# paper.pdf ## Metadata - PDFFormatVersion=1.4 - Language=en-US - IsLinearized=false - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - Creator=Arbortext Advanced Print Publisher 9.1.510/W Unicode - ModDate=D:20260515151741+05'30' - Title=WNL-2025-206811 1..13 - Producer=Acrobat Distiller 7.0 (Windows) - CreationDate=D:20260515130552+05'30' - xmp:modifydate=2026-05-15T15:17:41+05:30 - xmp:metadatadate=2026-05-15T15:17:41+05:30 - xmp:createdate=2026-05-15T13:05:52+05:30 - xmp:creatortool=Arbortext Advanced Print Publisher 9.1.510/W Unicode - pdf:producer=Acrobat Distiller 7.0 (Windows) - dc:title=WNL-2025-206811 1..13 ## Contents ### Page 1 ![Figure]() RESEARCH ARTICLE OPEN ACCESS # Autoimmune Encephalitis as Treatment-ResponsiveCause of Rapidly Progressive Dementia A Multicenter Prospective Cohort Study Robin W. van Steenhoven,1 Anna E.M. Bastiaansen,1 Jeroen Kerstens,1,2 Yvette S. Crijnen,1 Tessa Brand,1Juliette Brenner,1 Tessa M. Bienfait,1 Juna M. de Vries,1 Yoav D. Piura,3 Nihal Satyadev,3 Marienke A.A.M. De Bruijn,1,4 Mariska M.P. Nagtzaam,1 Ece Erdag Turgeon,1 Suzanne C. Franken,1Sharon Veenbergen,5 Rutger K. Balvers,6 Marc Langedijk,7 Robert M. Verdijk,8 Marcel M. Verbeek,9Charlotte E. Teunissen,10 Yolande A.L. Pijnenburg,11 Frank Jan De Jong,1 Annemieke J.M. Rozemuller,12Nina L. Fransen,12 Elise G.P. Dopper,1 Peter A.E. Sillevis Smitt,1 Gregory S. Day,3 Harro Seelaar,1 andMaarten J. Titulaer1 Neurology® 2026;106:e214933. doi:10.1212/WNL.0000000000214933 ## Abstract ### Background and Objectives Early recognition of patients with rapidly progressive dementia (RPD) attributed to autoimmune encephalitis (AE) is important because rapid initiation of immunotherapy improvesoutcomes. However, ancillary testing is often time-consuming because of the broad differentialdiagnosis, highlighting the need for accurate patient selection based on early clinical features.We aim to determine the frequency of AE subtypes in patients with RPD (AE-RPD) andcharacterize presenting subphenotypes (i.e., symptoms in addition to dementia) comparedwith other diagnoses. ### Methods This prospective multicenter observational cohort study was conducted from December 2019to December 2024 across centers in the Netherlands and included adult patients with RPD,defined as dementia beginning within 1 year of symptom onset. Clinical features and ancillarytesting data were collected and reviewed by 3 neurologists. Serum and CSF were evaluated forneuronal/glial autoantibodies. Correspondence Dr. Titulaer MORE ONLINE ![Figure]() #### Class of Evidence Criteria for ratingtherapeutic and diagnosticstudies #### Supplementary Material ### Results A total of 147 patients were included (46% female; median age 67 years, range 36–86). AE-RPDwas the largest diagnostic category (58/147; 39%) and most common treatment-responsive causeof RPD (58/95; 61%). Neurodegenerative diseases (20/147; 14%) and Creutzfeldt-Jakob disease(CJD, 17/147; 12%) accounted for the largest nonresponsive causes of RPD. RPD accompaniedby seizures at first presentation was more frequent in AE-RPD compared with other diagnoses(20/58; 34% vs 9/89; 10%; p < 0.001), with anti-LG1 (11/20; 55%) encephalitis representing themost frequent underlying subtype. Seizures manifested as subtle focal events in 9 of 20 patientswith AE-RPD (42%), complicating early clinical identification. Movement disorders were observed at presentation in similar proportions of patients with AE-RPD (17%) and CJD (35%; p =0.11) but were more likely to emerge within 3 months of symptom onset in patients with RPDAE (80% vs 25%; p = 0.004). In patients with AE-RPD without seizures, autoimmune glialfibrillary acidic protein (GFAP) astrocytopathy was the most common subtype (8/31; 26%),mostly presenting with prominent psychotic features or movement disorders (both 4/8; 50%). 1Department of Neurology, Erasmus University Medical Center, Rotterdam, the Netherlands;2Department of Neurology, Antwerp University Hospital, Belgium;3Department ofNeurology, Mayo Clinic in Florida, Jacksonville;4Department of Neurology, Elisabeth-TweeSteden Ziekenhuis, Tilburg, the Netherlands;5Department of Medical Immunology, ErasmusUniversity Medical Center, Rotterdam, the Netherlands;6Department of Neurosurgery, Erasmus University Medical Center, Rotterdam, the Netherlands;7Department of Neurology,Treant Hospital, Emmen, the Netherlands;8Department of Neuropathology, Erasmus University Medical Center, Rotterdam, the Netherlands;9Department of Neurology and HumanGenetics, Donders Institute for Brain Cognition and Behavior, Radboud University Medical Center, Nijmegen, the Netherlands;10Neurochemistry Laboratory, Department ofChemistry, Amsterdam University Medical Center, the Netherlands;11Department of Neurology, Amsterdam University Medical Center (VUMC), the Netherlands; and12Department ofNeuropathology, Amsterdam University Medical Center, the Netherlands. The Article Processing Charge was funded by the authors. This is an open access article distributed under the, which permits unrestricted use, distribution, and reproduction in any medium,provided the original work is properly cited. Copyright © 2026 The Author(s). Published by Wolters Kluwer Health, Inc. on behalf of the American Academy of Neurology. e214933(1) ### Page 2 ## Glossary AE = autoimmune encephalitis; AE-RPD = AE subtypes in patients with RPD; CBA = cell-based assay; CJD = CreutzfeldtJakob disease; EMC = Erasmus University Medical Center; FBDS = faciobrachial dystonic seizure; FLAIR = fluid-attenuatedinversion recovery; GFAP = glial fibrillary acidic protein; IHC = immunohistochemistry; IQR = interquartile range; LGI1 =leucine-rich, glioma-inactivated 1; mRS = modified Rankin Scale; p-tau = phosphorylated tau-181; RPD = rapidly progressivedementia; RT-QuIC = real-time quaking-induced conversion; SN-AE = seronegative AE; t-tau = total tau. ### Discussion In this study, AE was the most common treatment-responsive cause of RPD. Anti–leucine-rich, glioma-inactivated 1 encephalitisand autoimmune GFAP astrocytopathy were the predominant subtypes in AE-RPD. ## Introduction Rapidly progressive dementia (RPD) is generally defined asthe development of dementia within 1 or 2 years of symptomonset.1-3 The syndrome represents a significant challenge toclinicians because the differential diagnosis of RPD is broadand includes various treatment-responsive diseases, as opposed to dementia with a slowly progressive course.1,2,4-7Consequently, RPD often necessitates an extensive and timeconsuming diagnostic evaluation, while the time to act islimited because treatment effects are often most beneficial inearly disease stages.8 Consequently, organization and prioritization of ancillary testing is essential to support early recognition of patients with potentially treatment-responsivecauses of RPD and to prevent harmful diagnostic delays.1,8,9In addition, early diagnostic certainty of nontreatable diseases,most notably Creutzfeldt-Jakob disease (CJD), is crucial forpatients and caregivers because it enables effective counselingduring the early stages of the disease and enables enrollmentin clinical trials of putative disease-modifying therapies forprion disease.1,8 A major improvement in the diagnostic evaluation of RPD inthe past decades is the discovery of autoimmune encephalitis(AE) and various associated disease-specific autoantibodies.10Previous premortem and postmortem studies identified AE asimportant treatment-responsive mimic of CJD and neurodegenerative dementia.1,11-14 Early recognition of