Navigating the diagnostic and therapeutic complexities of autoimmune encephalitis

Characterized by immune-mediated inflammation of the brain, autoimmune encephalitis (AE) encompasses a diverse array of disorders, each presenting with unique as well as overlapping clinical manifestations, and underlying autoantibodies targeting neuronal and glial surface antigens. The clinical spectrum can range from subtle cognitive and behavioural changes to severe seizures, movement disorders, and altered consciousness. However, what truly sets AE apart from other encephalitides is its potential reversibility when identified early and managed appropriately. The identification of N-methyl-D-aspartate receptor (NMDAR) antibody encephalitis in 2007 heralded the discovery of a spectrum of immunotherapy-responsive AE. However, misdiagnosis remains prevalent despite diagnostic algorithms, resulting in unnecessary immunotherapies and delayed treatment of the correct diagnosis. AE diagnosis requires a judicious blend of diagnostic algorithms, antibody testing, and clinical acumen. Testing for antibodies against neuronal surface antigens, preferably in cerebrospinal fluid, is essential, but should be contextual to clinical relevance. Treatment guidelines are largely based on retrospective studies due to a dearth of randomized and prospective data. However, early immunotherapy initiation has proven to be pivotal for optimal outcomes. The therapeutic landscape in AE continues to evolve with the introduction of novel targeted immunotherapies. This review highlights the complexities, and proposes a systematic approach, to the diagnosis and treatment of AE.


INTRODUCTION
The first evidence that immunotherapy could improve some forms of encephalitis came from the identification of voltagegated potassium channel (VGKC) antibodies in association with limbic encephalitis (LE) in 2004. 1 Although the antibody targets were later identified as VGKC-associated proteins rather than VGKC itself, this study was the harbinger for the discovery of immunotherapy-responsive autoimmune encephalitis (AE).The subsequent discovery of N-methyl-D-aspartate receptor (NMDAR) antibody encephalitis (NMDARE) in 2007 and its detailed characterisation 2,3 was the catalyst for the discovery of an expanding group of immunotherapy-responsive AE mediated by autoantibodies directed against neuronal and glial cell surface antigenic targets (Table 1).Some of these syndromes are associated with tumours.In contrast, paraneoplastic neurological syndromes in which antibodies are directed against intracellular antigenic targets poorly respond to immunotherapy since antibodies directed against intracellular antigenic targets are not themselves pathogenic. 4However, some syndromes with antibodies against intracellular targets, such as glial fibrillary acidic protein (GFAP) 5 and glutamic acid SLJoN ⏐ CHANG decarboxylase (GAD), may respond to immunotherapy, perhaps due to coexistence of other yet unidentified pathogenic antibodies. 6 has emerged as the third most common cause of encephalitis after infections and acute disseminated encephalomyelitis. 7,8 us, AE is increasingly considered in the differential diagnoses of patients presenting with subacute memory loss, cognitive dysfunction, psychiatric symptoms or refractory seizures.Despite diagnostic algorithms for AE being proposed in 2016, 9 misdiagnosis remains a frequent clinical problem leading to morbidity resulting from unnecessary immunotherapies and delayed treatment of the correct diagnosis. 10eatment guidelines in AE syndromes are largely based on retrospective studies due to the paucity of randomised or prospective data. 11Despite the overall success of immunotherapy, treatment decisions in AE are hindered by the potential of immunomodulatory medications to cause serious complications and the lack of evidence to guide the optimal duration, and timing of escalation, of treatment.This review highlights the complexities, and proposes a systematic approach, to the diagnosis and treatment of AE.

Diagnosis of AE
3][14] The following four considerations are likely to reduce misdiagnosis.
First, the fulfilment of the three minimal requirements for possible autoimmune encephalitis (Panel 1) is crucial.
Most of the misdiagnoses occur when AE is considered in patients not fulfilling the three essential criteria (Panel 1). 10 The commonest pitfall was to consider AE in people with chronic symptoms (>3 months) without inflammatory changes in the brain Magnetic Resonance Imaging (MRI) scan or cerebrospinal fluid (CSF).The third requirement of the criteria is an imperative for rigorous consideration and careful exclusion of mimics of AE including infectious and postinfectious encephalitis/encephalopathy, Central Nervous System (CNS) vasculitis, CNS lymphoma, epileptic syndromes and psychiatric disorders.Adherence to the first two criteria for possible AE had high specificity of about 90% in ruling out primary psychiatric disorders. 15cond, the three requirements for possible AE should not be used as standalone diagnostic criteria.Instead, these criteria should be used as a checkpoint for entry into the algorithm for the differential diagnosis of AE.Clinical criteria to diagnose AE without the need for antibody testing are shown in Panel 2. A critical caveat to the use of these criteria is that they are not applicable to patients in the very early stages of the disease when they may have few symptoms or an isolated symptom (e.g., only seizures), as the differential diagnosis is too broad.
Third, testing for antibodies against neuronal surface antigens should include CSF.All neuronal surface antibodies occur more frequently is CSF than in serum although anti-LGI1 antibodies are better detected in serum when using currently available cell-based assays 16 .Myelin oligodendrocyte glycoprotein (MOG) antibodies occur more frequently in serum than CSF.Half of the diagnostic errors in patients misdiagnosed with AE occurred due to analysis of serum instead of CSF. 10 Fourth, detection of autoantibodies must be considered in context of clinical relevance.For example, the frequency of thyroid antibodies in patients with autoimmune CNS disorders is not different from that seen in patients with alternative diagnoses or people who are healthy while low titres of GAD65 antibodies are detected in about 8% of healthy people. 16,17 Subacute onset (rapid progression of <3 months) of working memory deficits (short-term memory loss), altered mental status, or psychiatric symptoms 2. At least one of the following: • New focal Central Nervous System (CNS) findings • Seizures not explained by a previously known seizure disorder • Cerebrospinal fluid (CSF) pleocytosis (white blood cell count of more than 5 cells per mm³) • Magnetic Resonance Imaging (MRI) features suggestive of encephalitis

