hall d 37. update on pathogenesis of nmosd · (2010 biochemistry protein projects) 139 aa 2ndary...
TRANSCRIPT
1
37. Update on pathogenesis of NMOSD
Kazuo Fujihara, M.D.Department of MS TherapeuticsFukushima Medical University
MS & NMO Center, STRINSTohoku University
Japan
Teaching Course 12: Neuromyelitis optica spectrum disordersHall D
(October 25, 2017, Paris, France)
Scientific Advisory BoardsBayer Schering, Biogen Idec, Mitsubishi Tanabe, Novartis, Chugai, Ono, Nihon, Merck Serono, Alexion, Medimmune, Medical Review
Speaker HonorariaBayer Schering Biogen Idec, Eisai, Mitsubishi Tanabe, Novartis, Astellas, Takeda, Asahi Kasei, Daiichi Sankyo, Nihon, Cosmic Corporation
Research SupportBayer Schering, Biogen Idec, Asahi Kasei, Chemo-Sero-Therapeutic Research Institute, Teva, Mitsubishi Tanabe, Teijin, Chugai, Ono, Nihon,Genzyme Japan
Government FundingGrants-in-Aid for Scientific Research from Ministry of Education, Science and Technology and Ministry of Health, Welfare and Labor of Japan
Disclosure (Kazuo Fujihara, M.D.)
2
International Consensus Diagnostic Criteria of NMOSD
NMOSD, the unifying term for the entire clin spectrum
1) AQP4-IgG-seropositive NMOSDOne Core Clinical Characteristic (ON, Acute MY, Brain Synd)
2) Seronegative NMOSDTwo or more Core Clinical CharacteristicsAdditional requirements (One of ON/MY/AP Synd, DIS, MRI)
No better explanation for the clin syndRed-flags: Findings atypical for NMOSD (OCB neg, <3VS, etc)
(Wingerchuk et al. Neurology 2015)
3
4
2. Environmental Factors1) Infections
Para- or Post-infectious -- case reportsCMV, VZV, EBV, HAV, HIV, Dengue, Mycoplasma p, Tbc
Clostridium perfringens ABC transporter (homology to T cell epitope of AQP4)(Varin-Dover. Ann Neurol 2012, Cree, Ann Neurol 2016)
1. Host Factors1) Multiple autoimmune diseases & F>>M (AQP4-IgG+cases)2) Ethnicity and AQP4-IgG seroprevalence
7.9/105 in Martinique vs. 3.3/105 in Olmsted (Flanagan. Ann Neurol 2016)
Asian 1.57/105 , Others 0.57//105 in Australia-NZNo North-South gradient (Bukhari, JNNP 2017)
3) Familial NMO: 1% (Japanese survey)4) HLA-DRB1*1501 -- no association with NMOSD5) Copy number variations at specific TCR loci (Sato. Ann Neurol 2015)
Host and Environmental Factors in NMOSD
5
1. NMO-IgG = AQP4-IgG (Lennon et al, JEM 2005)
2. AQP4 is a water channel richly expressed in CNS
3. AQP4-IgG binds to the extracellular domains of AQP4.
AQP4 (pink)
Astrocyte
AQP4-transfected CellsNMO-Patient’s IgG
4. AQP4-IgG predicts relapse. (Weinshenker, et al Ann Neurol 2006)
(Takahashi et al, Brain 2006)
Aquaporin 4 (AQP4) and AQP4-IgG
esp. in periventricular regions, spinal gray matter, andendfeet of astrocytes.