AE is importantbecause early treatment is associated with improved short-termand long-term outcomes.15-17 More specifically, neuronal autoantibody status provides relevant information regarding tumor association, preferred treatment strategies, relapse risk, andlonger term outcomes.18 In addition, identification of AE ata later onset age has been shown to be more complicatedbecause of absence of direct evidence of inflammation, such asmesiotemporal hyperintensities on brain MRI or pleocytosis inCSF, and a broader differential diagnosis including variousneurodegenerative diseases and comorbidities.19,20 Various RPD cohorts have been published since the firstdescriptions of AE, with the proportion of cases attributed to AE (AE-RPD) varying from approximately 5% to 30%.1,6,7,9,21However, extensive descriptions on the early clinical presentation of AE and different subtypes are scarce. Filling thisgap is important because proper targeted ancillary testing andautoantibody studies facilitate early diagnosis and treatment.We sought to determine the frequency of AE in a large prospective cohort of RPD. In addition, we comprehensivelycharacterized the early clinical presentation of different AEsubtypes in comparison with other causes of RPD to improveearly recognition of patients with AE-RPD. ## Methods ### Study Design and Participants The Erasmus University Medical Center (EMC) is the Dutchnational referral center for neuroinflammation and accreditedEuropean Reference Network site, as well as the academiccenter for patients with neurodegenerative diseases. Neurologists nationwide consult the EMC regarding patients withRPD or atypical dementia to coordinate diagnostic evaluationfor suspected rare causes, including AE and CJD. Patients were recruited between December 2019 and December2024 using 2 complementary approaches. First, patients referredto the EMC by neurologists, geriatricians, or general practitionersfor suspected RPD were evaluated by the authors (R.W.v.S.,M.J.T., H.S., J.d.V., E.G.P.D., F.J.d.J.). In addition, some patientswere referred to and seen at Amsterdam University MedicalCenter (location VUmc), another large academic memory clinicwithin the Netherlands (R.W.v.S., Y.A.L.P.). Second, patientswho were assessed by a neurologist at another Dutch hospital(both general and academic) and suspected of having RPD werediscussed with a neurologist from the EMC as part of the national consultation system, at the discretion of the assessingneurologist at the referring hospital. All referrals and nationalconsultations for suspected RPD during the study period wereprospectively and consecutively screened for eligibility bya neurologist from the research team (R.W.v.S., M.J.T., H.S.)through a standardized review of medical records and neuroimaging studies. Neurology | Volume 106, Number 11 | June 9, 2026 e214933(2) ### Page 3 We selected adult patients who developed dementia within - 1 year of symptom onset,3 with dementia defined accordingto 2011 NINCDS-ADRDA criteria.22 Patients with clearstructural intracranial lesions (e.g., subdural hematoma orspace-occupying mass) or typical presentations of psychiatric disorders or delirium with evident causes (e.g., infection) were excluded. Patients with cognitive dysfunctioncausing substantial interference with daily activities within - 1 year were enrolled after informed consent was obtained. ### Clinical and Imaging Data Once enrolled, information was prospectively collected concerningdemographic characteristics (i.e., age at presentation and sex),medical history, clinical presentation, level of functioning (modifiedRankin Scale [mRS]23), ancillary testing, and treatment strategiesfrom prospective review of medical records. Clinical symptomswere assessed at first presentation (i.e., presenting phenotype) andthroughout the course of clinical evaluation. Next to the prominentdementia symptoms, clinical presentations were categorized into 5different subphenotypes, according to the most dominant supportive symptom: seizures; movement disorders; psychiatric; purecognitive; or cerebellar/brainstem (i.e., infratentorial) localization.Faciobrachial dystonic seizures (FBDSs) were classified as seizures(not movement disorders).24 Cerebellar ataxia was assessed separately from other movement disorders because of differing etiology. Neuroimaging features were assessed through direct examinationof the MRI scans. Mesiotemporal T2/fluid-attenuated inversionrecovery (FLAIR) hyperintensities on brain MRI were consideredsuggestive of AE after excluding additional atypical features(i.e., enhancement, restricted diffusion) or MRI abnormalitiestypical for a specific AE subtype (i.e., perivascular enhancement asdescribed in autoimmune glial fibrillary acidic protein (GFAP)astrocytopathy or multifocal corticosubcortical T2/FLAIRhyperintensities as seen in anti–GABAA-R encephalitis).18,25-28 Inpatients who underwent brain biopsy as part of the evaluation,neuropathologic reports were reviewed and discussed witha neuropathologist and the findings were integrated with otherancillary tests, as previously described.29 Clinical diagnoses werevalidated through autopsy, when performed. ### Laboratory Studies All patients underwent extensive neuronal and glial autoantibodytesting in serum and CSF using cell-based assays (CBAs) andimmunohistochemistry (IHC).30,31 Cell surface autoantibodieswere tested using commercial CBAs (Euroimmun, L¨ubeck, Germany) at the ISO 15189–accredited diagnostic laboratory at EMCor in-house CBAs. Samples with positive CBAs were consideredpositive when confirmed by IHC.30,31 Anti-GFAP–positive CSFsamples were confirmed using a cerebellum (primate) slide (InovaDiagnostics, San Diego, CA). CSF total tau (t-tau), phosphorylatedtau-181 (p-tau), 14-3-3, and real-time quaking-induced conversion(RT-QuIC) on prion protein were performed on request by thetreating physician in the ISO 15189–accredited laboratory at theRadboud UMC. t-tau and p-tau were quantified using ELISAs(Lumipulse Fujibero, Ghent, Belgium). Abnormality was definedby cutoffs: t-tau >400 pg/mL, p-tau >64 pg/mL, and t-tau/p-tau ratio >30. t-tau levels above the level of quantification(i.e., >1,300 pg/mL) were considered to be extremely elevated. ### Definitions Medical records and neuroimaging were reviewed by 3 neurologists specialized in neurodegenerative diseases or neuroinflammatory disorders and final diagnoses established throughconsensus (R.W.v.S., H.S., M.J.T.). Patients with positive glial/neuronal autoantibodies and a compatible clinical syndrome wereclassified as AE-RPD. The 2016 clinical AE criteria were appliedfor probable seronegative AE (SN-AE).18 Established criteria wereused to define other neuroinflammatory disorders,18,32 neurodegenerative diseases,22,33-36 CJD,37 and psychiatric disorders.38Diagnoses were categorized as potential treatment-responsive ornonresponsive. Treatment responsiveness was defined as evidence from the published literature that disease-specific treatments (e.g., immunotherapy and antibiotics) have the potential toresult in sustained and clinically meaningful improvement incognitive function, with potential recovery toward functional independence.1 Autoimmune/inflammatory conditions, metabolic/toxic disturbances, primary psychiatric disorders, CNS lymphomas, CNS infections, epilepsy, and dural arteriovenous fistula wereclassified as treatment-responsive. Neurodegenerative diseases,vascular cognitive impairment and dementia, and World HealthOrganization grade 4 glioma were categorized as