Anti-neurofascin 155 Antibody-positive Neuropathy in a Human Immunodeficiency Virus-infected Patient

Abstract

Human immunodeficiency virus (HIV)-associated neuropathy is a common complication of HIV infection and has several clinical subtypes. HIV-associated chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) is a demyelinating neuropathy whose clinical features are known to differ from those of CIDP in the HIV-uninfected population. We herein report a case of CIDP in an HIV-infected patient who was finally diagnosed with anti-neurofascin 155 (NF155) antibody-positive neuropathy. The clinical features, including clinical findings and therapeutic responses, were typical of paranodal antibody-mediated neuropathy. To our knowledge, this is the first case of anti-NF155 antibody-associated neuropathy in an HIV-infected patient.

Introduction

Human immunodeficiency virus (HIV)-associated neuropathy is a common complication of HIV infection. While HIV itself causes macrophage dysregulation by proinflammatory cytokine release or direct neurotoxicity by its envelope protein, it also evokes autoimmune processes, such as chronic inflammatory demyelinating polyradiculoneuropathy (CIDP) (1).

HIV-associated CIDP is a chronic demyelinating neuropathy occurring in patients suffering from HIV and has distinct clinical features compared with the HIV-uninfected population, including a younger age of onset, a monophasic progressive course, slightly higher cerebrospinal fluid (CSF) protein levels, and a significantly better response to corticosteroid treatment (2). Although these distinct clinical features and therapeutic responses to immune modification therapy suggest that some specific autoimmune conditions might exist in this disease, no definite evidence supports this implication (2).

CIDP is a chronic inflammatory neuropathy with diverse subtypes. Recent progress in neuroimmunology has revealed a growing number of autoantibodies targeting the peripheral nervous system. There is increasing evidence that neuropathy associated with antibodies targeting the node and paranode of myelinated peripheral nerves, autoimmune ‘nodo-paranodopathies’, should be recognized as a pathologically distinct disease entity. In particular, anti-neurofascin 155 (NF155) antibody, a representative nodal and paranodal antibody, is associated with a specific phenotype versus seronegative CIDP. Characteristics of anti-NF155 antibody-positive neuropathy include predominantly distal weakness, marked elevation of CSF protein levels, and nerve root enlargement and enhancement on magnetic resonance imaging (MRI). Patients with this antibody tend to respond well to corticosteroids or rituximab, whereas immunoglobulins provide only transient or partial relief (3).

We herein report a case of neuropathy in an HIV-infected patient who was ultimately diagnosed with anti-NF155 antibody-positive neuropathy. To our knowledge, this is the first case of anti-NF155 antibody-positive neuropathy in an HIV-infected patient. HIV evokes a wide variety of autoantibody-mediated diseases, such as Graves' disease, antiphospholipid syndrome, and immune thrombocytopenic purpura (4,5), and anti-NF155 antibody in our case might be one such HIV-associated autoantibody. This case has clinical features of both anti-NF155 antibody-positive neuropathy and HIV-associated CIDP and suggests an autoantibody-mediated feature of HIV-associated CIDP.

Case Report

A 50-year-old man with HIV infection visited our institution with a 7-month history of dysesthesia and weakness of the extremities. At 36 years old, the patient was identified as infected with HIV. While combination antiretroviral therapy (cART) that was initiated at 40 years old had maintained HIV-RNA at an undetectable level, 6 months prior to the visit, HIV-RNA became detectable (37 copies/mL) for the first time in 10 years. Paresthesia in the soles of both feet gradually expanded proximally over 7 months accompanied by weakness. The patient's antiretroviral therapy regimen did not include any agents known to cause neuropathy. A neurological examination revealed mild symmetric weakness in the toe extensors, with diminished tendon reflexes in all four extremities. Symmetric dysesthesia and hypoesthesia in the distal extremities were noted. Deep sensation, including vibratory sensation and proprioception, was preserved. A cranial nerve examination was normal. Ataxia and tremor were absent. These findings suggested progressive polyneuropathy.

An examination of the CSF revealed prominent elevation of protein (164 mg/dL) but was otherwise normal (1 white cell/μL, IgG index 0.69). MRI revealed enlarged nerve roots in both the brachial and lumbosacral plexuses, suggesting nerve root involvement (Figure B). Nerve conduction studies (NCSs) showed prolonged distal latencies and slow conduction velocities (Figure C). The F-waves were delayed and reduced in occurrence. Sensory nerve action potential amplitudes were decreased (Table 1). These findings met the electrodiagnostic criteria for CIDP (6), and a preliminary diagnosis of HIV-associated CIDP was made.

Figure.

