CASE REPORT
Nonvasculitic Autoimmune Meningoencephalitis After Rituximab The Potential Downside of Depleting Regulatory B Cells in the Brain Indira Hadley, MD,*Þ Richa Jain, MD,þ and Antoine Sreih, MD§
Abstract: Nonvasculitic autoimmune meningoencephalitis (NAIM) is a rare condition describing a syndrome of steroid-responsive encephalopathy in patients with similar clinical and pathologic features. It can be associated with autoimmune diseases, such as Sjogren’s syndrome and autoimmune thyroiditis. Brain biopsies usually show inflammatory cells without evidence of vasculitis. In this article, we present a patient who developed NAIM after receiving rituximab, a B-cellYdepleting therapy for rheumatoid arthritis. The brain biopsy showed a lack of B lymphocytes in the brain tissue, and the patient responded well to intravenous immunoglobulins. We further discuss the role of B lymphocytes and specific regulatory B lymphocytes in suppressing autoimmunity in the brain and propose that the depletion of regulatory B cells may contribute to the pathogenesis of NAIM. This case illustrates a potential side effect of rituximab and demonstrates the importance of regulatory B cells in maintaining the immune response. Key Words: nonvasculitic autoimmune meningoencephalitis (NAIM), rituximab, regulatory B cells (J Clin Rheumatol 2014;20: 163Y166)
N
onvasculitic autoimmune meningoencephalitis (NAIM) is an uncommon but potentially fatal autoimmune disease. It was first described by Caselli et al1 in 5 patients with corticosteroidresponsive encephalopathy. Nonvasculitic autoimmune meningoencephalitis has been associated with thyroid autoimmunity, referred to as Hashimoto encephalopathy, and with Sjogren’s syndrome and possibly systemic lupus erythematosus.2,3 B lymphocytes are considered a vital part of the immune system. They play a major role in the adaptive immune response by producing antibodies and developing into memory cells. They are also antigen-presenting cells and express major histocompatibility complex class II and costimulatory molecules, CD80 and CD86, and produce inflammatory cytokines.4 More recently, a new suppressive function of B lymphocytes has been identified. This functional subset of B lymphocytes, called regulatory B cells, has the role of maintaining the fine equilibrium required for tolerance. They prevent the excessive inflammatory responses that occur in autoimmune diseases or prolonged infections.5 The depletion of regulatory B cells has been shown to worsen experimental autoimmune encephalomyelitis (EAE) in a From the *John H. Stroger Jr. Hospital of Cook County, Chicago, IL; Departments of †Medicine, and ‡Pathology, Rush University School of Medicine, Chicago, IL; and §Department of Medicine, The University of Pennsylvania Penn Tower, Philadelphia, PA. Supported by the Barbara and Jacob Rukel fund. The authors declare no conflict of interest. Correspondence: Indira Hadley, MD, John H. Stroger Jr. Hospital of Cook County, 1700 West Polk St, Suite 748, Chicago, IL 60612. E-mail: [email protected]. Copyright * 2014 by Lippincott Williams & Wilkins ISSN: 1076-1608/14/2003Y0163 DOI: 10.1097/RHU.0000000000000099
JCR: Journal of Clinical Rheumatology
&
mouse model of multiple sclerosis.6 Here, we present a human case, the first to our knowledge, in whom the depletion of B cells in the brain may have led to NAIM, a histopathology similar to murine EAE. This is a patient with rheumatoid arthritis (RA) who developed NAIM after rituximab treatment. We also discuss the potential protective role of regulatory B cells in the central nervous system (CNS) and the downside of their depletion.
CASE REPORT A 67-year-old white woman with a long-standing history of RA, diagnosed at age 20 years, who was receiving rituximab twice per year came to Chicago from Phoenix. She had cataract surgery 3 weeks before traveling and experienced self-limited diarrhea postoperatively. Three days after her arrival in Chicago, she began experiencing nausea, vomiting, abdominal pain, and diarrhea. On presentation, she had a fever of 102.6-F, respiratory rate of 18, pulse of 104, and blood pressure of 143/79 but was alert and oriented. On examination, she was noted to have ulnar deviation with metacarpophalangeal joint subluxation of both hands. There were no signs of active joint synovitis. Her initial laboratory tests showed a white blood cell count (WBC) of 8.0 (4.8Y10.8), hemoglobin of 13.1 (12.0Y16.0), and platelet count of 314 (150Y450) with a normal differential. Her creatinine was 1.2 (0.6Y1.4), and results of liver function tests were normal. The lipase level was 44 (8Y57). Two days into her hospital stay, she was noted to become increasingly confused and agitated. She developed jerking movements of her upper extremity and eventually became more incoherent. She was placed on broad-spectrum antibiotics with intravenous (IV) and oral vancomycin, linezolid, metronidazole, and antifungal antibiotics. She had an extensive infectious workup, including negative blood cultures, Herpes simplex virus 1 and 2, West Nile virus, QuantiFERON Gold, stool cultures, stool ova and parasite, and Clostridium difficile toxin. Her initial urinalysis showed protein greater than 300 and red blood cell count of 3 to 5, and her urine culture grew vancomycin-resistant enterococcus, which was later thought to be a contaminant. A repeat urinalysis done several weeks later showed normal findings. Other laboratory work included an elevated erythrocyte sedimentation rate of 35; a C-reactive protein of 5; and an elevated creatine kinase of 349, which rose to 2000 and then decreased to 508 during a course of 10 days. The thyroid-stimulating hormone level was normal, but the patient had mildly positive antithyroglobulin and antithyroid peroxidase antibodies. An autoimmune panel was negative for antinuclear, antiYSjogren’s syndrome A, antiYSjogren’s syndrome B, and antineutrophil cytoplasmic antibodies; showed an elevated rheumatoid factor of 84; and showed normal complement levels. A computed tomography (CT) of the chest showed fibrotic changes in the lung bases. A CT of the abdomen and the pelvis showed distal esophageal thickening, extensive thumbprinting, and mucosal edema throughout the distal sigmoid and the rectum, which were suggestive of acute infectious