nonresponsive.1In a few patients without known diagnoses (n = 7), treatmentresponsiveness was based on (sustained) effectiveness of previoustreatments (n = 1). ### Statistical Analysis Statistical analyses were performed with IBM SPSS 25.0(SPSS Inc., Chicago, IL) and R statistical software (version4.3.2; R Core Team, Vienna, Austria, 2023). The Pearson χ2test or the Fisher-Freeman-Halton test, when appropriate,was used for patient characteristic analysis and group comparisons, encompassing categorical data. p Values were 2sided and considered statistically significant when below 0.05. ### Standard Protocol Approvals, Registrations,and Patient Consents Written informed consent for usage of medical information forresearch purposes was obtained from participants or their caregivers. The study protocol was approved by the institutionalreview board (NL58014.078.18) and performed according tothe Strengthening the Reporting of Observational Studies inEpidemiology reporting guideline for observational research.39 ### Data Availability Anonymized study data will be shared pending review of a request to the corresponding author from qualified individuals. ## Results ### Patient Characteristics and ### Diagnostic Categories A total of 286 patients with suspected RPD were screened foreligibility, of whom 147 (51%) met criteria for RPD and were Neurology | Volume 106, Number 11 | June 9, 2026 e214933(3) ### Page 4 included. Ninety of 147 patients (61%) were evaluated andenrolled at the EMC or Amsterdam University MedicalCenter (location VUmc), or were seen by the authors. Sixtyseven patients (46%) were female (Table 1). The median agewas 67 years (interquartile range [IQR] 62–75 years, range36–86). Time from symptom onset to first presentation was 7 weeks (IQR 1–8 weeks, range 0–62); the median follow-upwas 33 months (IQR 19–44 months, range 0.5–62). Pairedserum and CSF were tested for neuronal/glial autoantibodiesin 146 of 147 patients (99%); only serum was tested in theremaining patient. Ninety-five of 147 patients (65%) werediagnosed with a potentially treatment-responsive cause ofRPD (Figure 1, Table 2). AE was the most commontreatment-responsive cause of RPD (58/95; 61%), followedby other neuroinflammatory disorders (13/95; 14%) andmetabolic disturbances, intoxications, or delirium (combined9/95; 10%). Neurodegenerative diseases (20/52; 38%) andCJD (17/52; 33%) represented the most common nonresponsive causes of RPD. Autopsy was performed in 8 of 147patients (5%), showing 100% concordance between clinicaland neuropathologic diagnoses (CJD 5/8; 62%, neurodegenerative disease 2/8; 25%, AE 1/8; 13%). ### Clinical Presentation and Ancillary Testing Seizures were more common in AE-RPD compared withother diagnoses (34% vs 10%; p < 0.001; Table 1), and AE wasthe most common underlying disorder in patients with RPDand seizures (20/29; 69%; Figure 2). Cerebellar/brainstemsigns were more common in non-AE diagnoses (19/89; 21%vs 5/58; 9%; p = 0.041), including 11 of 19 patients (58%)with RPD attributed to CJD (Figure 2). Hyponatremia,mesiotemporal T2/FLAIR hyperintensities on MRI, and CSFpleocytosis were more common in patients with AE-RPD vsother diagnoses (Table 1). By contrast, restricted diffusion onbrain MRI was more frequent in non-AE diagnoses (18% vs4%; p = 0.009), particularly in ≥2 cortical regions or caudate/putamen (8% vs 0%; p = 0.028)—findings exclusively seen inpatients with CJD (7/7; 100%). The median time from symptom onset to dementia wasshorter in patients with potentially treatment-responsive RPDthan in those with nonresponsive RPD (8 vs 22 weeks),corresponding to a median difference of 14 weeks (95% CI5.5–19.0; p < 0.001). Most patients with treatment-responsivecauses met RPD criteria within 3 months of symptomaticonset (58% vs 27%; p < 0.001; eTable 1). Abnormal movements were identified at first presentation in similar proportions of patients with AE-RPD (17%) and CJD (35%; p =0.11; eTable 2) but were more frequent over the course ofRPD in patients with CJD (71%) vs other causes (26%; p <0.001), with a higher hazard of developing abnormal movements in those with CJD compared with AE (hazard ratio 3.9,95% CI 1.8–8.5; Figure 3). Most patients with AE-RPD(80%) developed movement disorders within 3 months aftersymptom onset, whereas patients with CJD-RPD mostly developed these later (75% beyond 3 months; p = 0.004;Figure 3). CSF t-tau levels above the level of quantification Neurology | Volume 106, Number 11 | June 9, 2026 (>1300 pg/mL) and t-tau/p-tau >30 were found in 7 patientswith a diagnosis other than CJD (eTable 3). Five of thesepatients (63%) were diagnosed with AE, including SN-AE(n = 3), anti-GABAAR, and anti-GABABR (each, n = 1).These patients showed signs considered atypical for CJD,including seizures (2/5; 40%), MRI suggestive of AE withoutrestricted diffusion, and CSF pleocytosis (5/5; 100%). RTQuIC was exclusively positive in patients with CJD but wasalso negative in 3 patients with definite (i.e., autopsy-proven)sporadic CJD. ### AE Subtypes Anti–leucine-rich, glioma-inactivated 1 (LGI1) (19/58;33%), SN-AE (11/58; 19%), and autoimmune GFAPastrocytopathy (8/58; 14%) were the 3 most common subtypes in AE-RPD (Figure 1, Table 2). The seizure subphenotype represented the most common clinicalpresentation of AE-RPD (20/58; 34%, Figure 2). Amongthese, anti-LGI1 (11/20; 55%) was the most common underlying subtype, followed by anti–GABAA/B-R (4/20; 20%;Figure 2). At first presentation, 13 of 20 patients with AERPD and seizures (65%) had focal seizures, including 9 (69%)without generalized seizures (eFigure 1). Six of 13 patients(46%) presented with subtle nonmotor, autonomic, or dyscognitive seizures, and 3 of 13 patients (23%) presented withFBDS. Seizures were missed at first presentation in 5 of 20patients with AE (25%). Anti-LGI1 was the most commonunderlying AE subtype in patients with focal seizures (9/13;69%). Ultimately, 27 of 58 patients with AE-RPD (47%)developed seizures over the illness course (Figure 4A), with17 of 27 (63%) experiencing seizures within 3 months ofsymptom onset (median 4 weeks; IQR 0–17 weeks, range0–39; Figure 4B). Eleven of 13 seizure-positive anti-LGI1patients (85%) developed seizures within 3 months. In the pure cognitive subphenotype and among all seizurenegative patients, SN-AE was the most common AE-RPDsubtype (8/18; 44% and 7/31; 23%, respectively; Figures 2and 4A). Eight of 11 patients with SN-AE (73%) presentedwith a pure cognitive subphenotype, while additional neurologic symptoms were present in the remaining patients, including seizures (2/11; 18%) or movement disorders (1/11;9%). Ultimately, 4 of 12 patients with SN-AE (33%) developed seizures, all occurring more than 3 months aftersymptom onset (Figure 4B). ### Autoimmune GFAP Astrocytopathy Autoimmune GFAP astrocytopathy was the most commonsubtype in seizure-negative AE-RPD patients (8/31; 26%,Figure 4A, Table 3). Six of 8 patients with autoimmune GFAPastrocytopathy (75%) were male, with a median age at onset of66 years (IQR 60–72, range 56–83). Prodromal weight loss wasreported by 4 of 8 patients (50%). Seven of 8 patients (88%) hadclear neurologic or psychiatric signs, in addition to RPD. Psychiatric symptoms were present in 4 of 8 patients (50%), mainlydelusions and visual hallucinations dominating the phenotype in 3 of 8 patients (38%). Myoclonus (3/8; 38%) and tremor (2/8; e214933(4) ### Page 5 | All (N = 147) | AE (N = 58; 39%) | Other (N = 89; 61%) | p Value | | - | - | - | - | | Female sex, n (%) | 67 (46) | 26 (45) | 