Isolated Dystonia with NMDAR Antibodies: Expanding the Phenotype or Illustrating the Intricacies of Antibody Testing?

In this issue, two groups report the cases of two young women with isolated (cranio‐) cervical dystonia, who tested positive for N‐methyl‐D‐aspartate receptor (NMDAR) antibodies in the cerebrospinal fluid (CSF). ¹ , ²

The 23‐year‐old developed tremulous cervical and laryngeal dystonia over 5 months, without any other neurological or psychiatric features. Brain magnetic resonance imaging showed T2 hyperintensities in the basal ganglia, midbrain, and pons. NMDAR antibodies were tested positive in serum and CSF. A steroid pulse did not result in improvement of the symptoms or scan; rituximab, and shortly thereafter, botulinum toxin injections were started and significant improvement of the clinical symptoms, but only mild improvement of the brain scan, were observed.

The 35‐year‐old developed a sensation of eye strain and blepharospasm with a sensory trick, as well as jaw dystonia, and over the next 6 months also cervical dystonia. A 2‐month period of fatigue preceded the development of dystonia, but otherwise, there were no associated psychiatric or neurological symptoms. She tested weakly positive for NMDAR antibodies in the serum and strongly positive for NMDAR antibodies in the CSF, using a commercially available fixed cell‐based assay. Serum NMDAR antibodies were confirmed in a reference laboratory. Treatment was installed with an initial pulse of high‐dose steroids, followed by monthly intravenous immunoglobulins (IVIg) and rituximab. Symptomatic treatment with botulinum toxin injections was suspended after four courses because of limited return of symptoms, attributed to the immunotherapy. Six months after initiation of immunotherapy, the patient reported memory difficulties and occasional visual hallucinations (small animals in the periphery of her visual field), with only a minor deficit on formal psychological testing and resolution of the hallucinations without specific treatment. At the last 2‐year follow up, there was only mild blepharospasm, and the authors report a cyclical IVIg response with blepharospasm emerging the days leading up to the next infusion and resolving almost completely in the days following. Serum NMDAR antibodies remained strongly positive.

In neither case, an ovarian teratoma or other tumor was found, and the diagnostic work‐up did not encompass genetic causes of young‐onset dystonia. Both cases are not in keeping with the classic phenotype of anti‐NMDAR encephalitis, which consists in a rapidly progressive, multistage illness with psychosis, cognitive deficits, epilepsy leading up to a state of dysautonomia, and reduced consciousness. ³

On the other hand, the similarity—isolated dystonia in young women—is noteworthy, and partial manifestations, including isolated dystonia, have been described with NMDAR antibodies. ⁴ , ⁵ , ⁶ Moreover, NMDAR‐antibodies tested positive in CSF and a clinical response to immunotherapy was observed. Such cases beg the question if we are entering another era of “expanding the clinical spectrum,” as we have seen in neurogenetics, where the wider use of next‐generation sequencing techniques has greatly expanded the phenotypic spectra of many genes. We recognize later presentations, monosymptomatic manifestations, and other “formes frustes”. Part of that era was, however, also the recognition of the problem variants of unknown significance (VUS) pose in clinical practice. Similarly, with the wider ability of antibody testing, we encounter also more often “antibodies of unknown significance” (AUS). ⁷ In analogy to the criteria of the American College of Medical Genetics for VUS, we proposed some handles for the interpretation of AUS. ⁷ In contrast to VUS, a substantial proportion of AUS can be attributed to false positive test results, which occurs particularly with serum‐only testing. Therefore, one important indicator of a significant antibody finding is the positivity in CSF, and ideally, the proof of intrathecal synthesis of that specific antibody. In both cases in question here, the NMDAR antibodies were detected in CSF, even though we do not know if there was indeed intrathecal synthesis as defined by the antibody index. ⁷