Volume 20, Number 3, April 2014
www.jclinrheum.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
163
JCR: Journal of Clinical Rheumatology
Hadley et al
colitis or ischemic colitis. An esophageal gastroduodenoscopy and colonoscopy showed mild reflux esophagitis in the distal esophagus and a few erosions in the stomach. There was submucosal hemorrhage with edema and hyperemia in the sigmoid region and diffuse diverticulosis. An antral biopsy showed mild gastritis, and a sigmoid colon biopsy showed colonic mucosa with limited pathologic change and diffuse hyperemia. No evidence of vasculitis was seen. At the onset of the altered mental status, a magnetic resonance imaging (MRI) of the brain showed extensive patchy and fluffy areas of abnormal white matter signal within the deep white matter of the frontoparietal region, suggesting significant white matter disease (Fig. 1). A previous MRI done in 2002 was unremarkable. A lumbar puncture showed 33 red blood cells, 28 WBCs with 82% lymphocytes, glucose of 60 mg/dL, and protein of 40 mg/dL. Results of acid-fast bacilli, cryptococcal antigen, and John Cunningham (JC) polyoma virus polymerase chain reaction were all negative. Cerebrospinal fluid (CSF) was also negative for malignancy but showed small reactive lymphocytes. The immunoglobulin G to albumin ratio was within normal limits, and no oligoclonal bands were seen. Only 1% of CD19 cells were detected on flow cytometry in the CSF. The patient had 2 repeat lumbar punctures, which still showed an elevated WBC with elevated lymphocytes and mildly elevated protein. Additional infectious workup was negative for JC virus antibody in the CSF; West Nile virus, enterovirus RNA, and arbovirus antibodies; and Tropheryma whippeli, coccidioidies, and bartonella antibodies. Blood flow cytometry showed a high percentage of T cells and absent B cells. The T cells showed normal cell surface antigen expression, and there was no
FIGURE 1. Precontrast and postcontrast MRI of the brain showing multiple scattered nonenhancing T2/fluid attenuated inversion recovery hyperintense areas noted in the subcortical and periventricular white matter. Technique included sagittal and axial precontrast and postcontrast T1-weighted acquisitions, axial T2-weighted fluid attenuated inversion recovery susceptibility weighted acquisition, diffusion-weighted acquisition, and apparent diffusion coefficient map. Postcontrast examination was done after IV infusion of 13 mm of multihance.
164
www.jclinrheum.com
&
Volume 20, Number 3, April 2014
FIGURE 2. Hematoxylin and eosin, 400. There is white matter parenchymal and perivascular inflammation. The brain biopsy was fixed in 10% buffered formalin followed by paraffin embedding and hematoxylin and eosin staining.
evidence of lymphoma. A CT angiogram of the brain showed a normal circle of Willis, and the cerebral angiogram showed no evidence of vasculitis. The electroencephalogram (EEG) showed continuous, diffuse, moderate-to-severe slowing but no epileptiform activity. Eventually, the patient was treated with pulse IV methylprednisolone for 3 days because of the lack of clinical response to antibiotics and antifungals. A repeat MRI of the brain several days after the course of pulse corticosteroids revealed a new lesion in the left frontal pole resembling a demyelinating lesion. A stereotactic brain biopsy was performed and showed white matter tissue with perivascular and parenchymal inflammation comprising predominantly lymphocytes. The blood vessel walls were intact without endothelial cell swelling or inflammatory cells infiltrating the vessel wall, indicating the absence of vasculitis (Fig. 2). There was no evidence of demyelination as determined by a special stain for myelin (Luxol fast blue). Strong CD3 immunoreactivity was noted in the infiltrate characterizing the lymphocytes as T cells (Figs. 3, 4). As expected, there was complete absence of CD20-reactive B lymphocytes (Fig. 5).
FIGURE 3. 200. CD3 immunostain highlights T cells within the parenchymal infiltrate. All immunohistochemical stains were performed on Leica Autostainer 48. The monoclonal antibody used against CD3 antigen was a prediluted clone LM10 from Leica. * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JCR: Journal of Clinical Rheumatology
&
Volume 20, Number 3, April 2014
FIGURE 4. 400. The perivascular lymphocytes are also CD3-reactive T lymphocytes. The CD3 immunostain used here is the same as that in Figure 3.
In addition, there were foci of microgliosis, which were highlighted by CD68 immunostain. However, there were no microglial nodules, infarctions, hemorrhages, or intranuclear or cytoplasmic inclusions (viral cytopathic effects). On the basis of the brain biopsy results, a diagnosis of NAIM was made. Because of the brain biopsy results and persistent decline in cognitive function despite IV corticosteroids, she was treated with IV immunoglobulins (IVIG), with a total of 2 g/kg given for 2 days. By the second day of IVIG treatment, the patient was more responsive and following some verbal commands. By the third day after treatment, she was communicating verbally but with signs of expressive aphasia. By the fourth day, the patient was communicating with family members, recognized familiar faces, and was reading words. She eventually went back to Phoenix, where she continued to improve. Repeat blood work after treatment revealed both normal urinalysis results and creatine kinase of 53. A repeat CT of the chest, the abdomen, and the pelvis showed resolution of small and large bowel wall thickening. A brain MRI without contrast done 5 months after her initial presentation showed bihemispheric white matter disease thought to reflect sequelae of prior small vessel ischemic disease. There was no abnormal parenchymal or leptomeningeal enhancement.