41 (46) | 0.88 | | Age at presentation, y, median IQR; range | 67; 62–75; 36–87 | 68; 62–74; 45–87 | 67; 62–73; 36–87 | 0.48 | | Medical history, n (%) | | Autoimmune disease | 41 (28) | 15 (26) | 29 (29) | 0.66 | | Systemic malignancy | 32 (22) | 11 (19) | 21 (24) | 0.51 | | Initial presentation, n (%) | | Prodromal symptoms | 43 (29) | 18 (31) | 25 (28) | 0.70 | | Behavioral symptoms | 66 (45) | 21 (36) | 45 (51) | 0.08 | | Psychiatric symptoms | 53 (36) | 24 (41) | 29 (33) | 0.28 | | Hallucinations | 30 (20) | 15 (26) | 15 (17) | 0.19 | | Delusions | 19 (13) | 10 (17) | 9 (10) | 0.21 | | Agitation | 21 (14) | 11 (19) | 10 (11) | 0.19 | | Mania | 3 (2) | 1 (2) | 2 (2) | 0.83 | | Sleeping disorders | 36 (25) | 14 (24) | 22 (25) | 0.94 | | Movement disorders | 32 (22) | 10 (17) | 22 (25) | 0.28 | | Myoclonus | 12 (8) | 4 (7) | 8 (9) | 0.65 | | Parkinsonism | 9 (6) | 2 (2) | 7 (8) | 0.23 | | New-onset seizures | 29 (20) | 20 (34) | 9 (10) | <0.001c | | Focala | 17 (12) | 13 (22) | 4 (5) | <0.001c | | Generalized | 17 (12) | 11 (19) | 6 (6) | 0.024c | | Cerebellar/brainstem symptoms | 24 (16) | 5 (9) | 19 (21) | 0.041c | | Autonomous symptoms | 10 (7) | 4 (7) | 6 (7) | 0.61 | | Impaired consciousness | 10 (7) | 6 (10) | 4 (5) | 0.15 | | Hyponatremia (<136 mmol/L) | 36 (25) | 21 (36) | 15 (17) | 0.008c | | MRI brain performed, n (%) | 146 (99) | 57 (98) | 89 (100) | 0.21 | | Mesiotemporal hyperintensitiesb | 37 (25) | 30 (53) | 7 (8) | <0.001c | | Unilateral | 11 (8) | 7 (12) | 4 (5) | 0.082 | | Bilateral | 26 (18) | 23 (40) | 3 (3) | <0.001c | | Enhancement | 13/71 (18) | 6/30 (20) | 7/41 (17) | 0.75 | | Restricted diffusion | 18 (12) | 2 (4) | 16 (18) | 0.009c | | ≥2 cortical regions | 8 (6) | 0 (0) | 7 (8) | 0.028c | | Caudate/putamen | 7 (5) | 0 (0) | 7 (8) | 0.028c | | CSF performed, n (%) | 146 (99) | 58 (100) | 88 (99) | 0.42 | | WBC >5/μL | 61 (42) | 33 (58) | 28 (32) | 0.002c | | CSF-specific oligoclonal bands | 24 (38) | 12/23 (52) | 12/40 (30) | 0.081 | | EEG performed, n (%) | 101 (69) | 39 (67) | 62 (71) | 0.76 | | Epileptic abnormalities | 16 (16) | 10 (26) | 6 (10) | 0.025c | | PSWCs | 4 (4) | 0 (0) | 4 (6) | 0.15 | Continued Neurology | Volume 106, Number 11 | June 9, 2026 e214933(5) ### Page 6 | All (N = 147) | AE (N = 58; 39%) | Other (N = 89; 61%) | p Value | | - | - | - | - | | Systemic tumor, n (%) | 15 (10) | 9 (16) | 6 (7) | 0.076 | | mRS score, median; IQR; range | 3; 3–4; 2–5 | 3; 3–3; 2–5 | 3; 3–4; 2–5 | 0.39 | | Duration between onset and dementia, median; IQR; range | 13; 3–25; 0–54 | 13; 4–24; 0–54 | 13; 3–26; 0–50 | 0.64 | | Dementia <3 mo, n (%) | 69 (47) | 27 (47) | 42 (47) | 0.94 | Abbreviations: FLAIR = fluid-attenuated inversion recovery; IQR = interquartile range; mRS = modified Rankin Scale; PSWCs = periodic sharp-wave complexes;WBC = white blood cell count. a Three patients with faciobrachial dystonic seizure. b T2/FLAIR. eTable 2 lists the comparison between treatment-responsive and nonresponsive diagnoses. eTable 3 lists the comparison between AE andCreutzfeldt-Jakob disease. c p Values showing significance (p < 0.05) between AE patients and other. 25%) were the most common movement disorders. Brain MRIwas abnormal in 6 of 8 patients (75%), showing perivascularenhancement (2/8; 25%), mesiotemporal hyperintensities (5/8;63%), and pachymeningeal enhancement (1/8; 13%). CSFpleocytosis was observed in all patients. A good outcome (mRSscore ≤2 at 1 year after immunotherapy) was observed in 6 of 8patients (75%), including 1 patient who was treated withimmunotherapies for more than 1 year after symptom onset. ## Discussion In this prospective multicenter cohort study, we determined thefrequency of AE in patients presenting with RPD and comprehensively characterized the phenotypes of AE-RPD, comparedwith other causes of RPD. AE was the most common treatment responsive cause of RPD, accounting for approximately 40% ofdiagnoses, being higher than previously reported.1,6,7,9,21 Amongpatients with AE-RPD, anti-LGI1, SN-AE, and autoimmuneGFAP astrocytopathy were the 3 most common AE subtypes,altogether representing approximately 65% of patients with AERPD. We applied an extensive autoantibody testing methodology,as paired serum and CSF were tested in 99% of patients usingmultiple techniques, resulting in optimal autoantibody detection.30,31 Our findings emphasize that AE should be consideredin patients presenting with RPD; targeted ancillary testing shouldbe performed dependent on specific patient characteristics. The most apparent early clinical difference between AE andother causes of RPD was the high occurrence of seizures(34%) in patients with AE-RPD.15,24,26 Anti-LGI1 accounted ![Figure]() AMPAR = AMPA receptor; CASPR2 = contactin-associated protein-like 2; GFAP = glial fibrillary acidic protein; GlyR = glycine receptor; LGI1 = leucine-rich,glioma-inactivated 1; NMDAR = NMDA receptor; PNS = paraneoplastic neurologic syndrome; SN-AE = seronegative autoimmune encephalitis. *The categoriesincluded both treatment-responsive and nonresponsive diagnoses. Specific diagnoses per diagnostic subcategory are provided in eTable 1. Neurology | Volume 106, Number 11 | June 9, 2026 e214933(6) ### Page 7 Specific diagnoses (n = 147) | Treatment-responsive (n = 95; 65%) | Nonresponsive (n = 52; 35%) | | - | - | | Autoimmune encephalitisa 58 (61%)LGI1 19 (33%) | Neurodegenerative disease 20 (39%)Alzheimer disease 7 (35%) | | SN-AE 11 (19%) | Dementia with Lewy bodies 6 (30%)Frontotemporal dementia 4 (20%)Progressive supranuclear palsy 3 (15%) | | GFAP 8 (14%) | | NMDAR 7 (12%) | | GABAB-R 4 (7%) | | PNSb 3 (5%) | | CASPR2 2 (3%) | | AMPAR 2 (3%) | | GABAA-R 1 (2%) | | GlyR 1 (2%) | | Inflammatory CNS disorder 13 (14%) | Creutzfeldt-Jakob disease 17 (33%) | | Probable neuroinflammatory disorder 9 (69%)Hashimoto encephalopathy 2 (15%) | | PACNS 1 (8%) | | Anti–TNF-induced demyelination 1 (8%) | | Metabolic/toxic/delirium 9 (10%) | | Delirium with underlying systemic infection 3 (33%)Hyponatremia 2 (22%) | | Drug-induced cognitive impairment 3 (33%)Hyperthyroidism 1 (11%) | | Primary psychiatric disorder 4 (4%)Depressive disorder 2 (50%)Psychotic disorder 1 (25%)Anxiety disorder 1 (25%) | | Brain tumor 4 (4%)PCNSL 4 (100%) | Brain tumor 1 (2%) | | Glioma WHO grade 4 1 (100%) | | CNS infection 3 (3%)Neurosyphilis 2 (67%)Neuroborreliosis 1 (33%) | | Epilepsy with a noninflammatory cause 2 (2%) | | Vascular 1 (1%) | Vascular 8 (15%) | | Dural arteriovenous fistula 1 (100%) | Vascular dementia 8 (100%) | | Unknownc 1 (1%) | Unknownd 6 (12%) | Abbreviations: AMPAR = AMPA receptor; Anti-TNF = anti–tumor necrosis factor; CASPR2 = contactin-associated protein-like 2; GABAA-R = GABAA-receptor;GABAB-R = GABAB-receptor; GFAP = glial fibrillary acidic protein; GlyR = glycine receptor; LGI1 = leucine-rich, glioma-inactivated 1; NMDAR = NMDA receptor;PACNS = primary angiitis of the CNS; PCNSL = primary CNS lymphoma; SN-AE = seronegative autoimmune encephalitis; WHO = World Health Organization.a Autoantibody studies included testing for anti-IgLON5 antibodies by cell-based assays and immunohistochemistry. b Ma2 (1), KLHL11 (1), seronegative PNS (1). c Patient with a high suspicion of PCNSL with enhancement on brain MRI and mild pleocytosis deceased before brain biopsy. d All patients with a high suspicion of a neurodegenerative disease with atrophy on brain MRI and absence of signs indicative of a treatment-responsivedisorder. for almost one-third of AE-RPD cases and almost one-half ofcases in patients with seizures at