire (table 1). The detailed characteristics of the lupus cohort at the University Medical Centre Mainz are given in a online supplementary table.

### Supplemental material

Table 1

Demographic and clinical data

Figure 1

Fatigue and NR2 Abs correlate in SLE. (A) Serum NR2 antibody levels in patients with SLE and healthy controls quantified by ELISA. (B) Incidence of fatigue in patients with SLE evaluated by the FSMC score (self-assessment by the patient). (C) Correlation of serum NR2 antibody levels and the severity and type (motoric, cognitive) of fatigue (evaluated by the FSMC core). (D) Correlation analysis with clinical activity markers, graph displays SLEDAI-2K versus NR2 antibodies, and table displays correlation of NR2 antibodies and serological activity markers of SLE. (E) Correlation analysis of fatigue (by FSMC score), depression (evaluated by BDI) and NR2 antibody level in patients with SLE. Demographic and patient clinical characteristics are detailed in table 1. Statistics analysed by the Mann-Whitney U test. *P<0.05, **P<0.01, ***P<0.001. Values are means±SEM. BDI, Beck’s Depression Inventory; CRP, C-reactive protein; ESR, erythrocyte sedimentation rate; FSMC, Fatigue Scale for Motor and Cognitive functions; SLE, systemic lupus erythematosus; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.

### Correlation of circulating anti-NR2 antibodies with severity of fatigue in patients with SLE

Fatigue is present in 75% of 426 consecutively assessed patients with lupus (figure 1B). Among those, more than 50% are even graded as severe fatigue based on the FSMC score as determined by Penner et al 5 (figure 1C). Using four different grades of fatigue severity, we investigated the correlation between the clinical extent of fatigue and the titre of anti-NR2 antibodies (figure 1C–E). We detected a significant difference between the level of anti-NR2 antibodies and the severity of fatigue as depicted in figure 1C. Moreover, there was a correlation between the levels of anti-NR2 antibodies in sera with the fatigue score in total as well as with its partial components representing motoric and cognitive fatigue in specific (figure 1C).

Furthermore, we investigated the relationship of anti-NR2 antibodies with other parameters of disease activity in SLE. Interestingly, anti-NR2 antibody titres correlated with SLEDAI-2K and anti-dsDNA antibodies, whereas no significant correlation was found with complement factors, ESR, CRP or renal function (figure 1D). Depression and fatigue showed a clear correlation, while the level of anti-NR2 antibodies did not correlate with the depression score (Beck’s Depression Inventory)(figure 1E).

In addition, there was no correlation between the anti-NR2 antibody levels and the presence of neuropsychiatric disease manifestations (data not shown).

### Anti-NR2 antibodies can be detected in the cerebrospinal fluid of patients with SLE and bind to the hippocampus area in situ

In 22 patients with SLE, we were able to analyse the levels of anti-NR2 antibodies in cerebrospinal fluid (CSF) samples (figure 2A). We detected low levels of anti-NR2 antibodies in CSF and simultaneously high titres of anti-NR2 antibodies in the sera (figure 2A).

Figure 2

Detection of anti-NR2 antibodies in the cerebrospinal fluid of patients with SLE and binding of SLE sera within the brain. (A) In SLE sera, high levels of anti-NR2 antibodies were detected, while the anti-NR2 antibody titres of cerebrospinal fluid were low. (B) Immunohistochemical analyses showed a strong binding of SLE sera with high titres of anti-NR2B antibodies within the brain of MRL lupus mice, predominantly to the hippocampus area. Anti-GluN2B at a 1:50 dilution showed a strong staining of the hippocampus area (a). This staining pattern was also observed with SLE serum highly positive for anti-NR2 antibodies at a 1:20 dilution (b) but not in case of using a healthy control (c) or SLE serum which was blocked with peptide 2 (d), both 1:20 diluted. (C) Top panel: correlation analysis of the hippocampal volume and NR2 antibody levels, hippocampal volume and SLE activity determined by SLEDAI-2K. Bottom panel: correlation analysis of NR2 antibody levels, fatigue score, SLEDAI-2K and C3c. Demographic and patient clinical characteristics are detailed in table 2. Statistics analysed by the Mann-Whitney U test. *P<0.05,**P<0.01. Values are means±SEM. SLE, systemic lupus erythematosus; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.