Clinical findings in the patient. (A) The disease course of the patient. Following the administration of corticosteroids, sensory disturbance improved prominently. (B) Magnetic resonance imaging revealed nerve root hypertrophy with gadolinium enhancement in both the brachial plexus (a, b) and lumbosacral plexus (c, d) [a and c, magnetic resonance neurography (3D-NerveVIEW); b and d, gadolinium-enhanced T1-weighted imaging]. Dashed lines on neurography indicate the level of the axial image shown. Arrowheads indicate nerve root hypertrophy. (C) A nerve conduction study showed prominent demyelination, which improved after corticosteroid therapy. Distal latencies, conduction velocities and combined motor action potential (CMAP) amplitudes were improved in motor conduction studies. Sensory nerve action potential (SNAP) of the ulnar nerve was initially undetectable. Three months after initiating corticosteroids, SNAP was detected. IVIg: intravenous immunoglobulin, SCIg: subcutaneous immunoglobulin, IVMP: intravenous methylprednisolone, PSL: prednisolone, MCS: motor conduction studies, SCS: sensory conduction studies, DL: distal latency, dCMAP: distal combined motor action potential amplitude, MCV: motor conduction velocity, SCV: sensory conduction velocity, SNAP: sensory nerve action potential

Table 1.

Nerve Conduction Studies throughout the Clinical Course. All Studies were Conducted on the Left Side.

	Before treatment	1 month after IVIg	2 months of SCIg	5 months of SCIg	7 months of SCIg	After IVMP	3 months of oral PSL
Motor
Median nerve
DL (ms)	7.3	7.9	7.7	7.5	8.5	8.5	4.9
CV (m/s)	33.5	33.9	32.9	27.2	29.3	30.1	38.5
dCMAP (mV)	5.4	7.1	6.4	5.6	8.8	6.8	8.9
Ulnar nerve
DL (ms)	6.1	5.8	5.8	6.0	7.4	7	3.7
CV (m/s)	41.3	40	40.2	39.1	32.8	37.6	41.0
dCMAP (mV)	10.6	10.5	11.2	9.7	11.2	9.6	14.4
Tibial nerve
DL (ms)	16.8	14.2	12.5	11.7	14.2	16.0	8.7
CV (m/s)	27.8	29.5	30.6	22.0	17.4	16.2	28.6
dCMAP (mV)	1.9	3.3	3.4	1.4	1.6	1.5	4.0
Peroneal nerve
DL (ms)	13.6	15.2	14.7	17.1	19.2	18	10.6
CV (m/s)	19.6	23.6	25.5	20.8	14.9	14.7	27.1
dCMAP (mV)	0.26	1.1	0.96	0.11	0.18	0.87	0.9
Sensory
Median nerve
DL (ms)	5.6	5.9	6.0	7.1	N.D.	N.D.	3.4
CV (m/s)	23.2	23.5	24.8	20.3	N.D.	N.D.	36.9
SNAP (µV)	1.7	1.0	0.8	1.2	N.D.	N.D.	6.7
Ulnar nerve
DL (ms)	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.	3.2
CV (m/s)	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.	30.9
SNAP (µV)	N.D.	N.D.	N.D.	N.D.	N.D.	N.D.	3.3
Sural nerve
DL (ms)	2.8	3.1	3.3	3.4	5.2	5.1	4.0
CV (m/s)	52.8	47.7	45.2	44.6	38.8	39.5	37.9
SNAP (µV)	2.8	2.8	3.6	3.6	2.3	4.9	3.9
F-wave
Median nerve (left)
Occurrence (%)	0	87	87	100	81	100	75
Latency (ms)	N.D.	65.2	65.0	67.6	70.9	69.1	39.9
Tibial nerve (right)
Occurrence (%)	100	100	100	100	100	94	25
Latency (ms)	56.8	112	126	139	166	174	92.6

DL: distal latency, CV: conduction velocity, dCMAP: distal compound motor action potential, SNAP: sensory nerve action potential, IVIg: intravenous immunoglobulin, SCIg: subcutaneous immunoglobulin, IVMP: intravenous methylprednisolone, PSL: prednisolone, N.D.: not detected

High-dose intravenous immunoglobulin (IVIg) (2 g/kg) and subsequent weekly subcutaneous immunoglobulin (SCIg, 0.2 g/kg) administration mildly relieved the patient's symptoms; however, sensory disturbance continued to progress. During gradual neurological deterioration under SCIg, the patient was found to be positive for anti-NF155 IgG4 antibodies in a flow cytometric assay (7). Considering the more favorable response to corticosteroids in anti-NF155 antibody-positive neuropathy (7), intravenous corticosteroid administration followed by initiation of 0.4 mg/kg/day oral prednisolone (equivalent to 0.5 mg/kg/day in consideration of drug interaction with anti-retroviral agents) was administered.

The sensory disturbance resolved almost completely in three months, and amplitudes and conduction velocities improved in both the motor and sensory NCSs (Figure A). Notably, sensory nerve action potentials became detectable in the ulnar nerve for the first time, providing electrophysiological evidence of the therapeutic response. Although anti-NF155 antibody titers are reported to be a marker of disease activity (8), unfortunately, we did not have the opportunity to test posttreatment titers.