presentation. This observationemphasizes the importance of recognizing seizures in patientswith RPD.9,26 Seizures are also recognized in patients with RPDattributed to neurodegenerative diseases (;10%) but usuallyoccurred later in the illness course.40 Although seizures areusually easily recognized, we identified 2 important diagnosticpitfalls. First, around half of patients presented solely with focalseizures. Focal seizures may be easily missed or misdiagnosedbecause of the lower frequency of motor signs (absent ina substantial proportion of patients) and possible manifestationwith subtle signs, such as piloerection.41 In particular, frequent(>5 per day), stereotyped episodes should raise suspicion foranti-LGI1 encephalitis and prompt autoantibody testing, which can be reliably performed in serum because of its high diagnostic accuracy.24,26 Second, approximately 25% of patientsonly developed seizures after their initial presentation andaround 20% even more than 6 months after onset of symptoms.We recommend proactive clinical monitoring for potentialseizures in patients with RPD, even if absent during the firstevaluation. Early identification of patients with AE-RPD is particularlychallenging if seizures are absent because other symptoms(e.g., psychosis and focal deficits) are usually less specific. Autoimmune GFAP astrocytopathy was the most commonsubtype in seizure-negative AE-RPD patients, of whom most Neurology | Volume 106, Number 11 | June 9, 2026 e214933(7) ### Page 8 ![Figure]() AE = autoimmune encephalitis; AMPAR = AMPA receptor; CASPR2 = contactin-associated protein-like 2; CJD = Creutzfeldt-Jakob disease; GFAP = glial fibrillaryacidic protein; GlyR = glycine receptor; Infl. = inflammatory CNS disorder; LGI1 = leucine-rich, glioma-inactivated 1; ND = neurodegenerative disease; NMDAR =NMDA receptor; presenting phenotype = symptoms at first presentation; Psych. = psychiatric disorder; SN-AE = seronegative autoimmune encephalitis;Vasc. = vascular. (;90%) had striking additional (i.e., noncognitive) signs. Thepsychiatric phenotype was the most common presentation ofautoimmune GFAP astrocytopathy, representing a clearlydefined clinical syndrome characterized by severe psychoticsymptoms and rapid cognitive decline. In previous descriptions of autoimmune GFAP astrocytopathy, cognitiveand psychiatric symptoms were reported in 10%–35% ofpatients.27,42 In RPD accompanied by movement disorders,autoimmune GFAP astrocytopathy was the most commonsubtype, with myoclonus and tremor as most common ![Figure]() AE = autoimmune encephalitis; CJD = Creutzfeldt-Jakob disease. Last follow-up or death without the development ofmovement disorder was treated as censoring events and isindicated by plus signs (+). Neurology | Volume 106, Number 11 | June 9, 2026 e214933(8) ### Page 9 ![Figure]() (A) Autoimmune Encephalitis subtypes in seizure-positive vs seizure-negative patients (at first presentation and during complete period of diagnosticevaluation). (B) Time from symptom onset to seizures in seizure-positive rapidly progressive dementia due to autoimmune encephalitis. Presentingphenotype = symptoms at first presentation. Established phenotype = symptoms during complete period of diagnostic evaluation. AMPAR = AMPA receptor;CASPR2 = contactin-associated protein-like 2; GFAP = glial fibrillary acidic protein; GlyR = glycine receptor; LGI1 = leucine-rich, glioma-inactivated 1; NMDAR =NMDA receptor; PNS = paraneoplastic neurologic syndrome; SN-AE = seronegative autoimmune encephalitis. *Anti-GABAA/B receptor encephalitis. movement disorders. We observed that movement disordersare typically an early symptom of AE (i.e., within 3 months),whereas in CJD, these symptoms generally appear at later stages.43 MRI and CSF findings were mostly abnormal inautoimmune GFAP astrocytopathy, facilitating early detection. A striking radiologic feature in patients with Neurology | Volume 106, Number 11 | June 9, 2026 e214933(9) ### Page 10 | mRS | mRS1 yafterITX | | - | - | | Sex; | score | Time todementia(wk) | Time totreatment(wk) | | Patient | age | at | | no. | (range) | Subphenotype | Clinical | presentation | nadir | Ancillary testing | Tumor | ITX | | 1 | M; | Psychiatric | Cognitive decline, visualhallucinations, delusions,tremor, weight loss | 3 | 43 | MRI + C: perivascularenhancement, bilateralmesiotemporalhyperintensities | No | IVMP, IVIG,oralprednisone,rituximab | 67 | 1 | | 50–60 | | CSF: L 16/mm3, OCB+ | | 2 | V; | Psychiatric | Cognitive decline, visualhallucinations,depressive symptoms | 3 | 30 | MRI + C: bilateralmesiotemporalhyperintensities | No | IVMP, IVIG,oralprednisone,rituximab | 31 | 3 | | 60–70 | | CSF: L 35/mm3, OCB NA | | 3 | M; | Psychiatric | Cognitive decline,delusions, obsessivebehavior, weight loss,gait disorder | 3 34 | MRI + C: generalizedatrophy | No | IVMP, IVIG,oralprednisone | 35 | 3 | | 70–80 | | CSF: L 74/mm3, OCB+ | | 4 | M; | Cerebellar | Cognitive decline, weightloss, cerebellar ataxia,visual hallucinations,myoclonus | 4 | 8 | MRI + C: bilateralmesiotemporalhyperintensities | No | IVMP, IVIG,oralprednisone | 9 | 1 | | 60–70 | ataxia | | CSF: L 151/mm3, OCBNA | | 5 | M; | Cerebellar | Cognitive decline,cerebellar ataxia, weightloss | 3 14 | MRI + C: bilateralmesiotemporal andcortical hyperintensitiesCSF: 41, OCB NA | No | IVMP, IVIG,oralprednisone | 22 | 1 | | 60–70 | ataxia | | 6 | V; | Movement | Cognitive decline, tremor | 3 | 19 | MRI + C: perivascularenhancement,unilateralmesiotemporalhyperintensity | No | IVMP, IVIG,oralprednisone,MMF, AZA | 23 | 1 | | 60–70 | disorder | | CSF: L 49/mm3, OCB+ | | 7 | M; | Movement | Cognitive decline,myoclonus, fever,impaired consciousness | 5 1 | MRI + C:pachymeningealenhancement | No | IVIG, oralprednisone | 2 | 2 | | 40–50 | disorder | | CSF: L 593/mm3, OCB− | | 8 | M; | Pure cognitive | Cognitive decline | 3 | 3 | MRI + C: small vesseldisease | No | IVMP, IVIG,oralprednisone | 1 | 2 | | 80–90 | | CSF: L 207/mm3, OCB− | Abbreviations: AZA = azathioprine; GFAP = glial fibrillary acidic protein; ITX = immunotherapy; IVIG = IV immunoglobulin; IVMP = IV methylprednisone; L =leukocytes; MMF = mycophenolate mofetil; MRI + C = MRI with contrast; mRS = modified Rankin Scale; NA = not applicable; OCB = unique oligoclonal band. autoimmune GFAP astrocytopathy in this study was the highoccurrence (;60%) of mesiotemporal T2/FLAIR hyperintensities on brain MRI. Although the involvement of thetemporal pole has been reported in some patients withautoimmune GFAP astrocytopathy,42 a distinct radiologicpattern of limbic encephalitis has not been described earlier,highlighting the expanding clinicoradiologic profile of thisAE-RPD subtype. In accordance with previous research,particularly the presence of a CSF pleocytosis was highlysensitive.27,42 Identification of autoimmune GFAP astrocytopathy in RPD is crucial because immunotherapy resulted infavorable outcomes in 75% of patients, making this AE subtype a highly treatment-responsive cause of RPD. It is important to note that GFAP antibodies are not included withinall commercial test panels. Therefore, it is strongly recommended to test specifically for GFAP antibodies in seizurenegative RPD patients, especially when accompanied bypsychotic symptoms, early-onset movement disorders, or cerebellar ataxia in combination with CSF pleocytosis. GFAPantibodies should be tested in CSF and confirmed by tissueanalysis, in view of limited sensitivity and specificity in serum.27 SN-AE was a relatively common cause of RPD in our series,with 11 of 147 patients