DISCUSSION Nonvasculitic autoimmune meningoencephalitis is an uncommon but potentially fatal syndrome. The symptoms of NAIM were first described as progressive dementia in a 56-year-old woman with primary Sjogren’s syndrome whose brain biopsy showed perivascular lymphocytic inflammation in the leptomeningeal and parenchymal vessels.7 Nonvasculitic autoimmune meningoencephalitis was later found to be associated with thyroid autoimmunity, referred to as Hashimoto encephalopathy2; Sjogren’s syndrome; and possibly, systemic lupus erythematosus.3 The term NAIM was first proposed in 1999 by Caselli et al1 to describe a syndrome of corticosteroid-responsive encephalopathy in 5 patients with similar clinical and pathologic features. Four cases had an associated autoimmune disease, and all had progressive cognitive decline. Brain biopsy showed perivascular lymphocytic infiltration without vessel wall involvement. Typical EEG in patients with NAIM show moderate-to-severe nonspecific dysrhythmic slowing, and spinal fluid may show an elevated immunoglobulin G index and lymphocytic pleocytosis.8 Our patient had an underlying * 2014 Lippincott Williams & Wilkins
NAIM After Rituximab
autoimmune disease, mildly positive antithyroglobulin and antithyroid peroxidase antibodies, diffuse moderate-to-severe slowing on EEG, lymphocytic pleocytosis in her CSF, and a brain biopsy revealing parenchymal and perivascular inflammatory infiltrate without evidence of demyelination or vasculitis, all of which are suggestive of NAIM. Of interest, NAIM was previously described in a 56-year-old woman with an elevated rheumatoid factor who was initially given the diagnosis of Alzheimer dementia9; however, it has never been reported in a patient with RA. Our patient has received 5 cycles of rituximab, with the last infusion being 5 months before her clinical deterioration. Rituximab has been associated with CNS demyelination10 but has not been previously reported to cause NAIM. Finally, the presence of mild elevation in antithyroglobulin and thyroid peroxidase antibodies may have predisposed her to developing NAIM. Targeting B-cell activity has shown to be effective in patients with refractory RA. B cells have been known to play a key role in the pathogenesis of RA since the discovery of antibodies to rheumatoid factor and citrullinated proteins.11 The role of B cells in autoimmune diseases is varied and includes direct cytotoxic effects of autoantibodies, immune complexYmediated destruction of organs, and B-cell mediated costimulation of T and dendritic cells through upregulation of cell surface molecules.12 Rituximab targets CD20, a specific surface antigen expressed on pre-B and mature B cells, but CD20 is also expressed in low levels on T and NK cells. As a result, the numbers of T and NK cells have been shown to be reduced during rituximab treatment in RA patients.13 B cells play a role in regulating the T-cell response and in regulating the magnitude of the immune response through antigen presentation to CD4-positive autoreactive T cells.14 More recently, it has been shown to reduce the T helper 17 response.15 Regulatory B cells, a functional subset of B cells, play an important role in regulating immune responses. The role of regulatory B cells has been studied most in the mouse model of human multiple sclerosis called EAE. The disease is exacerbated in mice deficient in B cells because of genetic ablation. B cellY deficient mice were susceptible to EAE induction and exhibited a chronic disease course.6 Furthermore, preferential depletion of regulatory B cells in vivo during disease initiation has been found to increase EAE pathogenesis. Regulatory B cells directly influence production of proinflammatory cytokines by CD4+ T cells and suppress antigen-presenting function of dendritic cells.16 B cells also play a role in preventing autoimmunity by controlling the number of T regulatory cells. In the same mouse model, antiCD20 depletion of B cells caused a reduction in the absolute number of T regulatory cells and resulted in severe colitis, a T-cellYmediated autoimmune condition. Our patient was also
FIGURE 5. 200. There is absence of CD20-positive B lymphocytes in the parenchymal and perivascular inflammatory infiltrate. The monoclonal antibody used against CD20 antigen was a prediluted clone MJ1 from Leica. www.jclinrheum.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
165
JCR: Journal of Clinical Rheumatology
Hadley et al
found to have severe colitis on CT of the abdomen and the pelvis during her initial presentation. In addition, regulatory B cell subsets, in particular the interleukin 10Yproducing B cells (B10), have been shown to be protective of EAE. Removal of B10 cells subset before EAE induction causes exacerbation of disease symptoms by influencing T cells within the CNS.17,18 We hypothesize that the depletion of B cells, in particular regulatory B cells and possibly B10 cells, in our patient resulted in loss of the immune regulatory function and an attack against undetermined CNS autoantigens causing NAIM. Our patient received her last rituximab infusion 5 months before the onset of symptoms, which led to complete deletion of her B lymphocytes as evidenced by the absence of B lymphocytes on blood and CSF flow cytometry and brain biopsy immunostaining. The effects of rituximab on serum B-cell depletion lasts for roughly 3 to 6 months, and B cells may not begin to repopulate until at least 8 months after treatment.19 Furthermore, rituximab has been found to penetrate an intact blood-brain barrier up to 6 months after the first infusion20 and has been linked to progressive multifocal leucoencephalopathy (PML), a rare and often fatal