Reasonable exclusion of alternative causes
Panel 1. Diagnostic criteria for possible AE (all three of the following criteria must be met) Probable NMDARE 9 Diagnosis can be made when all three of the following criteria have been met: 1. Rapid onset (less than 3 months) of at least four of the six following major groups of symptoms: • Speech dysfunction (pressured speech, verbal reduction, mutism) • CSF with pleocytosis or oligoclonal bands

Reasonable exclusion of other disorders
Diagnosis can also be made in the presence of three of the above groups of symptoms accompanied by a systemic teratoma.

Panel 2. Clinical criteria for AE
Definite autoimmune limbic encephalitis (LE) 9 Diagnosis can be made when all four of the following criteria have been met: 1. Subacute onset (rapid progression of less than 3 months) of working memory deficits, seizures, or psychiatric symptoms suggesting involvement of the limbic system 2. Bilateral brain abnormalities on T2-weighted fluid-attenuated inversion recovery MRI highly restricted to the medial temporal lobes 3.At least one of the following: • CSF pleocytosis (white blood cell count of more than 5 cells per mm 3 ) • EEG with epileptic or slow-wave activity involving the temporal lobes

Reasonable exclusion of alternative causes
Probable Leucine-rich glioma-inactivated 1 (LGI1)-LE 15 1.Subacute onset (<3 months) of It must be noted that patients with encephalitis mediated by antibodies not included in standard diagnostic panels might be considered seronegative.MOG antibodies may not be routinely tested in patients presenting with focal encephalitis (FLAMES -FLAIR-hyperintense lesions in MOG antibody associated encephalitis with seizures) which may result in it being classified as antibody-negative AE.Gamma-aminobutyric acid A (GABA A ) receptor encephalitis (mainly in children) and anti-neurexin-3α encephalitis at its onset may clinically mimic NMDARE (Table 1), but may be misclassified if specific antibodies are not tested.
Given the severity of its clinical presentation, new-onset refractory status epilepticus (NORSE) is likely to be over diagnosed as antibody-negative AE.NORSE is a poorly defined syndrome that represents multiple disorders.Adherence to the diagnostic algorithm is likely to avoid misdiagnosis and escalation of immunotherapy that is ineffective in non-immune NORSE.
Lack of an objectively measured clinical response after 3 to 4 months of immunotherapy may be considered as evidence against a diagnosis of AE in patients initially diagnosed as antibody-negative AE. 16

Treatment of AE
Evidence from retrospective studies have demonstrated significant benefit of early initiation of first line immunotherapy in AE. 11 However, there is a lack of randomised or prospective data to guide optimal treatment regimens, timing of escalation to second-or third-line therapies and the duration for maintenance therapy in individual AE syndromes.The tiers of immunotherapy in AE is shown in Table 2.This review will focus on the two most common types of AE, namely NMDARE and Leucine-rich glioma-inactivated 1 (LGI1)-LE.

Panel 3. Autoantibody negative AE
Diagnosis can be made when all four of the following criteria have been met: 1. Rapid progression (< 3 months) of working memory deficits (short-term memory loss), altered mental status, or psychiatric symptoms 2. Exclusion of well-defined syndromes of AE (eg, limbic encephalitis, Bickerstaff brainstem encephalitis, acute disseminated encephalomyelitis) 3. Absence of well characterised autoantibodies in serum and CSF, and the presence of at least two (one if paediatric) of the following criteria: • MRI abnormalities suggestive of autoimmune encephalitis  AMPAR, alpha-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid receptor; CASPR2, contactin-associated protein-like 2; DPPX, dipeptidyl-peptidase-like protein-6; GABA, gamma-amino butyric acid; GlyR, glycine receptor; LE, limbic encephalitis; LGI1, leucine-rich glioma-inactivated 1; NAD, no abnormalities detected; NMDAR, N-methyl-D-aspartate; OCB, oligoclonal bands; SCLC, small cell lung cancer; T2H, T2 hyperintensities on MRI SLJoN ⏐ CHANG An understanding of the mechanisms by which the autoantibodies are generated and perpetuated have important implications for rationalising the choice of immunotherapy in AE.Evaluating serum and CSF antibody titres in individuals across multiple AE syndromes demonstrates that autoantibodies are persistently present at higher levels in serum compared to CSF, thereby suggesting that the initial generation of autoantibodies is peripheral. 21Hence, therapeutic options targeting peripheral B-cell compartments or soluble circulating IgG have significant potential and proven efficacy.However, the presence of antigen-specific B-cells in the CSF mediating intrathecal synthesis of autoantibodies highlight the need to simultaneously target the intrathecal compartment in the treatment of AE. 21The immunological targets of currently available immunotherapies are outlined in Figure 1.