of RPD (7%) meeting the 2016 clinicalcriteria,18,26 which maintains high specificity when appliedrigorously.26 Early identification of patients with RPD attributed to SN-AE is challenging, given that approximately75% of patients presented with isolated cognitive symptomsand specific biomarkers are not available. Seizures were relatively rare and occurred at a later stage (i.e., after 6 months),consistent with earlier reports.44 However, higher frequenciesof seizures (;80%) are also reported in SN-AE,15 possiblyexplained by differences in study selection criteria. Seizurenegative SN-AE might be more prone to diagnostic delays andsubsequent development of severe cognitive decline. Thiscould help explain the relatively high number of patients with Neurology | Volume 106, Number 11 | June 9, 2026 e214933(10) ### Page 11 SN-AE in our study because these patients were possibly morefrequently suspected of having RPD and referred to memoryclinics. In 2023, a large cohort study showed that the outcomeof SN-AE was worse compared with anti-NMDAR encephalitis and delay of immunotherapy (≥1 month) was associatedwith an unfavorable outcome,17 emphasizing the importanceof early identification. We demonstrate that SN-AE is animportant differential diagnostic consideration in patientswith RPD—especially patients with isolated cognitivesymptoms. However, it is important to emphasize that SN-AErepresents a heterogeneous group of diseases, with significantvariation in clinical phenotypes and likely diverse underlyingpathogenic mechanisms.45 Further research is needed toidentify specific biomarkers to improve early identification ofSN-AE and subcategorize within this heterogeneous group ofdiseases.45 Several patients with treatment-responsive causes of RPD,mostly AE, showed markedly elevated CSF t-tau levels andincreased t-tau/p-tau ratios—findings that are generallyconsidered specific for CJD.15,46 It is important to note that allthese patients exhibited clinical or radiologic signs that wereatypical for CJD. Tau levels most likely reflect the severity ofsynaptic dysfunction in these patients, rather than a specificetiology.47 These findings underscore the need for accuratediagnostic tests and emphasize the essential role of criticallyinterpreting test results in relation to the clinical presentation. Neuronal antibodies studies in this study included testing foranti-IgLON5 autoantibodies, as cognitive decline can be thepresenting symptom of this AE subtype.48 Notably, nopatients with anti-IgLON5 disease were identified in ourstudy, which is probably explained by the fact that the cognitive symptoms in anti-IgLON5 disease generally followa more insidious course,13 complicating early diagnosis indementia patients. In patients with a slowly progressive courseof cognitive decline, testing for anti-IgLON5 autoantibodiesshould specifically be considered if additional signs are present, in particular movement disorders, sleep disorder, orbulbar symptoms.13,48 The EMC is the national referral center for neuronal autoantibody testing and AE in the Netherlands, which likelycontributed to the high proportion of patients with AE-RPDin our cohort vs others.6,21 However, comparable frequencies(;30%) of AE-RPD have been described previously incohorts from the United States and Brazil.1,7,9 Collectively,these findings affirm the relevancy of this group of diseases inpatients undergoing evaluation for RPD. Selection bias toward atypical CJD cases probably occurred because typicalCJD cases might have never been referred nor discussed withus. The lower sensitivity of the RT-QuIC in our study compared with previous research may reflect this bias.37,43Therefore, our findings should be interpreted with cautionbecause they may not be directly extrapolated to other settings (e.g., specialized prion centers and lower and middleincome countries), which strongly influences the prevalence of RPD causes. This is illustrated by previous research: neuroinfections were the leading cause of RPD in a large Indiancohort (39%),49 whereas neurodegenerative diseases and CJDwere more frequent in another RPD cohort.6 However, in thismulticenter study, patients were enrolled in many differentsettings, both specialized and nonspecialized, contributing tothe heterogeneity of the cohort. Specific paraclinical features that could aid in diagnosingneurodegenerative diseases were not included because ourfocus was on AE-RPD. For example, substantial brain atrophyon MRI has been linked to ongoing decline in suspected RPDcases,9 whereas CSF biomarkers other than t-tau and p-tau(e.g., amyloid-β 42/40 ratio, GFAP, and neurofilament light)may provide valuable diagnostic insight into patients withRPD.50 Furthermore, neuropsychological assessments werenot included in this study because they were only conductedin a minority of patients, primarily due to the rapidity ofdecline and severity of cognitive deficits, which limits interpretation of these tests in patients with RPD. Further exploration and comparison of profiles of cognitive dysfunctionacross different causes of RPD would be valuable for futureresearch. In summary, AE is the most common treatment-responsivecause of RPD and anti-LGI1 encephalitis is the most commonsubtype, in particular in patients with RPD and early seizures.Comprehensive testing for disease-associated autoantibodiesis strongly recommended in all patients with RPD and shouldinclude testing for anti-GFAP antibodies in CSF. ### Acknowledgment The authors thank all patients and their caregivers for theirparticipation. ### Author Contributions R.W. van Steenhoven: drafting/revision of the manuscript forcontent, including medical writing for content; major role inthe acquisition of data; study concept or design; analysis orinterpretation of data. A.E.M. Bastiaansen: drafting/revisionof the manuscript for content, including medical writing forcontent; study concept or design. J. Kerstens: drafting/revision of the manuscript for content, including medicalwriting for content. Y.S. Crijnen: drafting/revision of themanuscript for content, including medical writing for content.T. Brand: drafting/revision of the manuscript for content,including medical writing for content. J. Brenner: drafting/revision of the manuscript for content, including medicalwriting for content. T.M. Bienfait: drafting/revision of themanuscript for content, including medical writing for content.J.M. de Vries: drafting/revision of the manuscript for content,including medical writing for content. Y.D. Piura: drafting/revision of the manuscript for content, including medicalwriting for content. N. Satyadev: drafting/revision of themanuscript for content, including medical writing for content.M.A.A.M. De Bruijn: drafting/revision of the manuscript forcontent, including medical writing for content. M.M.P. Neurology | Volume 106, Number 11 | June 9, 2026 e214933(11) ### Page 12 Nagtzaam: drafting/revision of the manuscript for content,including medical writing for content. E.E. Turgeon: drafting/revision of the manuscript for content, including medicalwriting for content. S.C. Franken: drafting/revision of themanuscript for content, including medical writing for content. S. Veenbergen: drafting/revision of the manuscript for content, including medical writing for content. R.K. Balvers:drafting/revision of the manuscript for content, includingmedical writing for content. M. Langedijk: drafting/revisionof the manuscript for content, including medical writing forcontent. R.M. Verdijk: drafting/revision of the manuscript