infection of the CNS caused by the JC polyoma virus. In 1 case series of rituximab-associated PML in RA patients, the majority developed PML 5 to 7 months after several cycles of rituximab infusion were given.21 Our patient responded well to IV immunoglobulins. We suspect that IVIG replaced the immune-mediated effects of regulatory B cells by inhibiting cytokines and complement activation, interfering with Fc receptor binding, and blocking immunoregulatory and adhesion molecules. The 1 previous case report of a patient with CNS demyelination after rituximab therapy also responded to IVIG, as did our patient.10 It is unclear as to why NAIM is not more frequently seen with rituximab therapy. We suspect that genetic predisposition or the presence of specific autoimmune diseases such as Hashimoto thyroiditis or Sjogren’s disease may increase the risk for developing NAIM. To our knowledge, this is the first human case of autoimmune meningoencephalitis that may have resulted from the depletion of B cells and specific regulatory B cells. Although much is still unknown about the exact role of regulatory B cells and the association between the depletion of regulatory B cells and NAIM, we hypothesize that their depletion with prior rituximab therapy played a role in the pathogenesis of NAIM in our patient. Specific depletion of cytotoxic B cells while preserving the regulatory B cells may potentially prove to be a safer approach for the treatment of autoimmune diseases. REFERENCES 1. Caselli RJ, Boeve BF, Scheithauer BW, et al. Nonvasculitic autoimmune inflammatory meningoencephalitis (NAIM): a reversible form of encephalopathy. Neurology. 1999;53:1579Y1581. 2. Chong JY, Rowland LP. What’s in a NAIM? Hashimoto encephalopathy, steroid-responsive encephalopathy associated with autoimmune thyroiditis, or nonvasculitic autoimmune meningoencephalitis? Arch Neurol. 2006;63:175Y176.
166
www.jclinrheum.com
&
Volume 20, Number 3, April 2014
3. Caselli RJ, Scheithauer BW, O’Duffy JD, et al. Chronic inflammatory meningoencephalitis should not be mistaken for Alzheimer’s disease. Mayo Clin Proc. 1993;68:846Y853. 4. Kessel A, Rosner I, Toubi E. Rituximab: beyond simple B cell depletion. Clin Rev Allergy Immunol. 2008;34:74Y79. 5. Mauri C, Bosma A. Immune regulatory function of B cells. Annu Rev Immunol. 2012;30:221Y241. 6. Ray A, Mann MK, Basu S, et al. A case for regulatory B cells in controlling the severity of autoimmune-mediated inflammation in experimental autoimmune encephalomyelitis and multiple sclerosis. J Neuroimmunol. 2011;230:1Y9. 7. Caselli RJ, Scheithauer BW, Bowles CA, et al. The treatable dementia of Sjogren’s syndrome. Ann Neurol. 1991;30:98Y101. 8. Hoffman Snyder C, Mishark KJ, Caviness JN, et al. Nonvasculitic autoimmune inflammatory meningoencephalitis imitating Creutzfeldt-Jakob disease. Arch Neurol. 2006;63:766Y768. 9. Lyons MK, Caselli RJ, Parisi JE. Nonvasculitic autoimmune inflammatory meningoencephalitis as a cause of potentially reversible dementia: report of 4 cases. J Neurosurg. 2008;108:1024Y1027. 10. Stubgen JP. Central nervous system inflammatory demyelination after rituximab therapy for idiopathic thrombocytopenic purpura. J Neurol Sci. 2010;288:178Y181. 11. Cohen SB. Targeting the B cell in rheumatoid arthritis. Best Pract Res Clin Rheumatol. 2010;24:553Y563. 12. Chan AC. B cell immunotherapy in autoimmunityV2010 update. Mol Immunol. 2011;48:1344Y1347. 13. Liossis SN, Sfikakis PP. Rituximab-induced B cell depletion in autoimmune diseases: potential effects on T cells. Clin Immunol. 2008;127:280Y285. 14. Nagafuchi S. The role of B cells in regulating the magnitude of immune response. Microbiol Immunol. 2010;54:487Y490. 15. Van de Veerdonk FL, Lauwerys B, Marijnissen RJ, et al. The anti-CD20 antibody rituximab reduces the Th17 cell response. Arthritis Rheum. 2011;63:1507Y1516. 16. Matsushita T, Horikawa M, Iwata Y, et al. Regulatory B cells [B10 cells] and regulatory T cells have independent roles in controlling experimental autoimmune encephalomyelitis initiation and late-phase immunopathogenesis. J Immunol. 2010;185:2240Y2252. 17. Ray A, Basu S, Williams C, et al. A novel IL-10Yindependent regulatory role for B cells in suppressing autoimmunity by maintenance of regulatory T cells via GITRL. J Immunol. 2012;188:3188Y3198. 18. Matsushita T, Yanaba K, Bouaziz JD, et al. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease progression. J Clin Invest. 2008;118:3420Y3430. 19. Leandro MJ, Cambridge G, Ehrenstein MR, et al. Reconstitution of peripheral blood B cells after depletion with rituximab in patients with rheumatoid arthritis. Arthritis Rheum. 2006;54:613Y620. 20. Petereit HF, Rubbert-Roth A. Rituximab levels in cerebrospinal fluid of patients with neurological autoimmune disorders. Mult Scler. 2009;15:189Y192. 21. Clifford DB, Ances B, Costello C, et al. Rituximab-associated progressive multifocal leukoencephalopathy in rheumatoid arthritis. Arch Neurol. 2011;68:1156Y1164.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Nonvasculitic Autoimmune Meningoencephalitis After Rituximab The Potential Downside of Depleting Regulatory B Cells in the Brain Indira Hadley, MD,*Þ Richa Jain, MD,þ and Antoine Sreih, MD§