Discussion

In our case, the patient initially presented with HIV-associated neuropathy, but the clinical features, including distal dominant weakness, enlarged nerve roots on MRI, and therapeutic response to corticosteroids, were consistent with anti-NF155 antibody-positive CIDP (3). To our knowledge, this is the first case of anti-NF155 antibody-positive neuropathy in an HIV-infected patient.

HIV-associated neuropathy has several subtypes with a complex pathophysiology (1). The most common type of HIV-associated neuropathy is distal sensory neuropathy due to HIV-associated macrophage dysregulation (9) or direct neurotoxicity of the HIV envelope protein gap120 (10). Another type of HIV-associated neuropathy, inflammatory demyelinating polyradiculoneuropathy, has an autoimmune aspect and definite characteristics in its clinical course and therapeutic response, as presented in our case. Interestingly, HIV evokes a wide variety of autoantibody-mediated diseases, such as Graves' disease, antiphospholipid syndrome, and immune thrombocytopenic purpura (4). Neurological diseases, such as myasthenia gravis and Guillain-Barré syndrome, have also been reported (5), and anti-NF155 antibody in our case might be have been one such HIV-associated autoantibody. Autoimmune conditions are more likely to be present in the early phase of HIV infection or after the initiation of cART, when CD4 cell counts are relatively preserved and immunodeficiency is not severe (11). In our case, HIV infection was relatively well-controlled, which might support the existence of HIV-associated autoimmune conditions.

NF155 is an essential adhesion molecule located in paranodal septate-like junctions of peripheral and central myelinated axons (12). Recent studies have reported that anti-NF155 antibody evokes a specific type of inflammatory neuropathy, including a relatively young onset of age, predominantly distal weakness, frequent tremors and ataxia, marked elevation of CSF protein levels, and nerve root enlargement and enhancement on MRI (3,7). However, CIDP in the HIV-infected population has distinct features compared to CIDP in the HIV-uninfected population, including a relatively young age of onset, a monophasic progressive course, a significantly better response to corticosteroid treatment, and slightly higher CSF protein levels (2). Although ataxia and tremor were absent and CSF protein levels were lower than in typical cases of anti-NF155 antibody-associated neuropathy, the distal dominant phenotype and therapeutic response were findings consistent with those of anti-NF155 antibody-associated neuropathy.

Our case showed findings consistent with anti-NF155 antibody-associated neuropathy and its clinically relevant autoantibody in an HIV-infected patient for the first time. Furthermore, our case has clinical features of both anti-NF155 antibody-positive neuropathy and HIV-associated CIDP (Table 2) (2,7), suggesting a possible link between these two disease entities.

Table 2.

Comparison between Features of HIV-positive CIDP, Anti-NF155 Antibody Positive Neuropathy and This Case.

	HIV-CIDP [2]	Anti-NF155 antibody positive neuropathy [7]	This case
Median age at onset	37 [30-42] Younger than non-HIV CIDP	25 [13-50] Younger than seronegative CIDP	50
Sex	Female predominant	Male predominant	Male
Typical weakness distribution	Symmetric, proximal dominant	Symmetric, frequently distal dominant	Symmetric, distal dominant
Tremor	No difference with non-HIV CIDP	Significantly frequent compared to seronegative CIDP	Absent
Sensory ataxia	Not available	Significantly frequent compared to seronegative CIDP	Absent
CSF protein [mg/dL]	138 [88-220] Slightly higher but no significant difference with non-HIV CIDP	317 [175.9-458.1] Significantly higher than seronegative CIDP	164
Nerve conduction studies	Not available	Pronounced prolongation of distal and F-wave latencies	Pronounced prolongation of distal and F-wave latencies
Thickened nerves	No difference with non-HIV CIDP	Greater spinal root diameter	Spinal root thickening on MRI
Therapeutic response	Good response to corticosteroids	Limited response to IVIg Favorable response to corticosteroids	Limited response to IVIg/SCIg Favorable response to corticosteroids

HIV: human immunodeficiency virus, CIDP: chronic inflammatory demyelinating polyradiculoneuropathy, NF155: neurofascin-155, CSF: cerebrospinal fluid, MRI: magnetic resonance imaging, IVIg: intravenous immunoglobulin, SCIg: subcutaneous immunoglobulin

Conclusion

Neuropathy in an HIV-infected patient may present as anti-NF155 antibody-positive neuropathy. HIV-induced immune dysregulation causes multiple autoantibody-mediated diseases, and anti-NF155 antibody-positive neuropathy should be considered an important clinical presentation of HIV-associated neuropathy. Further studies should be conducted to determine the frequency of positive anti-nodal and paranodal antibodies in HIV-infected patients with CIDP and their therapeutic response to delineate the differences in clinical features of CIDP with or without HIV infection.

The authors state that they have no Conflict of Interest (COI).