forcontent, including medical writing for content. M.M. Verbeek: drafting/revision of the manuscript for content, including medical writing for content. C.E. Teunissen: drafting/revision of the manuscript for content, including medicalwriting for content. Y. Pijnenburg: drafting/revision of themanuscript for content, including medical writing for content.F.J. De Jong: drafting/revision of the manuscript for content,including medical writing for content. A.J. Rozemuller:drafting/revision of the manuscript for content, includingmedical writing for content. N. Fransen: drafting/revision ofthe manuscript for content, including medical writing forcontent. E.G. Dopper: drafting/revision of the manuscript forcontent, including medical writing for content. P.A.E. SillevisSmitt: drafting/revision of the manuscript for content, including medical writing for content. G.S. Day: drafting/revision of the manuscript for content, including medicalwriting for content. H. Seelaar: drafting/revision of themanuscript for content, including medical writing for content.M.J. Titulaer: drafting/revision of the manuscript for content,including medical writing for content; major role in the acquisition of data; study concept or design; analysis or interpretation of data. ### Study Funding This study was funded by ZonMw (Memorabel Program;733050508 and VIMP Program; 07330462210002), theErasmus MC Foundation, and the ItsME Foundation. ### Disclosure R.W. van Steenhoven has been funded by ZonMw(PARADE-VIMP program) and Alzheimer Nederland (InterAct Grant) and has received an Autoimmune EncephalitisAlliance Seed Grant (2024) and a Van Leersum Travel Grantfrom the Royal Dutch Academy of Science (2024). A.E.M.Bastiaansen, J. Kerstens, Y. Crijnen, T. Brand, T.M. Bienfait,J.M. de Vries, Y.D. Piura, N. Satyadev, M.A.A.M. de Bruijn,M.M.P. Nagtzaam, E.E. Erdag, S.C. Franken, S. Veenbergen, R. Balvers, M.J.H. Langedijk, R.M. Verdij, M.M. Verbeek, C.Teunissen, Y.A.L. Pijnenburg, F.J. de Jong, A. Rozemuller,N.L. Fransen, E.G.P. Dopper, and H Seelaar report no disclosures. P.A.E. Sillevis Smitt, J.M. de Vries, and M.J. Titulaerof this publication are members of the European ReferenceNetwork for Rare Immunodeficiency, Autoinflammatory andAutoimmune Diseases, Project ID No. 739543 (ERN-RITA;HCP Erasmus MC). H. Seelaar, E.G.P. Dopper, and F.J. deJong of this publication are members of the European Reference Network for Rare Neurological Diseases (ProjectID No. 739510). P.A.E. Sillevis Smitt holds a patent for thedetection of anti-DNER and received research support fromEuroimmun. M.M. Verbeek is supported by research grantsfrom ParkinsonNL (project P2021-18), the Dutch DigestiveFoundation (project WOO 21-05), Stichting Woelse Waardand Stichting Alkemade-Keuls, ZonMW - Dementia program(10510032120006 and 10510032120003), and AlzheimerNederland (WE032023-11). G.S. Day is supported byNIH (R01AG089380, U01AG057195, U01NS120901,U19AG032438, P30AG062677); serves as a consultant forArialys Therapeutics; serves as a topic editor (Dementia) forDynaMed (EBSCO); is a co-project PI for a clinical trial inanti-NMDAR encephalitis, which receives support fromNIH/National Institute of Neurological Disorders and Stroke(U01NS120901) and Amgen Pharmaceuticals; has developededucational materials for Continuing Education Inc. and IonisPharmaceuticals; and owns stock in ANI Pharmaceuticals.G.S. Day’s institution has received in-kind contributions forradiotracer precursors for tau-PET neuroimaging in studies ofmemory and aging (through Avid Radiopharmaceuticals,a wholly owned subsidiary of Eli Lilly). M.J. Titulaer is a coprincipal investigator of the ExTINGUISH study; is a steeringcommittee member of LEGIONE; has filed a patent on behalfof the Erasmus MC for methods for typing neurologic disorders and cancer, and devices for use therein; has receivedresearch funds for serving on scientific advisory boards forHorizon Therapeutics/Amgen and Argenx, for consultationat Guidepoint Global LLC, and for consultation at UCB;has received an unrestricted research grant from Euroimmune AG and CSL Behring; has received funding fromEpilepsieNL; and receives royalties from UpToDate. Go to for full disclosures. ### Publication History Received by Neurology® November 21, 2025. Accepted in final formFebruary 20, 2026. Submitted and externally peer reviewed. Thehandling editor was Associate Editor Linda Hershey, MD, PhD, FAAN. ## References - 1. Satyadev N, Tipton PW, Martens Y, et al. Improving early recognition of treatmentresponsive causes of rapidly progressive dementia: the STAM(3) P score. Ann Neurol.2024;95(2):237-248. doi:10.1002/ana.26812 - 2. Day GS, Musiek ES, Morris JC. Rapidly progressive dementia in the outpatient clinic:more than prions. Alzheimer Dis Assoc Disord. 2018;32(4):291-297. doi:10.1097/WAD.0000000000000276 - 3. Satyadev N, Piura YD, Tipton PW, et al. Standardizing “rapid”: applying the clinicaldementia rating to define rapidly progressive dementia. Neurology. 2026;106(1):e214439. doi:10.1212/WNL.0000000000214439 - 4. Geschwind MD, Shu H, Haman A, Sejvar JJ, Miller BL. Rapidly progressive dementia.Ann Neurol. 2008;64(1):97-108. doi:10.1002/ana.21430 - 5. Papageorgiou SG, Kontaxis T, Bonakis A, Karahalios G, Kalfakis N, VassilopoulosD. Rapidly progressive dementia: causes found in a Greek tertiary referral center inAthens. Alzheimer Dis Assoc Disord. 2009;23(4):337-346. doi:10.1097/WAD.0b013e31819e099b - 6. Stamatelos P, Kontokostas K, Liantinioti C, et al. Evolving causes of rapidly progressive dementia: a 5-year comparative study. Alzheimer Dis Assoc Disord. 2021;35(4):315-320. doi:10.1097/WAD.0000000000000472 - 7. Studart Neto A, Soares Neto HR, Simabukuro MM, et al. Rapidly progressive dementia:prevalence and causes in a neurologic unit of a tertiary hospital in Brazil. Alzheimer DisAssoc Disord. 2017;31(3):239-243. doi:10.1097/WAD.0000000000000170 - 8. Day GS. Rapidly progressive dementia. Continuum (Minneap Minn). 2022;28(3):901-936. doi:10.1212/CON.0000000000001089 - 9. Piura YD, Satyadev N, Corriveau-Lecavalier N, et al. Decoding the clinical featuresthat associate with progression, causes, and outcomes in patients with suspected Neurology | Volume 106, Number 11 | June 9, 2026 e214933(12) ### Page 13 rapidly progressive dementia. Ann Neurol. 2025;98(4):764-776. doi:10.1002/ana.27297 - 10. Dalmau J, Gleichman AJ, Hughes EG, et al. Anti-NMDA-receptor encephalitis: caseseries and analysis of the effects of antibodies. Lancet Neurol. 2008;7(12):1091-1098.doi:10.1016/S1474-4422(08)70224-2 - 11. Maat P, de Beukelaar JW, Jansen C, et al. Pathologically confirmed autoimmuneencephalitis in suspected Creutzfeldt-Jakob disease. Neurol Neuroimmunol Neuroinflamm. 2015;2(6):e178. doi:10.1212/NXI.0000000000000178 - 12. Grau-Rivera O, S´anchez-Valle R, Saiz A, et al. Determination of neuronal antibodies insuspected and definite Creutzfeldt-Jakob disease. JAMA Neurol. 2014;71(1):74-78.doi:10.1001/jamaneurol.2013.4857 - 13. Bastiaansen AEM, van Steenhoven RW, Te Vaarwerk ES, et al. Antibodies associatedwith autoimmune encephalitis in patients with presumed neurodegenerative dementia. Neurol Neuroimmunol Neuroinflamm. 2023;10(5):e200137. doi:10.1212/NXI.0000000000200137 - 14. Lazar EB, Porter AL, Prusinski CC, et al. Improving early recognition of CreutzfeldtJakob disease mimics. Neurol Clin Pract. 