Abstract: Nonvasculitic autoimmune meningoencephalitis (NAIM) is a rare condition describing a syndrome of steroid-responsive encephalopathy in patients with similar clinical and pathologic features. It can be associated with autoimmune diseases, such as Sjogren’s syndrome and autoimmune thyroiditis. Brain biopsies usually show inflammatory cells without evidence of vasculitis. In this article, we present a patient who developed NAIM after receiving rituximab, a B-cellYdepleting therapy for rheumatoid arthritis. The brain biopsy showed a lack of B lymphocytes in the brain tissue, and the patient responded well to intravenous immunoglobulins. We further discuss the role of B lymphocytes and specific regulatory B lymphocytes in suppressing autoimmunity in the brain and propose that the depletion of regulatory B cells may contribute to the pathogenesis of NAIM. This case illustrates a potential side effect of rituximab and demonstrates the importance of regulatory B cells in maintaining the immune response. Key Words: nonvasculitic autoimmune meningoencephalitis (NAIM), rituximab, regulatory B cells (J Clin Rheumatol 2014;20: 163Y166)
N
onvasculitic autoimmune meningoencephalitis (NAIM) is an uncommon but potentially fatal autoimmune disease. It was first described by Caselli et al1 in 5 patients with corticosteroidresponsive encephalopathy. Nonvasculitic autoimmune meningoencephalitis has been associated with thyroid autoimmunity, referred to as Hashimoto encephalopathy, and with Sjogren’s syndrome and possibly systemic lupus erythematosus.2,3 B lymphocytes are considered a vital part of the immune system. They play a major role in the adaptive immune response by producing antibodies and developing into memory cells. They are also antigen-presenting cells and express major histocompatibility complex class II and costimulatory molecules, CD80 and CD86, and produce inflammatory cytokines.4 More recently, a new suppressive function of B lymphocytes has been identified. This functional subset of B lymphocytes, called regulatory B cells, has the role of maintaining the fine equilibrium required for tolerance. They prevent the excessive inflammatory responses that occur in autoimmune diseases or prolonged infections.5 The depletion of regulatory B cells has been shown to worsen experimental autoimmune encephalomyelitis (EAE) in a From the *John H. Stroger Jr. Hospital of Cook County, Chicago, IL; Departments of †Medicine, and ‡Pathology, Rush University School of Medicine, Chicago, IL; and §Department of Medicine, The University of Pennsylvania Penn Tower, Philadelphia, PA. Supported by the Barbara and Jacob Rukel fund. The authors declare no conflict of interest. Correspondence: Indira Hadley, MD, John H. Stroger Jr. Hospital of Cook County, 1700 West Polk St, Suite 748, Chicago, IL 60612. E-mail: [email protected]. Copyright * 2014 by Lippincott Williams & Wilkins ISSN: 1076-1608/14/2003Y0163 DOI: 10.1097/RHU.0000000000000099
JCR: Journal of Clinical Rheumatology
&
mouse model of multiple sclerosis.6 Here, we present a human case, the first to our knowledge, in whom the depletion of B cells in the brain may have led to NAIM, a histopathology similar to murine EAE. This is a patient with rheumatoid arthritis (RA) who developed NAIM after rituximab treatment. We also discuss the potential protective role of regulatory B cells in the central nervous system (CNS) and the downside of their depletion.
CASE REPORT A 67-year-old white woman with a long-standing history of RA, diagnosed at age 20 years, who was receiving rituximab twice per year came to Chicago from Phoenix. She had cataract surgery 3 weeks before traveling and experienced self-limited diarrhea postoperatively. Three days after her arrival in Chicago, she began experiencing nausea, vomiting, abdominal pain, and diarrhea. On presentation, she had a fever of 102.6-F, respiratory rate of 18, pulse of 104, and blood pressure of 143/79 but was alert and oriented. On examination, she was noted to have ulnar deviation with metacarpophalangeal joint subluxation of both hands. There were no signs of active joint synovitis. Her initial laboratory tests showed a white blood cell count (WBC) of 8.0 (4.8Y10.8), hemoglobin of 13.1 (12.0Y16.0), and platelet count of 314 (150Y450) with a normal differential. Her creatinine was 1.2 (0.6Y1.4), and results of liver function tests were normal. The lipase level was 44 (8Y57). Two days into her hospital stay, she was noted to become increasingly confused and agitated. She developed jerking movements of her upper extremity and eventually became more incoherent. She was placed on broad-spectrum antibiotics with intravenous (IV) and oral vancomycin, linezolid, metronidazole, and antifungal antibiotics. She had an extensive infectious workup, including negative blood cultures, Herpes simplex virus 1 and 2, West Nile virus, QuantiFERON Gold, stool cultures, stool ova and parasite, and Clostridium difficile toxin. Her initial urinalysis showed protein greater than 300 and red blood cell count of 3 to 5, and her urine culture grew vancomycin-resistant enterococcus, which was later thought to be a contaminant. A repeat urinalysis done several weeks later showed normal findings. Other laboratory work included an elevated erythrocyte sedimentation rate of 35; a C-reactive protein of 5; and an elevated creatine kinase of 349, which rose to 2000 and then decreased to 508 during a course of 10 days. The thyroid-stimulating hormone level was normal, but the patient had mildly positive antithyroglobulin and antithyroid peroxidase antibodies. An autoimmune panel was negative for antinuclear, antiYSjogren’s syndrome A, antiYSjogren’s syndrome B, and antineutrophil cytoplasmic antibodies; showed an elevated rheumatoid factor of 84; and showed normal complement levels. A computed tomography (CT) of the chest showed fibrotic changes in the lung bases. A CT of the abdomen and the pelvis showed distal esophageal thickening, extensive thumbprinting, and mucosal edema throughout the distal sigmoid and the rectum, which were suggestive of acute infectious
Volume 20, Number 3, April 2014
www.jclinrheum.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
163