In addition, immunohistochemistry revealed a strong binding of SLE sera with high titres of anti-NR2 antibodies within the brain of MRL lupus mice (figure 2B). The expression is predominantly located at the hippocampus area (figure 2B) and can be specifically blocked by coincubation with NR2 peptides (figure 2B).

### Anti-NR2 antibodies affect the hippocampus volume in patients with lupus with fatigue

Based on the in situ results, we sought to determine whether there is a long-term effect of anti-NR2 antibodies on the hippocampus volume of 33 patients with lupus with severe fatigue (figure 2C). Demographic and clinical data of patients (MRI-cohort) and controls are given in table 2.

Table 2

Demographic and clinical data of the MRI cohort of patients with SLE

Following a 2-year observation time, a reduction in hippocampal volume was associated with high titres of anti-NR2 antibodies, the SLEDAI-2K score and low C3c complement levels, while there was no correlation with the fatigue score reported by patients (figure 2C, online supplementary table).

### Functional effects of anti-NR2 antibodies on different neuronal cells in vitro

Based on the initially illustrated strong correlation between anti-NR2 antibodies and severity of fatigue in patients with lupus, in vitro tests were performed to reveal possible pathophysiological effects.

For this purpose, two different cell lines (astrocyte CCF-STTG1, neuroblastoma CHP-126) were exposed to anti-NR2 antibody positive and negative serum samples of patients with lupus.

Those cell lines are good candidates for the in vitro experiments since both express NMDA receptors 1 and 2 as shown by North et al and Lee et al 16 17

Anti-NR2 antibodies showed a remarkable effect on cell activity by reducing the ATP production in neuronal cells (figure 3A, upper left). Moreover, this effect is mediated via the NR2 receptor since coincubation with NR2-blocking peptides completely abrogated the modulation of cell activity (figure 3A, upper right). However, a similar experimental setup measuring cytotoxicity did not show any enhanced cytotoxic effects of cultured astrocytes (figure 3B). Thus, anti-NR2 antibodies exert functional impairment of astrocytes in vitro without inducing cell death suggesting a reversible mechanism. Comparable results were observed for neuroblastom cell line (CHP-126) (data not shown).

Figure 3

Functional effects of anti-NR2 antibodies on astrocytes in vitro. (A) Influence of anti-NR2 antibodies on cell activity of cultured astrocytes. Left panel: ATP production of astrocytes incubated with SLE serum was clearly lower compared with untreated cells or astrocytes which were incubated with healthy control serum. Right panel: cell activity of astrocytes which were incubated with peptide 2 blocked SLE serum was significantly higher compared with astrocytes treated with unblocked SLE serum. (B) Cytotoxicity of anti-NR2 antibodies on astrocytes. Anti-NR2 antibodies showed no enhancing effect on cytotoxicity of cultured astrocytes. Statistics analysed by the t-test. *P<0.01. Values are means±SEM. SLE, systemic lupus erythematosus.

### Effect of belimumab therapy on fatigue and anti-NR2 antibodies

Based on the reported effects of belimumab on fatigue, we analysed the subgroup of belimumab treated patients in our SLE cohort.18 Eighty-six patients with lupus with at least 6 months under belimumab therapy performed the clinical fatigue score before and after belimumab therapy simultaneously to the evaluation of anti-NR2 antibody titre (figure 4).

Figure 4

Treatment with belimumab results in reduction of serum NR2 antibodies accompanied by an amelioration of fatigue in SLE. (A) Serum NR2 antibody levels in patients with SLE before treatment with belimumab and at least 6 month following the first treatment with belimumab quantified by ELISA. (B) Severity of total fatigue, motoric and cognitive fatigue before and under belimumab treatment. (C) Clinical activity (SLEDAI-2K, dsDNA, ESR (1 hour), C3c, C4 ad CRP) of patients with SLE before and under belimumab treatment. Demographic and patient clinical characteristics are detailed in table 1. Statistics analysed by the Mann-Whitney U test. *P<0.05, **P<0.01, ****P<0.0001. Values are means±SEM. CRP, C-reactive protein; dsDNA, double stranded DNA; ESR, erythrocyte sedimentation rate; SLE, systemic lupus erythematosus; SLEDAI-2K, Systemic Lupus Erythematosus Disease Activity Index 2000.

1. In the course of belimumab therapy (mean time under therapy 12±8.5, range from 6 to 36 months), a significant decline in anti-NR2 antibody levels was monitored (figure 4A).