2022;12(6):406-413. doi:10.1212/CPJ.0000000000200097 - 15. Bastiaansen AEM, van Steenhoven RW, de Bruijn MAAM, et al. Autoimmune encephalitis resembling dementia syndromes. Neurol Neuroimmunol Neuroinflamm.2021;8(5):e1039. doi:10.1212/NXI.0000000000001039 - 16. Titulaer MJ, McCracken L, Gabilondo I, et al. Treatment and prognostic factors forlong-term outcome in patients with anti-NMDA receptor encephalitis: an observational cohort study. Lancet Neurol. 2013;12(2):157-165. doi:10.1016/S14744422(12)70310-1 - 17. Lee WJ, Lee HS, Kim DY, et al. Seronegative autoimmune encephalitis: clinicalcharacteristics and factors associated with outcomes. Brain. 2022;145(10):3509-3521.doi:10.1093/brain/awac166 - 18. Graus F, Titulaer MJ, Balu R, et al. A clinical approach to diagnosis of autoimmuneencephalitis. Lancet Neurol. 2016;15(4):391-404. doi:10.1016/S1474-4422(15)00401-9 - 19. Bastiaansen AEM, de Bruijn MAAM, Schuller SL, et al. Anti-NMDAR encephalitis in theNetherlands, focusing on late-onset patients and antibody test accuracy. Neurol Neuroimmunol Neuroinflamm. 2022;9(2):e1127. doi:10.1212/NXI.0000000000001127 - 20. Titulaer MJ, McCracken L, Gabilondo I, et al. Late-onset anti-NMDA receptor encephalitis. Neurology. 2013;81(12):1058-1063. doi:10.1212/WNL.0b013e3182a4a49c - 21. Liu X, Sun Y, Zhang X, et al. Prevalence and outcomes of rapidly progressive dementia: a retrospective cohort study in a neurologic unit in China. BMC Geriatr. 2023;23(1):142. doi:10.1186/s12877-023-03841-1 - 22. McKhann GM, Knopman DS, Chertkow H, et al. The diagnosis of dementia due toAlzheimer’s disease: recommendations from the National Institute on AgingAlzheimer’s Association workgroups on diagnostic guidelines for Alzheimer’s disease.Alzheimers Dement. 2011;7(3):263-269. doi:10.1016/j.jalz.2011.03.005 - 23. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserveragreement for the assessment of handicap in stroke patients. Stroke. 1988;19(5):604-607. doi:10.1161/01.str.19.5.604 - 24. van Sonderen A, Thijs RD, Coenders EC, et al. Anti-LGI1 encephalitis: clinicalsyndrome and long-term follow-up. Neurology. 2016;87(14):1449-1456. doi:10.1212/WNL.0000000000003173 - 25. Kelly MJ, Grant E, Murchison AG, et al. Magnetic resonance imaging characteristics ofLGI1-antibody and CASPR2-antibody encephalitis. JAMA Neurol. 2024;81(5):525-533. doi:10.1001/jamaneurol.2024.0126 - 26. Van Steenhoven RW, de Vries JM, Bruijstens AL, et al. Mimics of autoimmuneencephalitis: validation of the 2016 clinical autoimmune encephalitis criteria.Neurol Neuroimmunol Neuroinflamm. 2023;10(6):e200148. doi:10.1212/NXI.0000000000200148 - 27. Flanagan EP, Hinson SR, Lennon VA, et al. Glial fibrillary acidic protein immunoglobulin G as biomarker of autoimmune astrocytopathy: analysis of 102 patients. AnnNeurol. 2017;81(2):298-309. doi:10.1002/ana.24881 - 28. Spatola M, Petit-Pedrol M, Simabukuro MM, et al. Investigations in GABAA receptorantibody-associated encephalitis. Neurology. 2017;88(11):1012-1020. doi:10.1212/WNL.0000000000003713 - 29. van Steenhoven RW, Salih S, de Vries JM, et al. Clinical impact and safety of brainbiopsy in unexplained central nervous system disorders: a real-world cohort study.Ann Clin Transl Neurol. 2025;12(4):792-804. doi:10.1002/acn3.70000 - 30. van Coevorden-Hameete MH, Titulaer MJ, Schreurs MWJ, de Graaff E, Sillevis SmittPAE, Hoogenraad CC. Detection and characterization of autoantibodies to neuronalcell-surface antigens in the central nervous system. Front Mol Neurosci. 2016;9:37. doi:10.3389/fnmol.2016.00037 - 31. Kerstens J, Schreurs MWJ, de Vries JM, et al. Autoimmune encephalitis and paraneoplastic neurologic syndromes: a nationwide study on epidemiology and antibodytesting performance. Neurol Neuroimmunol Neuroinflamm. 2024;11(6):e200318. doi:10.1212/NXI.0000000000200318 - 32. Hajj-Ali RA, Singhal AB, Benseler S, Molloy E, Calabrese LH. Primary angiitis of theCNS. Lancet Neurol. 2011;10(6):561-572. doi:10.1016/S1474-4422(11)70081-3 - 33. McKeith IG, Boeve BF, Dickson DW, et al. Diagnosis and management of dementiawith Lewy bodies: fourth consensus report of the DLB Consortium. Neurology. 2017;89(1):88-100. doi:10.1212/WNL.0000000000004058 - 34. Rascovsky K, Hodges JR, Knopman D, et al. Sensitivity of revised diagnostic criteriafor the behavioural variant of frontotemporal dementia. Brain. 2011;134(pt 9):2456-2477. doi:10.1093/brain/awr179 - 35. Gorelick PB, Scuteri A, Black SE, et al. Vascular contributions to cognitive impairmentand dementia: a statement for healthcare professionals from the American HeartAssociation/American Stroke Association. Stroke. 2011;42(9):2672-2713. doi:10.1161/STR.0b013e3182299496 - 36. Jack CR Jr, Andrews JS, Beach TG, et al. Revised criteria for diagnosis and staging ofAlzheimer’s disease: Alzheimer’s Association Workgroup. Alzheimers Dement. 2024;20(8):5143-5169. doi:10.1002/alz.13859 - 37. Hermann P, Appleby B, Brandel JP, et al. Biomarkers and diagnostic guidelines forsporadic Creutzfeldt-Jakob disease. Lancet Neurol. 2021;20(3):235-246. doi:10.1016/S1474-4422(20)30477-4 - 38. Diagnostic and Statistical Manual of Mental Disorders, 5th Edition, Text Revision (DSM5-TR). American Psychiatric Association; 2022. - 39. von Elm E, Altman DG, Egger M, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reportingobservational studies. J Clin Epidemiol. 2008;61(4):344-349. doi:10.1016/j.jclinepi.2007.11.008 - 40. Beagle AJ, Darwish SM, Ranasinghe KG, La AL, Karageorgiou E, Vossel KA. Relativeincidence of seizures and myoclonus in Alzheimer’s disease, dementia with Lewybodies, and frontotemporal dementia. J Alzheimers Dis. 2017;60(1):211-223. doi:10.3233/JAD-170031 - 41. Rocamora R, Becerra JL, Fossas P, et al. Pilomotor seizures: an autonomicsemiology of limbic encephalitis? Seizure. 2014;23(8):670-673. doi:10.1016/j.seizure.2014.04.013 - 42. Gravier-Dumonceau A, Ameli R, Rogemond V, et al. Glial fibrillary acidic proteinautoimmunity: a French cohort study. Neurology. 2022;98(6):e653-e668. doi:10.1212/WNL.0000000000013087 - 43. Shir D, Lazar EB, Graff-Radford J, et al. Analysis of clinical features, diagnostic tests,and biomarkers in patients with suspected Creutzfeldt-Jakob disease, 2014-2021.JAMA Netw Open. 2022;5(8):e2225098. doi:10.1001/jamanetworkopen.2022.25098 - 44. Graus F, Escudero D, Oleaga L, et al. Syndrome and outcome of antibody-negativelimbic encephalitis. Eur J Neurol. 2018;25(8):1011-1016. doi:10.1111/ene.13661 - 45. van Steenhoven RW, Titulaer MJ. Seronegative autoimmune encephalitis: exploringthe unknown. Brain. 2022;145(10):3339-3340. doi:10.1093/brain/awac338 - 46. Skillback T, Ros´en C, Asztely F, Mattsson N, Blennow K, Zetterberg H. Diagnosticperformance of cerebrospinal fluid total tau and phosphorylated tau in CreutzfeldtJakob disease: results from the Swedish Mortality Registry. JAMA Neurol. 2014;71(4):476-483. doi:10.1001/jamaneurol.2013.6455 - 47. Soares C, Bellaver B, Ferreira PCL, et al. CSF total tau as a proxy of synaptic degeneration. Nat Commun. 2025;16(1):8076. doi:10.1038/s41467-025-63545-5 - 48. Gaig C, Compta Y, Heidbreder A, et al. Frequency and characterization of movementdisorders in anti-IgLON5 disease. Neurology. 2021;97(14):e1367-e1381. doi:10.1212/WNL.0000000000012639 - 49. Anuja P, Venugopalan V, Darakhshan N N, Awadh P, Wilson V, Manoj G, Manish M,Vivek L. Rapidly progressive dementia: An eight year (2008-2016) retrospectivestudy. PLoS One. 2018;18(13(1) e0189832). doi:10.1371/journal.pone.0189832.29346380 - 50. Kuchenbecker LA, Tipton PW, Martens Y, et al. Diagnostic utility of cerebrospinalfluid biomarkers in patients with rapidly progressive dementia. Ann Neurol. 2024;95(2):299-313. doi:10.1002/ana.26822 Neurology | Volume 106, Number 11 | June 9, 2026 e214933(13)