JCR: Journal of Clinical Rheumatology
Hadley et al
colitis or ischemic colitis. An esophageal gastroduodenoscopy and colonoscopy showed mild reflux esophagitis in the distal esophagus and a few erosions in the stomach. There was submucosal hemorrhage with edema and hyperemia in the sigmoid region and diffuse diverticulosis. An antral biopsy showed mild gastritis, and a sigmoid colon biopsy showed colonic mucosa with limited pathologic change and diffuse hyperemia. No evidence of vasculitis was seen. At the onset of the altered mental status, a magnetic resonance imaging (MRI) of the brain showed extensive patchy and fluffy areas of abnormal white matter signal within the deep white matter of the frontoparietal region, suggesting significant white matter disease (Fig. 1). A previous MRI done in 2002 was unremarkable. A lumbar puncture showed 33 red blood cells, 28 WBCs with 82% lymphocytes, glucose of 60 mg/dL, and protein of 40 mg/dL. Results of acid-fast bacilli, cryptococcal antigen, and John Cunningham (JC) polyoma virus polymerase chain reaction were all negative. Cerebrospinal fluid (CSF) was also negative for malignancy but showed small reactive lymphocytes. The immunoglobulin G to albumin ratio was within normal limits, and no oligoclonal bands were seen. Only 1% of CD19 cells were detected on flow cytometry in the CSF. The patient had 2 repeat lumbar punctures, which still showed an elevated WBC with elevated lymphocytes and mildly elevated protein. Additional infectious workup was negative for JC virus antibody in the CSF; West Nile virus, enterovirus RNA, and arbovirus antibodies; and Tropheryma whippeli, coccidioidies, and bartonella antibodies. Blood flow cytometry showed a high percentage of T cells and absent B cells. The T cells showed normal cell surface antigen expression, and there was no
FIGURE 1. Precontrast and postcontrast MRI of the brain showing multiple scattered nonenhancing T2/fluid attenuated inversion recovery hyperintense areas noted in the subcortical and periventricular white matter. Technique included sagittal and axial precontrast and postcontrast T1-weighted acquisitions, axial T2-weighted fluid attenuated inversion recovery susceptibility weighted acquisition, diffusion-weighted acquisition, and apparent diffusion coefficient map. Postcontrast examination was done after IV infusion of 13 mm of multihance.
164
www.jclinrheum.com
&
Volume 20, Number 3, April 2014
FIGURE 2. Hematoxylin and eosin, 400. There is white matter parenchymal and perivascular inflammation. The brain biopsy was fixed in 10% buffered formalin followed by paraffin embedding and hematoxylin and eosin staining.
evidence of lymphoma. A CT angiogram of the brain showed a normal circle of Willis, and the cerebral angiogram showed no evidence of vasculitis. The electroencephalogram (EEG) showed continuous, diffuse, moderate-to-severe slowing but no epileptiform activity. Eventually, the patient was treated with pulse IV methylprednisolone for 3 days because of the lack of clinical response to antibiotics and antifungals. A repeat MRI of the brain several days after the course of pulse corticosteroids revealed a new lesion in the left frontal pole resembling a demyelinating lesion. A stereotactic brain biopsy was performed and showed white matter tissue with perivascular and parenchymal inflammation comprising predominantly lymphocytes. The blood vessel walls were intact without endothelial cell swelling or inflammatory cells infiltrating the vessel wall, indicating the absence of vasculitis (Fig. 2). There was no evidence of demyelination as determined by a special stain for myelin (Luxol fast blue). Strong CD3 immunoreactivity was noted in the infiltrate characterizing the lymphocytes as T cells (Figs. 3, 4). As expected, there was complete absence of CD20-reactive B lymphocytes (Fig. 5).
FIGURE 3. 200. CD3 immunostain highlights T cells within the parenchymal infiltrate. All immunohistochemical stains were performed on Leica Autostainer 48. The monoclonal antibody used against CD3 antigen was a prediluted clone LM10 from Leica. * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
JCR: Journal of Clinical Rheumatology
&
Volume 20, Number 3, April 2014
FIGURE 4. 400. The perivascular lymphocytes are also CD3-reactive T lymphocytes. The CD3 immunostain used here is the same as that in Figure 3.
In addition, there were foci of microgliosis, which were highlighted by CD68 immunostain. However, there were no microglial nodules, infarctions, hemorrhages, or intranuclear or cytoplasmic inclusions (viral cytopathic effects). On the basis of the brain biopsy results, a diagnosis of NAIM was made. Because of the brain biopsy results and persistent decline in cognitive function despite IV corticosteroids, she was treated with IV immunoglobulins (IVIG), with a total of 2 g/kg given for 2 days. By the second day of IVIG treatment, the patient was more responsive and following some verbal commands. By the third day after treatment, she was communicating verbally but with signs of expressive aphasia. By the fourth day, the patient was communicating with family members, recognized familiar faces, and was reading words. She eventually went back to Phoenix, where she continued to improve. Repeat blood work after treatment revealed both normal urinalysis results and creatine kinase of 53. A repeat CT of the chest, the abdomen, and the pelvis showed resolution of small and large bowel wall thickening. A brain MRI without contrast done 5 months after her initial presentation showed bihemispheric white matter disease thought to reflect sequelae of prior small vessel ischemic disease. There was no abnormal parenchymal or leptomeningeal enhancement.
