In 2021, the European Society of Neurology issued guidelines to propose autoimmune nodopathies, noting that anti-CNTN-1 autoimmune nodopathies seem more likely to occur in elderly individuals. Additionally, autoimmune nodopathies include four kinds of antibody-related autoimmune nodular diseases: contact-1 (CNTN1), neurofascin-155 (NF155), contact-associated protein 1 (Caspr1), and neurofascin isoforms (NF140/186). The incidence of these autoantibody diseases in CIDP is low, ranging from 2% in Europe to 10% in Asian countries[9].
Anti-CNTN-1 autoimmune nodopathies mainly present subacute or acute CIDP-like manifestations, as well as sensory ataxia, tremor, cranial nerve involvement, etc, which are mainly motor symptoms[2, 10]. Taieb et al showed that patients with anti-CNTN-1 autoimmune nodopathies were prone to kidney involvement and nephrotic syndrome[11]. We report that this patient initially presented with dysarthria and tremor of the limbs. Previous reports of anti-CNTN-1 autoimmune nodopathies with dysarthria were rare. Only Li Q et al reported one patient with unclear speech, but the specific performance was unclear[5]. The dysarthria of our patient manifested as voice prosodic disorder (voice intensity is different from that of normal people, and there is slight tremor). The biggest feature is that the voice interruption is obvious, and the patient will stop after a few words of pronunciation. After treatment, the patient's voice improved after telephone follow-up, leaving a slight sense of explosion. Generally, dysarthria is analysed from the perspectives of speech length, phonation, respiration, resonance, rhythm, etc. It can be divided into atony, spasticity, hypokinesia, hyperactivity, ataxia, and mixed dysarthria[12]. Among them, ataxia dysarthria is mainly seen in cerebellar diseases and mainly manifests as syllable explosion, damaged rhythm, and pause after syllable. This is also the case for this patient. Some studies have shown that CNTN1 is not only limited to paranode expression but also found in dorsal root ganglion (DRG) neurons and cerebellar granule neurons (CGN), in which cerebellar binding is suspected to be a factor related to tremor in patients with CNTN-1 autoantibodies[13]. At present, there is no evidence of intrathecal synthesis of autoantibodies against accessory nodules[13], but protein elevation in the cerebrospinal fluid of patients with anti-CNTN-1 AN is common, and it is believed that it is most likely caused by the destruction of the blood‒brain barrier[13]. The patients we reported also showed a significant increase in the protein level in the cerebrospinal fluid, as well as positive antibody indicators in the cerebrospinal fluid, indicating that this patient may have damaged the blood‒brain barrier. The patient showed weakness of the lower limbs, which was gradually aggravated, and there was muscle atrophy and numbness of both lower limbs and the feeling of stepping on cotton, which was also obviously consistent with the clinical manifestations of this disease. The 24-hour urine protein quantity of the patient was significantly increased, and other urine-related protein indicators suggested renal glomerular damage. Some studies have shown that CNTN-1 protein also exists in podocytes, and IgG4 and CNTN1 antigens coexist in glomerular deposition[11, 14]. At present, the diagnosis of the disease includes the history, electrophysiological diagnosis and detection of antibodies to nodal/paranodal areas, the most important of which is the detection of autoantibodies.
The patient was hospitalized for the first time in our case. We did not fully understand the nodal/paranodal disease, and although CIDP was highly suspected when neuroelectrophysiology suggested that demyelinating motor nerves and sensory nerves were involved, the patient did not improve significantly after hormone shock and third-ball therapy, and because the patient was associated with obvious deep sensory disorders, it was possible to consider SCA combined with peripheral neuropathy. The neuroelectrophysiology of SCA with peripheral neuropathy is nonspecific and can manifest as sensory and axonal neuropathy, sensorimotor axonal neuropathy, demyelinating sensorimotor neuropathy, sensorimotor polyneuropathy with axonal and demyelinating sensations, etc[15]. Later, the patient's condition slowly worsened, and the diagnosis was finally confirmed through antibody detection. Neuroelectrophysiological characteristics of anti-CNTN-1 autoimmune nodopathies are demyelination, prolonged latency of distal motor and slow nerve conduction velocity, conduction block, prolonged or absent latency of the F wave, significant decrease in CMAP amplitude, early axonal damage, slow SCV, and decreased amplitude of SNAP[16]. Notably, the patient's two neuroelectrophysiological examinations suggested that the motor nerve and sensory nerve had progressive conduction disorder, the time limit of the motor unit of needle electromyography was increased in the early stage of the course of the disease, and the spontaneous potential and obvious motor unit potential wave amplitude of needle electromyography were increased in the late stage, indicating denervation performance and suggesting secondary axonal damage after severe myelin sheath injury. The summary of the electrophysiological examination showed that the patient had multiple types of peripheral nerve damage (motor and sensory damage), mainly myelin sheath damage, and myelin sheaths and axons were involved. On magnetic resonance imaging (MRI) of spinal nerve roots in autoimmune nodopathies,they usually show enhancement or thickening of the lumbosacral nerve roots[2].Interestingly, our case showed enhancement not only in the lumbar nerve root but also in the cervical plexus. At the same time, this patient also had multiple nerve root sheath cysts, which had not been reported before.