OUT
IN
NH2COOH
Loop A Loop ELoop C
M1
M23
Factors to influence sensitivity in AQP4-IgG assay
1. Cell-Based Assay > ELISA > Mouse Tissue-Based Assay
2. M23 > M1 (M23 forms orthogonal array of particles)
3. Other factors to lower sensitivity in Cell-Based Assay(1) Pre-fixed cells, (2) GFP tagging at N-terminus
OUT
IN
NH2COOH
Loop A Loop ELoop C
M1
M23
AQP4-M23 AQP4-M1
AQP4: Transmembrane protein
(Verkman et al, Methods Enzymol2012)
Cell-Based Assay
(Takahashi et al, Brain 2007)
AQP4 : Tetramer
Orthogonal Array of Particles (OAP)
6
Pathological Features of NMO
(Lucchinetti et al, Brain 2002)
General Features:•Severe Demyelination
•Edema, Necrosis & Cavity
•Acute axonal swelling, Spheroids
•Myelin-laden Macrophages
Humoral Immunity & Vascular Pathology•Deposition of Immunoglobulins and
Complements in Vasculocentric Pattern
•Thickened and Hyalinized Vessel Walls
IgG C9neo C9neo
IgM KiM1P KiM1P7
8
NMO MSLoss of AQP4 & GFAP in NMO Lesions
MBP
AQP4
GFAP
MBP
AQP4
GFAP
Lost & well demarcated
Preserved
Lostextensively
Relatively preserved
(Misu, TJEM 2006, Brain 2007)
(AQP4 & Ig, C9neo)
Higher magnification
Perivascular Deposit of Ig & C9neo & AQP4 loss
9(Yaguchi et al. MSJ 2012)
Pathological study of subacute encephalopathyin a pregnant woman with AQP4-IgG
Consciousness disturbance, aphasia, Rt hemiparesis, Extensive brain MRI lesions, AQP4-IgG-positiveOpen biopsy from Lt temporal lobe:
1) demyelination, loss of AQP4 and GFAP in WM,2) loss of AQP4 also observed in GM
consistent with the NMO SC pathology
Normal WM
Lesion
AQP4 GFAP
10
Type 1: Active NMO lesions with C activation and Gra infiltration
Type 2: Cystic lesions with extensive tissue destruction
Type 3: Lesions resembling secondary Wallerian degeneration
Type 4: Lesions with selective loss of AQP4
Type 5: Active NMO lesions with astrocytic clasmatodendrosis
Type 6: Lesions with astrocyte dystrophy and primary demyelination
Six Different Lesion Types in NMO
(Misu et al, Acta Neuropathol 2013)
Six Lesion Types in NMO
Diverse mechanisms of tissue injury operate in the same patient
Complement-mediated cytotoxicity, necrotic
No complement deposition, apoptosis-like
11
Pathomechanisms of AQP4-IgG+NMOSD
(modified from Verkman et al, Brain Pathol 2013)
Plasmabast
IL6
C5
CD27+Memory B Cell
MBP
NF, thin RNFL
GFAP, S100b
AQP4-IgG1
T Cell(Th17, Th1,
AQP4-T cell, etc)
protease
BBB Breakdown
(CD19intCD27highCD38highCD180--)Neutrophil Eosinophil
Mac
Mechanisms of NMO pathogenesis and therapeutic targets
(Pittock, Ann NY Aca Sci 2016)
(1) IL-6 inhibitorstocilizumab, SA237
(2) anti-CD20rituximab
(3) anti-CD19MEDI-551
(4) glutamate antagonists(5) AQP4-IgG blocking
aquaporumab(6) C5 inhibition
eculizumabC1 inhibitor
Cinryse(7) anti-neutrophil
sivelestat(8) anti-eosinophil
cetirizine, ketotifen
Blood
BBB
CNS
12
Astrocyte
Severe Astrocytic Damage in AQP4-IgG+NMOSD
GFAP: glial fibrillary acidic protein (an astrocytic protein)
1) Pathogenicity of AQP4 Ab in vitro and in vivo
2) Astrocyte Pathology (AQP4 Loss, etc) in the Lesions
3) Low Myo-Isositol/Creatine Value on 1H-MRS
4) Remarkable High CSF-GFAP in Relapse
13
(Ciccarelli et al, Ann Neurol2013)
(MIsuet al, Brain 2007, etc)
(Hinson et al, Neurology 2007, Bradlet al. Ann Neurol2009, etc)
(Takano et al, Neurology 2011)
GFAP MBP
Remarkably Elevated CSF-GFAP Levels in NMOSD
14
2476
0.8 0.7
(Takano et al, Neurology 2011)
Astrocytic damage is far more severe than demylination in NMOSD
All AQP4-IgG+
1) CNS regions at high AQP4 expression
(Matiello et al, JAMA Neurol 2013)
15
3) Astrocytes are devoid of Complement Regulators (CD46, 55, 59)(Peripheral organs express them)
Lesion distribution in AQP4-IgG+NMOSD
M1
M23
SC ONBSBr
(Pittock et al, Arch Neurol 2006)
2) Higher AQP4 expression in Optic nerve and spinal cord(at mRNA & protein levels)
(Saadoun et al, MSJ 2015)
4) Muscle: an Extra-CNS involvement in NMOSD ?HyperCKemia (Suzuki et al, Neurology 2010)
1) Increase of relapse in 6 months postpartum- An International Collaboration of Korea, Japan, UK and Portugal -
(Kim et al, Neurology 2011)
* P<0.005
Pregnancy PostpartumBefore Pregnancy
16
2) Higher rate of miscarriage (Nour et al, Neurology 2016)
Pregnancy and AQP4-IgG+NMOSD
Experimental Study: AQP4-IgG can cause placental inflammation & fetal death (Saadoun et al, J Immunol 2013)
1) Axonal degenerative pathology of NMO ?