DISCUSSION Nonvasculitic autoimmune meningoencephalitis is an uncommon but potentially fatal syndrome. The symptoms of NAIM were first described as progressive dementia in a 56-year-old woman with primary Sjogren’s syndrome whose brain biopsy showed perivascular lymphocytic inflammation in the leptomeningeal and parenchymal vessels.7 Nonvasculitic autoimmune meningoencephalitis was later found to be associated with thyroid autoimmunity, referred to as Hashimoto encephalopathy2; Sjogren’s syndrome; and possibly, systemic lupus erythematosus.3 The term NAIM was first proposed in 1999 by Caselli et al1 to describe a syndrome of corticosteroid-responsive encephalopathy in 5 patients with similar clinical and pathologic features. Four cases had an associated autoimmune disease, and all had progressive cognitive decline. Brain biopsy showed perivascular lymphocytic infiltration without vessel wall involvement. Typical EEG in patients with NAIM show moderate-to-severe nonspecific dysrhythmic slowing, and spinal fluid may show an elevated immunoglobulin G index and lymphocytic pleocytosis.8 Our patient had an underlying * 2014 Lippincott Williams & Wilkins
NAIM After Rituximab
autoimmune disease, mildly positive antithyroglobulin and antithyroid peroxidase antibodies, diffuse moderate-to-severe slowing on EEG, lymphocytic pleocytosis in her CSF, and a brain biopsy revealing parenchymal and perivascular inflammatory infiltrate without evidence of demyelination or vasculitis, all of which are suggestive of NAIM. Of interest, NAIM was previously described in a 56-year-old woman with an elevated rheumatoid factor who was initially given the diagnosis of Alzheimer dementia9; however, it has never been reported in a patient with RA. Our patient has received 5 cycles of rituximab, with the last infusion being 5 months before her clinical deterioration. Rituximab has been associated with CNS demyelination10 but has not been previously reported to cause NAIM. Finally, the presence of mild elevation in antithyroglobulin and thyroid peroxidase antibodies may have predisposed her to developing NAIM. Targeting B-cell activity has shown to be effective in patients with refractory RA. B cells have been known to play a key role in the pathogenesis of RA since the discovery of antibodies to rheumatoid factor and citrullinated proteins.11 The role of B cells in autoimmune diseases is varied and includes direct cytotoxic effects of autoantibodies, immune complexYmediated destruction of organs, and B-cell mediated costimulation of T and dendritic cells through upregulation of cell surface molecules.12 Rituximab targets CD20, a specific surface antigen expressed on pre-B and mature B cells, but CD20 is also expressed in low levels on T and NK cells. As a result, the numbers of T and NK cells have been shown to be reduced during rituximab treatment in RA patients.13 B cells play a role in regulating the T-cell response and in regulating the magnitude of the immune response through antigen presentation to CD4-positive autoreactive T cells.14 More recently, it has been shown to reduce the T helper 17 response.15 Regulatory B cells, a functional subset of B cells, play an important role in regulating immune responses. The role of regulatory B cells has been studied most in the mouse model of human multiple sclerosis called EAE. The disease is exacerbated in mice deficient in B cells because of genetic ablation. B cellY deficient mice were susceptible to EAE induction and exhibited a chronic disease course.6 Furthermore, preferential depletion of regulatory B cells in vivo during disease initiation has been found to increase EAE pathogenesis. Regulatory B cells directly influence production of proinflammatory cytokines by CD4+ T cells and suppress antigen-presenting function of dendritic cells.16 B cells also play a role in preventing autoimmunity by controlling the number of T regulatory cells. In the same mouse model, antiCD20 depletion of B cells caused a reduction in the absolute number of T regulatory cells and resulted in severe colitis, a T-cellYmediated autoimmune condition. Our patient was also
FIGURE 5. 200. There is absence of CD20-positive B lymphocytes in the parenchymal and perivascular inflammatory infiltrate. The monoclonal antibody used against CD20 antigen was a prediluted clone MJ1 from Leica. www.jclinrheum.com
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
165
JCR: Journal of Clinical Rheumatology
Hadley et al
found to have severe colitis on CT of the abdomen and the pelvis during her initial presentation. In addition, regulatory B cell subsets, in particular the interleukin 10Yproducing B cells (B10), have been shown to be protective of EAE. Removal of B10 cells subset before EAE induction causes exacerbation of disease symptoms by influencing T cells within the CNS.17,18 We hypothesize that the depletion of B cells, in particular regulatory B cells and possibly B10 cells, in our patient resulted in loss of the immune regulatory function and an attack against undetermined CNS autoantigens causing NAIM. Our patient received her last rituximab infusion 5 months before the onset of symptoms, which led to complete deletion of her B lymphocytes as evidenced by the absence of B lymphocytes on blood and CSF flow cytometry and brain biopsy immunostaining. The effects of rituximab on serum B-cell depletion lasts for roughly 3 to 6 months, and B cells may not begin to repopulate until at least 8 months after treatment.19 Furthermore, rituximab has been found to penetrate an intact blood-brain barrier up to 6 months after the first infusion20 and has been linked to progressive multifocal leucoencephalopathy (PML), a rare and often fatal infection of the CNS caused by the JC polyoma virus. In 1 case series of rituximab-associated PML in RA patients, the majority developed PML 5 to 7 months after several cycles of rituximab infusion were given.21 Our patient responded well to IV immunoglobulins. We suspect that IVIG replaced the immune-mediated effects of regulatory B cells by inhibiting cytokines and complement activation, interfering with Fc receptor binding, and blocking immunoregulatory and adhesion molecules. The 1 previous case report of a patient with CNS demyelination after rituximab therapy also responded to IVIG, as did our patient.10 It is unclear as to why NAIM is not more frequently seen with rituximab therapy. We suspect that genetic predisposition or the presence of specific autoimmune diseases such as Hashimoto thyroiditis or Sjogren’s disease may increase the risk for developing NAIM. To our knowledge, this is the first human case of autoimmune meningoencephalitis that may have resulted from the depletion of B cells and specific regulatory B cells. Although much is still unknown about the exact role of regulatory B cells and the association between the depletion of regulatory B cells and NAIM, we hypothesize that their depletion with prior rituximab therapy played a role in the pathogenesis of NAIM in our patient. Specific depletion of cytotoxic B cells while preserving the regulatory B cells may potentially prove to be a safer approach for the treatment of autoimmune diseases. REFERENCES 1. Caselli RJ, Boeve BF, Scheithauer BW, et al. Nonvasculitic autoimmune inflammatory meningoencephalitis (NAIM): a reversible form of encephalopathy. Neurology. 1999;53:1579Y1581. 2. Chong JY, Rowland LP. What’s in a NAIM? Hashimoto encephalopathy, steroid-responsive encephalopathy associated with autoimmune thyroiditis, or nonvasculitic autoimmune meningoencephalitis? Arch Neurol. 2006;63:175Y176.