2. In parallel to the decrease of anti-NR2 antibodies under therapy with belimumab, we observed a clinically significant reduction of the fatigue score, regarding the total score as well as the motoric and cognitive components (figure 4B).

3. Concerning our subgroup of patients with available data on fatigue score and anti-NR2 antibody levels under therapy with belimumab, a significant improvement in disease activity, as measured by SLEDAI-2K, was demonstrated. Laboratory parameters, such as anti-dsDNA antibodies, ESR, complement factors and CRP did not show significant differences (figure 4C).

## Discussion

Our data reveal that anti-NR2 antibodies correlated with fatigue severity in patients with SLE. To our knowledge, this is the first study to examine this relationship in a large SLE cohort. Interestingly, the found correlation was not restricted to patients with NPSLE.

While anti-NR2 antibodies could be found in 529 patients with SLE, 103 patients were excluded from further analysis if (1) the questionnaire was not completed or (2) if the date of the questionnaire did not match the date of the serum sample in our biobank (range from <2 weeks apart was allowed). However, both groups as well as the subgroup of patients with MRI or belimumab treatment did not differ regarding the clinical activity of SLE.

In general, the role of anti-NR2 antibodies in cerebrospinal fluid (CSF) and/or serum has been examined in studies regarding different NPSLE forms.9 19–24 Level of anti-NR2 antibodies in CSF have been found to correlate with both diffuse and focal NPSLE.20 21 In the study of Gono et al, serum anti-NR2A antibodies correlated with higher disease activity (SLEDAI-2K), low leucocyte counts, low haemoglobin values and higher frequency of NPSLE.22 Furthermore, Lapteva et al showed an association of serum anti-NR2 antibodies with depressive mood (but not with cognitive impairment),23 whereas Omdal et al showed an additional correlation of these antibodies with decreased short-time memory and learning abilities.24

Thus, since the pioneering studies of DeGiorgio et al 25 on the cross-reactivity of anti-dsDNA antibodies with NMDAR subunits GluN2A and GluN2B, it has been well established that these antibodies are closely associated with neurocognitive impairment in NPSLE.26–30

Here, we now report that anti-NMDAR antibodies strongly correlate with fatigue even in patients with lupus without overt signs of neuropsychiatric manifestations according to the ACR classification criteria.13

However, our data clearly show that the anti-NMDAR antibodies found in our patients had a similar pathogenic potential: (1) they could be detected both in CSF and serum, (2) the isolated IgGs from patients with lupus with fatigue bound to NMDAR in the brain in situ and (3) anti-NR2 antibodies downregulated the energy metabolism of cultured neuronal cells. Further studies of anti-NR2-mediated fatigue will elucidate whether this mechanism is unique to patients with SLE.

Increased anti-NR2 antibodies have been reported in 25%–38% of patients with SLE.22 31–33 It is until now unclear whether elevation of anti-NR2 antibodies in the CSF of patients with SLE is due to increased intrathecal synthesis or to a transport from the peripheral blood circulation to the CSF through a damaged blood–brain barrier (BBB).34 35 In the study of Kowal et al, mice which were antigen-induced to express anti-NR2 antibodies experienced apoptotic cell death only after administration of lipopolysaccharide (LPS), a substance known to lead to BBB breakdown.36 Anti-NR2 antibodies bound mainly to neurons of the hippocampus and led to cell death causing cognitive dysfunction and altered hippocampal metabolism.36 The same study group showed that mice given a combination of serum of patient with SLE with reactivity to DNA/NMDAR and LPS demonstrated cognitive impairment.37

Interestingly, states of BBB disruption such as septic meningitis associate with rise of all intrathecal autoantibodies levels, including anti-NMDAR antibodies.35 Moreover, SLE (and especially NPSLE) is characterised by increased permeability of the BBB.38 Similarly, in our study, the presence of anti-NR2 antibodies and fatigue is not restricted to patients with SLE with NPSLE.

