Autoimmune encephalitis (AE) is a form of encephalitis resulting from autoimmune reactions (1). It is generally acute or subacute and has an annual incidence of approximately 0.8/100,000 (2). The typical presentation includes behavioral disorders, psychiatric symptoms, cognitive impairment, seizures, and impaired consciousness (1, 3). Since the first description of AE, characterized as anti-N-methyl-d-aspartate receptor (NMDAR) antibodies in 2007 (4), a variety of encephalitis-associated autoantibodies directed against components of the neuron have been discovered. These include antibodies against leucine-rich glioma-inactivated 1 protein (LGI1), contactin-associated protein-like 2 (CASPR2), γ-aminobutyric acid B receptor (GABA_BR), α-amino-3-hydroxy-5-methyl-4-isoxazole-propionic acid receptor (AMPAR), glycine receptor (GlyR), and so on.

Amphiphysin is a Bin/Amphiphysin/Rvs (BAR) domain containing protein that is involved in clathrin-dependent endocytosis during inhibitory neurotransmission (5). Its SH3 domain acts as a binding site during the formation of clathrin-coated intermediates (6). Amphiphysin has been associated with paraneoplastic neurological syndromes (PNS) in breast or small-cell lung cancers (7). The presence of anti-amphiphysin autoantibodies is most commonly associated with limbic encephalitis (LE) and stiff-person syndrome (8–10). Apart from the above, a number of case reports and small case series have reported a link between anti-amphiphysin antibodies and brainstem encephalitis (11), myelopathy (12, 13), peripheral neuropathy (11, 14), and cerebellar dysfunction (11). Anti-amphiphysin encephalitis is usually treated in the same way as other AE. First-line therapies are corticosteroids, intravenous gamma globulin (IVIG), and plasmapheresis (PLEX) and second-line treatment involves the addition of rituximab or cyclophosphamide in patients unresponsive to first-line therapy (15). Immunotherapy is effective for treating encephalitis patients with anti-amphiphysin antibodies, especially those without tumors (15). Treatment of tumors should be performed without delay to improve outcomes (1).

However, an in-depth understanding of amphiphysin-IgG-associated encephalitis is lacking. Here, we describe the clinical features of 10 anti-amphiphysin encephalitis patients and their outcomes after immunotherapy. A literature review of amphiphysin-antibody-associated neuropathy was also performed.

Patients with anti-amphiphysin antibody positivity in serum or cerebrospinal fluid (CSF) admitted to the Beijing Tiantan Hospital between 2018 and 2022 were reviewed. Ten patients fitting the criteria of autoimmune encephalitis updated in 2016 were retrospectively enrolled (1).

We retrospectively enrolled 10 patients diagnosed with amphiphysin-antibody-related encephalitis. The male: female ratio was 4:6, which is consistent with previous studies showing a greater frequency among female patients (17–19). The age of onset was during middle or old age, consistent with previously published data (17–19).

Anti-amphiphysin encephalitis has a variety of clinical manifestations, including limbic encephalitis, stiffness, limb weakness/numbness, paresthesia, ataxia, sleep disorders, and dysautonomia, amongst others. Of these, limbic encephalitis is the most common symptom. Notably, patients who conform to the AE criteria should be suspected of having anti-amphiphysin antibodies when presenting with stiffness or paresthesia, even though it is rare.

Stiff-person syndrome (SPS) is a characteristic manifestation in amphiphysin-antibody-positive cases, although it did not appear as frequently as expected in our cohort (18–20). Amphiphysin-antibody-related SPS involves neck and arm instead of axial muscles of trunk and thus usually confused clinicians (18). One patient in our study presented unilateral upper limb stiffness, accompanied by paresthesia and dysautonomia. She fitted the diagnosis of progressive encephalomyelitis with rigidity and myoclonus (PERM), a severe immunophenotype of stiff-person spectrum disorder (SPSD) (21). This patient showed strong anti-amphiphysin antibody positivity in her serum, much higher than that seen in the other patients, suggesting a potential link between higher antibody titers and SPS severity. However, the relationship between antibody titers and the severity of clinical manifestation is still not clear in AE and requires future in-depth research.

Patients also demonstrated symptoms of myelitis. In Patient #9, the MRI showed the presence of a lesion in the posterior horn (C2-7). Similarly, Galassi et al. reported longitudinally extensive transverse myelitis (LETM) in a 40-year-old woman with amphiphysin antibody positivity (22). A study also demonstrated post-mortem histopathological evidence of the presence of CD8+ T cell-mediated immune responses in the spinal cord and dorsal root ganglia (17), confirming the involvement of the spinal cord.

Sensory neuronopathy has been found to be closely related to the presence of amphiphysin antibodies by previous studies (23). Our two patients, however, despite their complaints of paresthesia, did not meet the criteria of sensory neuronopathy (24). Besides, it had been considered that involvement of motor nerve or CSF inflammatory may indicate a paraneoplastic origin (25).