(Kawachi and Lassmann. JNNP 2017)
2) Glucose-regulated protein (GRP) 78 autoantibody associates with BBB disruption in NMOAn auto-ab to bind brain microvascular endothelial cellscauses extravasation of serum Alb, IgG & Fibrinogen into brain
(Shimizu et al. Sci Transl Med 2017)
Recent topics in NMOSD
17
19
(Hemmer B, et al. Nature Rev Neurosci, 2002) (Sinmaz S, et al. J Neuroinflamm 2016)
Myelin oligodendrocyte glycoprotein (MOG)
(2010 Biochemistry Protein Projects)
139 AA
2ndary structure Human MOG-IgG Reactivity1. CC’-loop most commonly recognized2. No epitope spreading over time3. not recognize rodent MOG
(Mayer et al. J Immunol 2013)
21
Severe Demyelination but No Astrocytopathyin MOG-IgG-Positive Definite NMO
31yo, MaleRt ON - Good recoveryAcute Myelitis (2W later) – Good recovery
CSF Cells 90, OCB negT2 lesions [C3-5, C6-T5 (central)]AQP4-IgG-Neg, MOG-IgG-Pos
CSF-MBP 1190pg/ml (normal <102)CSF-GFAP < 0.004ng/ml (Ikeda et al, Mult Scler, 2015)
a n t i - M O G + a n t i - A Q P 4 + M S
1
1 0
1 0 0
1 0 0 0
1 0 0 0 0
1 0 0 0 0 0
1 0 0 0 0 0 0
CSF-GFAP(ng/ml)
p < 0 .0 0 0 1 p < 0 .0 0 0 1
a n t i - M O G + a n t i - A Q P 4 + M S
0
2 5 0
5 0 0
7 5 0
1 0 0 0
1 5 0 0
3 0 0 0
4 5 0 0CSF-MBP(pg/ml)
p = 0 .0 4 1 8
p = 0 .0 1 6 8
〇:Myelitis●:Optic neuritis△:ADEM
(Kaneko K et al. JNNP 2016)
CSF-GFAP and -MBP (anti-MOG vs anti-AQP4+ vs MS)-- an international collaborative study (8 countries) --
I. MS type- II Pathology in biopsied brains of MOG-IgG+ cases
1. Konig et al. Arch Neurol. 2008
Early active inflammatory demyelination with deposition of Ig & C
49-yo white woman“CD-RRMS”, MOG-IgG positive
22
2. Spadalo M, et al. Ann Clin Transl Neurol, 2015
II. Potential pathogenicity of MOG-IgG
1. NMO-MOG-IgG causes reversible lesions in mouse brain(independent of complement) (Saadoun et al, Acta Neuropathol Comm 2015)
2. MOG-IgG affects oligodendrocyte cytoskeleton (Dale et al, Neurol N2. 2014)
3. MOG-IgG+complement induces demyelination (depends on type I IFN) (Berg et al, J Neuroinflammation. 2017)
MRI and OCT in anti-MOG+ vs anti-AQP4+ ON
OCT (Anti-MOG+ vs Anti-AQP4+ ON eyes, 6 month follow-up)• cpRNFL 90.2 ± 10.5 vs 74.1 ± 14.9 µm (P = 0.022)• GCIP 57.0 ± 6.2 vs 46.1 ± 12.3 µm (P = 0.027)
(Akaishi T, J Neurol Neurosurg Psychiatry 2015)
23
(Ogawa et al, Neurol Neuroimmunol Neuroinflamm, 2017)
4 adult male (23-39yo)
Encephalopathy(ON in two)
CSF cell ↑Cereb cort lesions(FLAIR high)
IVMP effectiveNo relapse
MOG-IgG+ unilateral cerebral cortical encephalitis with epilepsy
26c
AQP4-IG+, MOG-IgG+ and Seronegative NMOSDAQP4-IgG+(n = 166)
MOG-IgG+(n = 35)
Seronegative(n = 89)
p value
Onset Age 36 (4 – 78) 35 (3 – 79) 33 (10 – 80) 0.6770
Female Sex 88.6% (147) 54.3% (19) 64.0% (57) <.0001
Single Attack 16.3% (27) 42.9% (15) 38.2% (34) <.0001
Clinical Phenotype
NMO 60.8% (101) 11.4% (4) 21.4% (19)
LETM 29.5% (49) 25.7% (9) 55.1% (49) <.0001
Recur ON / Bil ON 9.7% (16) 62.9% (22) 23.6% (21)
EDSS (last visit)
Monophasic 6 (2 – 8.5) 2 (0 – 6) 4 (0 – 8.5) 0.0009
Recurrent 5 (1 – 8.5) 2 (0 – 8) 4 (0 – 7) <.0001
(Sato et al, AAN, Neurology 2014)27
Summary
28
1. AQP4-IgG+NMOSDis an autoimmune astrocytopathic diseasehas distinct pathogenetic factors from MS
2. MOG-IgG+diseaseis a demyelinating diseasemy be caused by pathogenic MOG-IgGappear to have a wide clinical spectrum
3. Double seronegative NMOSDmay be heterogeneous