Abstracts

Distinguishing Features and Outcomes of non-Rasmussen and Rasmussen Autoimmune Epilepsia Partialis Continua

Abstract number : 3.308
Submission category : 4. Clinical Epilepsy / 4D. Prognosis
Year : 2025
Submission ID : 1084
Source : www.aesnet.org
Presentation date : 12/8/2025 12:00:00 AM
Published date :

Authors :
Presenting Author: Andrea Stabile, MD – Mayo Clinic, Rochester MN, USA.

Haidara Kherbek, MD – Mayo Clinic, Rochester MN, USA.
Naveen Paramasivan, MD – Mayo Clinic, Rochester MN, USA.
Jeffrey Britton, MD – Mayo Clinic, Rochester MN, USA.
Kelsey M. Smith, MD – Mayo Clinic, Rochester MN, USA.
Elaine Wirrell, MD – Mayo Clinic, Rochester MN, USA.
Lily C. Wong-Kisiel, MD – Mayo Clinic, Rochester, MN, USA.
Gregory Cascino, MD – Mayo Clinic, Rochester MN, USA.
Andrew McKeon, MD – Mayo Clinic, Rochester MN, USA.
Sean Pittock, MD – Mayo Clinic, Rochester MN, USA.
Divyanshu Dubey, MD – Mayo Clinic, Rochester MN, USA.

Rationale: Epilepsia partialis continua (EPC) arises from diverse etiologies, including autoimmune causes as in Rasmussen Syndrome (RS). We aimed to compare clinical features and epileptic outcomes in autoimmune EPC related and unrelated to RS (nRS).

Methods: Patients with autoimmune EPC seen at Mayo Clinic between 2005-2025 were identified from Mayo Data Explore; EPC was defined per Mameniskiene et al.1 Patients were selected as meeting 2005 RS criteria2, 2016 autoimmune encephalitis (AIE) criteria3, or proposed APE2 autoimmune epilepsy (AE) criteria4.

Results: Twenty patients were identified: 13 females, 7 RS and 13 nRS. The median age at onset was 18 (1-65) years, 11 patients < 19 years. Neural antibodies were detected in 50% of cases. The most common was high-titer GAD65 IgG (n=4), followed by ANNA1 IgG (n=3), CRMP5 IgG (n=2), and KLHL11 IgG (n=1). Other patients were seronegative AE (n=3) and RS (n=7, 5 late-onset). A tumor was present in 20% of cases.

Only 1 nRS patient met criteria for definite AIE, another for possible AIE. In contrast, 10/13 (77%) met definite AE criteria, and 2 for probable AE. In only 1 ANNA1-IgG case AE criteria were unmet; however, the patient was immunotherapy responsive. In 3 nRS cases, EPC occurred at onset, alongside other symptoms in 1. Otherwise, EPC followed other epileptic and non-epileptic manifestations. Motor EPC predominated (RS 86%; nRS 92%). While EPC showed a typically unilateral, multifocal pattern in RS, nRS EPC was more heterogeneous being either limited to 1 body segment (n=4) or rarely bilateral (n=2). As in RS, nRS EPC was often persistent or repetitive. Rarely episodic EPC was also seen in nRS patients. Notably, nRS EPC showed more frequently a progressive course (8/13, 62%), as in all RS cases.

All patients were treated with antiseizure medications (ASMs) and all but 1 received immunotherapy, with a median delay of 11 (0-309) months from onset. Corticosteroids were used in all cases, IV immunoglobulins in 15, and plasma exchange in 8. Second-line agents, used in 9, included rituximab (n=6), cyclophosphamide (n=3), mycophenolate (n=3), tocilizumab (n=1), methotrexate (n=1), azathioprine (n=1). Additional treatments were ketogenic diet (n=3), botulinum toxin (n=5), epilepsy surgery (n=4), neuromodulation (n=4), more frequently in RS. Immunotherapy improved EPC in 89% (17/19) of cases. All RS and 8 nRS patients reported transient reduction in EPC intensity, with resolution in 3 nRS (1 transient, 2 persistent). Two nRS patients had no benefit. At last follow-up, 2 ANNA1-IgG patients died. Seizures and EPC continued in 57% (4/7) of RS cases, while in nRS seizures persisted in 64% (7/11), though EPC was controlled in 55% (6/11).

Conclusions: Autoimmune nRS EPC can be associated with different neural antibodies and can be seen in pediatric patients. Compared to RS, it shows greater heterogeneity and potential progression. Immunotherapy can exert a transient benefit in most of EPC cases, though complete resolution is rare.
1Mameniskiene et al.Epilepsia 2011.2Bien et al.Brain 2005.3Graus et al.Lancet Neurol 2016.4Dubey et al.Epilepsia 2019.

Funding: No

Clinical Epilepsy