Amphiphysin is an intraneuronal protein that is targeted in 5-10% of PNS cases, and the presence of anti-amphiphysin antibodies has been listed as a significant risk factor for tumor occurrence (>80%) (23). Amphiphysin antibodies are widely known to be associated with both breast and small-cell lung cancers. Associations with other cancers have also been reported, including angiosarcoma and thyroid adenoma ( Supplementary Table 1 ). Nevertheless, a much lower rate of cancer (2/10) was found in this study (19). No new neoplasms were diagnosed after discharge in our patient cohort. It is possible that insufficient follow-up may partially account for our low detection rate. Repeated screening every four or six months for two years is recommended (23). Accordingly, the necessity of cancer screening must be emphasized, and systemic tumor screening requires whole-body PET-CT as the first choice.

Amphiphysin antibodies are usually detected by immunoblotting, the gold standard for the identification of intracellular antibodies. Previous studies have reported the co-existence of amphiphysin antibodies with anti-neuronal nuclear antibody type 1 (ANNA1) and CRMP5 antibodies (17, 19). Here, we observed a co-occurrence of amphiphysin antibodies with CV2 and AMPAR antibodies. This overlap between amphiphysin antibodies and other subtypes of autoantibodies suggests that AE has both a complex etiology and mechanism. Furthermore, although not included in the present study, we have noticed amphiphysin antibody positivity in some AD (Alzheimer’s disease) and PD (Parkinson’s disease) patients. Although cognitive impairment may be a common manifestation, the progressive course of the disease suggests neurodegeneration rather than encephalitis. Thus, amphiphysin antibodies could be either causative agents or nonpathogenic bystanders caused by other CNS diseases. In terms of clinical diagnosis, the identification of this disease should thus be based on a thorough understanding of amphiphysin antibody pathology as well as clinical experience.

Patients with anti-amphiphysin encephalitis did not show distinctive features on either EEG or brain MRI. Involvement of the mesial temporal lobe was most common, which may mimic other subtypes of AE. Furthermore, involvement of the thalamus was found in one patient, consistent with the findings of a previous report (10). PET-CT revealed evidence of abnormal metabolism in several brain regions, although these were not specific. With the development of imaging analyses, more attention focusing on characteristic structural and functional imaging patterns is needed.

In our study, most patients received IVIG or corticosteroids along with ASMs, and their mRS scores reduced at follow-up. In animal experiments, animals exposed to purified amphiphysin antibodies showed symptom relief after plasmapheresis (26, 27). Clinically, a combination of immunotherapy and tumor treating is required (28). Besides, responses to high-dose benzodiazepine have also been noted (18). Follow-ups of the patients showed that the duration of symptoms varied under standard therapy. Although the seizure frequencies were observed to be reduced during follow-ups of 1 to 33 months, there was minimal improvement in cognitive impairment, sleep disturbances, or psychotic mood disorders. By tracking for 13 years, Taube et al. observed a similar outcome in a patient who was treated promptly with immunotherapy (29). As in other forms of AE, anti-amphiphysin encephalitis presenting with acute symptomatic seizures has a lower risk of subsequent unprovoked seizure development, which perhaps explains the disappearance of seizures in most patients.

There are some limitations in this study. Firstly, the sample size may have been too small. Secondly, immunoblotting is a semi-quantitative assay that does not show the exact titers of antibodies and may be subject to false-positive or false-negative results. In addition, in some cases, the absence of important examination findings such as electromyography and polysomnography hampered physician awareness of the disease and its treatment. In the future, the sample size should be expanded for more in-depth research.

Amphiphysin-antibody-related encephalitis is a rare type of autoimmune encephalitis. The most common manifestation was limbic encephalitis and a lower association with cancer was observed compared with that reported in other PNS cohorts. Although seizure frequencies declined in response to immunotherapy, psychiatric and cognitive symptoms, as well as sleep disturbances, often remained. The awareness of the clinical features of amphiphysin antibody-related encephalitis provides valuable information for a better understanding of the disease and may help to facilitate its early diagnosis, treatment strategy, and prognosis prediction.

The original contributions presented in the study are included in the article/ Supplementary Material . Further inquiries can be directed to the corresponding authos.

The studies involving human participants were reviewed and approved by KY2021-088-03. The patients/participants provided their written informed consent to participate in this study. Written informed consent was obtained from the individual(s) for the publication of any potentially identifiable images or data included in this article.

QW and YS concepted, designed, and supervised the study. YS, XQ and DH acquired the data. YS and XQ analyzed and interpreted the data, provided statistical analysis, had full access to all of the data in the study, and are responsible for the integrity of the data and the accuracy of the data analysis. YS and XQ drafted the manuscript. ZZ, YZ and QW critically revised the manuscript for important intellectual content. All authors contributed to the article and approved the submitted version.