166
www.jclinrheum.com
&
Volume 20, Number 3, April 2014
3. Caselli RJ, Scheithauer BW, O’Duffy JD, et al. Chronic inflammatory meningoencephalitis should not be mistaken for Alzheimer’s disease. Mayo Clin Proc. 1993;68:846Y853. 4. Kessel A, Rosner I, Toubi E. Rituximab: beyond simple B cell depletion. Clin Rev Allergy Immunol. 2008;34:74Y79. 5. Mauri C, Bosma A. Immune regulatory function of B cells. Annu Rev Immunol. 2012;30:221Y241. 6. Ray A, Mann MK, Basu S, et al. A case for regulatory B cells in controlling the severity of autoimmune-mediated inflammation in experimental autoimmune encephalomyelitis and multiple sclerosis. J Neuroimmunol. 2011;230:1Y9. 7. Caselli RJ, Scheithauer BW, Bowles CA, et al. The treatable dementia of Sjogren’s syndrome. Ann Neurol. 1991;30:98Y101. 8. Hoffman Snyder C, Mishark KJ, Caviness JN, et al. Nonvasculitic autoimmune inflammatory meningoencephalitis imitating Creutzfeldt-Jakob disease. Arch Neurol. 2006;63:766Y768. 9. Lyons MK, Caselli RJ, Parisi JE. Nonvasculitic autoimmune inflammatory meningoencephalitis as a cause of potentially reversible dementia: report of 4 cases. J Neurosurg. 2008;108:1024Y1027. 10. Stubgen JP. Central nervous system inflammatory demyelination after rituximab therapy for idiopathic thrombocytopenic purpura. J Neurol Sci. 2010;288:178Y181. 11. Cohen SB. Targeting the B cell in rheumatoid arthritis. Best Pract Res Clin Rheumatol. 2010;24:553Y563. 12. Chan AC. B cell immunotherapy in autoimmunityV2010 update. Mol Immunol. 2011;48:1344Y1347. 13. Liossis SN, Sfikakis PP. Rituximab-induced B cell depletion in autoimmune diseases: potential effects on T cells. Clin Immunol. 2008;127:280Y285. 14. Nagafuchi S. The role of B cells in regulating the magnitude of immune response. Microbiol Immunol. 2010;54:487Y490. 15. Van de Veerdonk FL, Lauwerys B, Marijnissen RJ, et al. The anti-CD20 antibody rituximab reduces the Th17 cell response. Arthritis Rheum. 2011;63:1507Y1516. 16. Matsushita T, Horikawa M, Iwata Y, et al. Regulatory B cells [B10 cells] and regulatory T cells have independent roles in controlling experimental autoimmune encephalomyelitis initiation and late-phase immunopathogenesis. J Immunol. 2010;185:2240Y2252. 17. Ray A, Basu S, Williams C, et al. A novel IL-10Yindependent regulatory role for B cells in suppressing autoimmunity by maintenance of regulatory T cells via GITRL. J Immunol. 2012;188:3188Y3198. 18. Matsushita T, Yanaba K, Bouaziz JD, et al. Regulatory B cells inhibit EAE initiation in mice while other B cells promote disease progression. J Clin Invest. 2008;118:3420Y3430. 19. Leandro MJ, Cambridge G, Ehrenstein MR, et al. Reconstitution of peripheral blood B cells after depletion with rituximab in patients with rheumatoid arthritis. Arthritis Rheum. 2006;54:613Y620. 20. Petereit HF, Rubbert-Roth A. Rituximab levels in cerebrospinal fluid of patients with neurological autoimmune disorders. Mult Scler. 2009;15:189Y192. 21. Clifford DB, Ances B, Costello C, et al. Rituximab-associated progressive multifocal leukoencephalopathy in rheumatoid arthritis. Arch Neurol. 2011;68:1156Y1164.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
![[Regulatory B cells in human autoimmune diseases].](/assets/img/hold.png)