The experimental studies point to different pathomechanisms of anti-NR2 antibodies including irreversible cell death or reduced energy metabolism.7 In our in vitro studies, we did not find a cytotoxic effect of the isolated anti-NR2 antibody IgGs, but a clear impact on ATP metabolism of neuronal cells. This fits quite well to the clinical experience that fatigue in patients with lupus responds to immunosuppressive therapy.18

By comparison, a different pathomechanism seems to be involved in limbic encephalitis caused by autoantibodies to NMDAR subunit NR1 or voltage-gated potassium channel complex resulting in receptor internalisation. However, removing the antibodies often similarly results in complete remission of the neuropsychiatric manifestation.39 40

On the other hand, other experimental studies show that pathological changes in the brain can occur long after the exposition of brain-reactive antibodies.41

To address the putative effects of long time exposure of anti-NMDAR antibodies, we investigated the changes in hippocampus volume in patients with fatigue and anti-NMDAR antibodies. Interestingly, in our study, we found a decline in the hippocampus volume in patients with lupus correlating with fatigue and circulating anti-NMDAR antibody titres over a 2-year time period. Clearly, the impact of anti-NMDAR antibodies on ‘Neuro-Lupus’ may be one factor in a complex pathophysiology involving microglia, cytokines and T cells.42

In recent years, plenty of autoantibodies against different neuronal and non-neuronal structures in the brain were discovered. The clinical importance of those autoantibodies remains often unclear. Opposite to this, our discussed link between NMDAR antibodies and fatigue is from major importance for the clinical practice: even though fatigue can be one of the most common and most challenging symptoms of SLE, it is so far not taken into consideration in the established activity scores such as SLEDAI-2K or SLICC. This can be traced back to the lack of objective measurement parameters. In controlled clinical studies, the Facit-Fatigue questionnaire is often used to assess the fatigue in the course of the study.18 We are using the FSMC questionnaire in our centre, originally introduced into MS research and recently validated for patients with SLE, since it can differentiate between cognitive and motoric fatigue.4 5 The results of our study offer a sustained clinical advantage: to add an objective measurement of fatigue in lupus patients to a subjective questionnaire. Anti-NMDAR antibodies should be identified routinely for patients with lupus suffering from fatigue.

Furthermore, the approved therapy with belimumab for patients with fatigue symptoms and proven anti-NMDAR antibodies might offer a reasonable therapeutic option.

More studies are needed to clearly define the diagnostic and therapeutic impact of the growing number of brain-reactive autoantibodies in patients with SLE.

View Abstract

• ## Lay summary

Disclaimer : This is a summary of a scientific article written by a medical professional (“the Original Article”). The Summary is written to assist non medically trained readers to understand general points of the Original Article. It is supplied “as is” without any warranty. You should note that the Original Article (and Summary) may not be fully relevant nor accurate as medical science is constantly changing and errors can occur. It is therefore very important that readers not rely on the content in the Summary and consult their medical professionals for all aspects of their health care and only rely on the Summary if directed to do so by their medical professional. Please view our full Website Terms and Conditions.
Copyright © 2019 BMJ Publishing Group Ltd & European League Against Rheumatism. Medical professionals may print copies for their and their patients and students non commercial use. Other individuals may print a single copy for their personal, non commercial use. For other uses please contact our Rights and Licensing Team.

## Footnotes

• Handling editor Josef S Smolen

• Contributors AS and JW conceived and supervised the study. AS, AW, JW, SG and MM maintained the SLE database. TM, MM, CS, AF, IS, SG and FL contributed to the acquisition of data. All authors have made substantial contributions to the analysis or interpretation of data and revised the manuscript critically for important intellectual content and approved the version for publication.

• Funding The Investigator-initiated-study was supported by GSK (data acquisition) and the Deutsche Forschungsgemeinschaft.

• Competing interests AS, TM, FL, KT, SG, SB, AW, FL, MM, CS, IS and AF have nothing to declare; AS has received speaker fees (less than US$10 000) and grant/research support by AbbVie, Novartis, Roche and GSK; KT has received speaker fees (less than US$10 000) and research support from Pfizer outside the submitted work.

• Patient and public involvement statement Indirect Patient and Public Involvement. We did not directly include PPI in this study, but the initiative of the study was driven by lupus patients and their representatives.

• Patient consent for publication Not required.

• Ethics approval The study protocol was approved by the local ethics committee (Ethics Commission of the State Chamber of Medicine in Rheinland-Pfalz, Mainz, Germany) in adherence to the Declaration of Helsinki.

• Provenance and peer review Not commissioned; externally peer reviewed.

• Data sharing statement All data relevant to the study are included in the article or uploaded as supplementary information.

## Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.