Another AUS handle is the double‐checking with a second test method, for example, demonstrating an immunohistochemistry staining pattern on tissue pertinent to that particular antibody. This confirmation would have been desirable in the two cases discussed here, more so in light of a recent study highlighting the high rates of autoimmune encephalitis misdiagnosis in adults. ⁸ In this large multicenter study, approximately one‐third of patients suspected to have autoimmune encephalitis were in fact misdiagnosed. The antibodies involved included, not surprisingly thyroid antibodies, but also various neuronal antibodies in the serum, including contactin‐associated protein‐like 2 (Caspr2), leucine‐rich glioma‐inactivated 1 (LGI1), glycine receptor, amphiphysin, glutamic acid decarboxylase (GAD), and NMDAR antibodies. Importantly, NMDAR antibodies were also found in the CSF in some cases, but were not validated with a second test method. The main causes for misdiagnosis were overinterpretation of positive serum antibody test results (20%), and misinterpretation of functional, psychiatric, or non‐specific cognitive dysfunction as encephalopathy (38%). The correct diagnoses included functional neurological disorders (25%), neurodegenerative disease (21%), primary psychiatric disease (18%), cognitive deficits from comorbidities (10%), cerebral neoplasms (10%), and other causes (17%) such as mitochondriopathy, infectious disease, vasculitis, and multiple sclerosis. A summary of red flags that should caution against a misdiagnosis of autoimmune encephalitis based on these findings is presented in Table 1. Moreover, the high rate of adverse effects to immunotherapy (20%) is something to carefully consider even in the want of not missing treatable causes. Overall, this study highlights the importance of carefully interpreting antibody test results. The two cases in this issue highlight the difficulties encountered in clinical practice, among which there is also the interpretation of the good response to immunotherapy. We often cannot stage treatment initiations, as one would in an experimental setting to dissect the effect of different therapies. Moreover, we are still learning about the placebo effect in immunotherapies and the role of the immune system in not‐primarily autoimmune disorders. ⁹ , ¹⁰

TABLE 1.

Red flags cautioning against a diagnosis of autoimmune encephalitis according to Flanagan et al. ⁸

Clinical

Insidious onset

Multiple comorbidities that cause cognitive deficits such as polypharmacy, chronic pain, fibromyalgia, sleep disorders

Examination results consistent with a functional neurological disorder

Features of mitochondrial disease

Normal neuropsychological test results

MRI

Normal

Progressive atrophy without signal abnormalities or enhancement

Lesion(s) continuing to expand despite immunotherapy

CSF

Non‐inflammatory (normal cell count, no CSF‐restricted oligoclonal bands)

Serology

TPO antibodies of any titer

Low‐positive anti‐GAD

VGKC antibodies but negative for Caspr2 and LGI1

Low titer antibodies by older generation lab techniques (eg, radio immunoprecipitation assays)

Isolated serum anti‐NMDAR negative in CSF

Immunoblot or line blot antibody positivity in isolation

Low titer anti‐Caspr2

Antibody detection in non‐certified laboratories

Abbreviations: Caspr2, contactin‐associated protein‐like 2; CSF, cerebrospinal fluid; LGI1, leucine‐rich glioma‐inactivated 1; MRI, magnetic resonance imaging; NMDAR, N‐methyl‐d‐aspartate receptor; VGKC, voltage gated potassium channel.

Finally, with the facts we currently have, it appears difficult to confidently reconcile the aspects in favor or against anti‐NMDAR mediated isolated dystonia. They certainly highlight the intricacies of dealing with unusual movement disorder cases with neuronal antibodies and bring out the importance of applying and further developing the proposed handles of AUS.

Author Roles

(1) Research project: A. Conception, B. Organization, C. Execution; (2) Statistical Analysis: A. Design, B. Execution, C. Review and Critique; (3) Manuscript Preparation: A. Writing of the First Draft, B. Review and Critique

B.B.: 3A.

Disclosures

Ethical Compliance Statement: I confirm that I have read the Journal's position on issues involved in ethical publication and affirm that this work is consistent with those guidelines. Informed patient consent was not necessary for this work, and neither was the approval of an institutional review board.

Funding Sources and Conflicts of Interest: B.B. has no conflicts of interest and received no specific funding for this work.

Financial Disclosures for Previous 12 Months: B.B. is supported by the Hurka Foundation and the Koetser Foundation and received royalties from Oxford University Press.