Immunotherapy tier Agent
The best outcomes in NMDARE have been reported in patients who received early immunotherapy 22,23 and early removal of an associated tumour, particularly ovarian teratomas that contain B cells with NMDAR reactivity. 23A lack of immunotherapy within 30 days of disease onset was associated with an almost 3-fold increased odds of poor outcome. 22 patients who failed first-line immunotherapies, second-line treatments have been shown to result in good outcomes. 22,23 erall, there is a 10-15% risk of relapse at 2 years, which is significantly lowered in those who receive second-line immunotherapies. 24Emerging evidence show that rituximab administration is associated with a greater likelihood of achieving a good outcome (mRS 2) 25 and 6-fold reduced odds of relapse after ≥24 months follow-up. 22Although there is insufficient evidence to guide the exact timing of escalating treatment, it has been observed that patients receiving earlier initiation of second-line immunotherapy (within 60 days of disease onset) was associated with 7-fold reduced odds of poor outcome compared with later initiation. 22Generally, it is reasonable to escalate to second-line immunotherapy, particularly rituximab, if there is no improvement within 14 days of commencing first-line immunotherapy.In NMDARE, it is usually not necessary to give maintenance doses of rituximab after an initial course.Cyclophosphamide is considered if there is no response within 4 weeks after the first dose of rituximab, or simultaneously with rituximab in patients with severe disease.
Third line immunotherapy is considered in patients with refractory AE not responding to rituximab and/or cyclophosphamide.Tocilizumab, which blocks the interleukin-6 (IL-6) receptor (Figure 1), has been used in combination with teratoma removal, corticosteroids, IVIg and rituximab in a large prospective cohort, demonstrating superior outcomes when compared to either the use of rituximab monotherapy, or the use of only first-line therapies. 26limitation of all cluster of differentiation (CD)-20-targeted approaches such as rituximab is that they do not eliminate the non-CD20 expressing, long-lived plasma cells, which has led to the consideration of monoclonal antibody targeting of antigens specific to these cells.Bortezomib, which targets the proteosome, shows a theoretical predilection for plasma cells, and has been reported to be efficacious in small series of severe refractory cases. 27

LGI1-LE
Immunotherapy has been shown to significantly cause cessation of faciobrachial dystonic seizures (FBDS), prevent cognitive impairment and improve long term outcomes. 28,29 though first-line immunotherapies have been shown to be efficacious in the treatment of LGI1-LE, 30 the evidence for rituximab has been conflicting. 25,31 zures in LGI1-antibody encephalitis are thought to represent acute symptomatic seizures rather than a persisting epileptogenic focus.Thus, the prolonged use of antiseizure medications is often not required in these patients.

Chronic immunotherapy in AE
The role for chronic immunotherapy in AE has not been adequately addressed in the literature.AE can follow a monophasic or relapsing course.Rapid discontinuation of immune suppression, particularly corticosteroids, has been noted to result in early recurrence.Thus, a tapering strategy is recommended once patients have achieved maximal clinical response to the initial immunomodulatory therapy.
Common tapering strategies include an oral prednisolone taper, monthly intravenous methylprednisolone, monthly intravenous immunoglobulin, and rituximab.The latter two agents are preferred when corticosteroids cause intolerable adverse effects.The duration of the taper typically extends over a 2-to-6-month period.Patients who received either rituximab or cyclophosphamide in their initial immunotherapy may not need a taper since these agents achieve a longersustained immunosuppression and have been shown to be associated with a lower risk of relapse 22 .
The optimal duration of maintenance therapy in patients with relapsed disease is unknown.Expert opinion supports a treatment period of three years. 32Azathioprine and mycophenolate mofetil have traditionally been used in this setting, overlapping with oral glucocorticoids for at least 2-to-6 months.

Conclusion
Mimics of AE are common.AE diagnosis should be guided by careful application of diagnostic algorithms and by judicious antibody testing.However, clinical reasoning should prevail in interpreting antibody results, and to do so, clinicians must become familiar with the syndromes of AE and related inflammatory disorders of the CNS.The best outcomes in AE are achieved with early initiation of immunotherapy.The therapeutic landscape of AE is expanding with the introduction of targeted immunotherapy.

FIGURE 1 B
FIGURE 1 • Seizures • Movement disorder, dyskinesias, or rigidity/abnormal postures • Decreased level of consciousness • Autonomic dysfunction or central hypoventilation 2. At least one of the following laboratory study results: • Abnormal electroencephalogram (EEG) (focal or diffuse slow or disorganised activity, epileptic activity, or extreme delta brush)