According to a Japanese study, corticosteroid therapy is effective in 73% of CNTN-1 IgG4-positive patients. Plasma exchange also has a certain effect on CNTN-1 IgG4 antibody-positive patients. Moreover, rituximab can target IgG4 antibodies and achieve good therapeutic effects by depleting B lymphocytes and downregulating the humoral immune response[5, 17]. Notably, some studies have found that the antibody in some small CNTN-1 patients is IgG3 rather than IgG4, and anti-CNTN-1 IgG3 may also appear in the acute phase, while CNTN-1 IgG3 AN patients will have a good response to immunoglobulin, while IgG4 AN patients have a poor response[18, 19]. The patient had an IgG4 antibody, which had a poor effect on corticosteroids and plasma exchange. However, the treatment response after using rituximab is good.
The patient had dysarthria, and we analysed three reasons. (1) Previous studies have shown that cerebellar granule cells also have CNTN-1 expression[13]. However, the patient's cerebrospinal fluid indicated the presence of antibodies, and the cerebrospinal fluid protein level was elevated. Theoretically, it may be possible for antibodies to attack the CNTN-1 protein on the cerebellum, which may lead to cerebellar dysarthria. Interestingly, the patient's cerebellum MRI was normal. Of course, many autoimmune antibody diseases attack the cerebellum and have symptoms that may also have negative MRI. For example, anti-NMDAR autoimmune encephalitis shows cerebellar symptoms, but it is normal on MRI[20]. In addition, nystagmus was elicited in this patient, considering the presence of brainstem and cerebellar involvement, which may explain the cerebellar injury caused by CNTN-1 antibody. (2) The disease may involve the cranial nerve. It has been reported that anti-CNTN-1 may involve the cranial nerve, including the glossopharyngeal nerve, vagus nerve, trigeminal nerve or facial nerve injury, which may show facial paralysis, eye movement problems, dysphagia, etc[6, 7]. If the patient has anti-CNTN-1 antibodies that damage the vestibular nerve and glossopharyngeal nerve, they may have dysarthria and nystagmus. However, if the patient has dysphagia, choking on drinking water and other symptoms, it is more supportive of cranial nerve injury. However, the patient's peripheral nerve dysarthria is generally characterized by flaccid dysarthria, slow speech, and hypernasia and may be accompanied by facial paralysis and drooling. This patient is not like this[12]. (3) The patient's GD1b antibody is positive, and the antibody can affect the brainstem and peripheral nerves. Therefore, it may show peripheral nerve dysarthria.
Overall, our findings have several guiding significance: (1) The onset of anti-CNTN-1 autoimmune nodopathies may not appear as symmetrical limb sensory/motor symptoms similar to the CIDP acute phase but may show dysarthria, tremor or other nonperipheral neuropathy signs as the initial symptoms. Some studies have found that CNTN-1 can be detected in the retina, spinal cord, cerebral cortex, hippocampus, cerebellum and oligodendrocytes[21]. Therefore, nonperipheral neuropathy may also occur in anti-CNTN-1 autoimmune nodopathies, such as decreased memory when attacking the hippocampus, which suggests that the disease caused by CNTN-1 antibody is specific, and this reminds clinicians not to misdiagnose or miss diagnosis. (2) It was previously reported that the disease mainly occurred in middle-aged and elderly people, but this case reminds us that the disease may also be manifested in young people. Of course, the patient developed at a young age, and we speculate that it may be positive for GD1b antibody and anti-CNTN-1 antibody, and GD1b antibody may also involve Ranvier knot, so the patient may have an early onset. (3) Enhanced magnetic resonance imaging of anti-CNTN-1 autoimmune nodopathies may involve not only the lumbosacral region but also the cervical plexus. In addition, this patient also had nerve root sheath cysts. It is temporarily impossible to differentiate congenital nerve root sheath cysts from nerve root swelling-induced cysts, and more cases and longer follow-up observations are needed.
In summary, the first symptom of anti-CNTN-1 autoimmune nodopathies with dysarthria is not common, which may be just a coincidence, but it may also expand a new phenotype of a class of diseases. Regardless, the nature of the relationship between this manifestation and the disease we have considered is still speculative, and we encourage further research. In the end, we hope that if clinicians encounter patients with similar clinical manifestations, the differential diagnosis of cerebellar dysarthria can suggest the disease and try to improve the relevant examinations, especially the differentiation with spinocerebellar ataxia, to avoid misdiagnosing or misdiagnosing the disease to maximize the treatment that patients can help.
Table 1
Neuroelectrophysiological/electromyographic results of patients with anti-CNTN-1 autoimmune nodopathies
Detect nerve | Time point detection | Normal value |
Left/right ulnar nerve | Detection 5 years after onset Left/Right | Detection 8 years after onset Left/Right | |
Motor nerve conduction velocity (m/s) Upper elbow wrist | 13.9/14.3 | 16.6/18.2 | ≥ 50 |
Motor nerve conduction velocity amplitude (mV) Upper elbow wrist | 7.3/8.4 | 2.2/2.2 | ≥ 7.0 |
Latency of motor end (mS) Wrist abductor digiti minimus | 5.35/5.07 | 6.27/5.21 | ≤ 4.0 |
Sensory nerve action potential (uV) Upper elbow wrist | Not elicited | Not elicited | ≥ 7.1 |
Sensory nerve conduction velocity (m/s) Upper elbow wrist | Not checked | Not checked | ≥ 50 |
F wave latency (ms) Wrist abductor digiti minimus | 61.4/64.2 | 81.5/88.5 | ≤ 30 |
F wave occurrence rate (%) Wrist abductor digiti minimus | 58.5/64.5 | 90.0/90.0 | ≥ 75 |
Median nerve | | | |
Motor nerve conduction velocity (m/s) Elbow wrist | 28.8/30.8 | 15.7/15.5 | ≥ 50 |
Motor nerve conduction velocity amplitude (mV) wrist opponens pollicis muscle | 9.7/8.1 | 4.0/1.2 | ≥ 6.0 |
Latency of motor end (mS) Wrist opponens pollicis | 6.04/6.42 | 8.85/9.00 | ≤ 4.2 |
Sensory nerve action potential (uV) Wrist opponens pollicis | Not elicited | Not elicited | ≥ 9.5 |
Sensory nerve conduction velocity (m/s) Wrist opponens pollicis | Not checked | Not checked | ≥ 50 |
F wave latency (ms) Wrist abductor digiti minimus | 60.1/65.7 | 89.4/131.0 | ≤ 30 |
F wave occurrence rate (%) Wrist abductor digiti minimus | 90.0/90.0 | 95.0/40.0 | ≥ 75 |
Tibial nerve | | | |
Motor nerve conduction velocity (m/s) Popliteal fossa ankle | 30.1/31.8 | Not elicited/13.3 | ≥ 40 |
Motor nerve conduction velocity amplitude (mV) Ankle abductor hallucis muscle | 1.10/1.09 | Not elicited/1.04 | ≥ 7 |
Latency of motor end (mS) Ankle abductor hallucis | 9.77/9.55 | Not elicited/11.5 | ≤ 5 |
Sensory nerve action potential (uV) | Not checked | Not checked | ≥ 0.4 |
Sensory nerve conduction velocity (m/s) | Not checked | Not checked | ≥ 35.1 |
F wave latency (ms) Ankle abductor hallucis | Not elicited/not elicited | Not elicited/not elicited | ≤ 51.0 |
F wave occurrence rate (%) Ankle abductor hallucis | Not elicited/not elicited | Not elicited/not elicited | ≥ 75.0% |
Common peroneal nerve | | | |
Motor nerve conduction velocity (m/s) Fibula capitulum ankle | 33.9/35.2 | Not elicited/not elicited | ≥ 45 |
Motor nerve conduction velocity amplitude (mV) Ankle extensor digitorum brevis | 0.57/0.24 | Not e elicited/not elicited | ≥ 3 |
Latency of motor end (mS) Ankle abductor hallucis | 9.35/10.4 | Not elicited/not elicited | ≤ 5 |
Sural nerve | | | |
Sensory nerve conduction velocity (m/s) Middle outer ankle of lower leg | Not checked | Not elicited/not elicited | ≥ 50 |
Sensory nerve action potential (uV) Middle outer ankle of lower leg | Not elicited/not elicited | Not elicited/not elicited | ≥ 3.3 |
electromyogram | | | |
Left abductor digiti minimus | | | |
Average amplitude (uV) | 502 | 1843 | ≤ 1000 |
Average time limit (ms) | 10.2 | 13.0 | ≤ 9.5 |
Resting potential | Not elicited | Not elicited | - |
Left tibialis anterior muscle | | | |
Average amplitude (uV) | 2408 | 1377 | ≤ 1000 |
Average time limit (ms) | 23.5 | 15.0 | ≤ 12.1 |
Resting potential | Not elicited | Lead out P++, F++ | - |