To compare the efficacy and safety of different-course corticosteroids plus mycophenolate mofetil (MMF) as maintenance therapy in autoimmune encephalitis (AE) with neuronal surface antibodies (NSAbs) without tumor and explore the optimal course of corticosteroids.
Fifty-five patients with definite AE without tumor were enrolled consecutively between June 2015 and November 2020 and retrospectively divided three groups according to the course of treatment with corticosteroid, i.e., a group of patients with a course of 3-6 months (Group 3-6mo), 6-12 months (Group 6-12mo), and >12 months (Group >12mo). Demographic data, clinical manifestation and ancillary tests results were recorded. The dosage and courses of corticosteroid treatment, the recovery of neurological function, the occurrence of adverse effects, and relapses were followed up.
A total of 55 patients were included in the final analysis. The numbers of patients in Group 3-6 mo, Group 6-12 mo, and Group >12 mo was 14, 17, and 24, respectively. A significantly higher proportion of patients in Group >12 mo showed a decreased level of consciousness at the onset (12, 50%) than in Group 3-6 mo and Group 6-12 mo (2,14.3%; 3, 17.6%) (p = 0.033). The incidence of MRI abnormalities was significantly higher in Group 6-12 mo and Group >12 mo (10, 58.8%; 16, 66.7%) than in Group 3-6 mo (3, 21.4%) (P=0.023). Ordinal regression analysis indicated that decreased level of consciousness was associated with the course of corticosteroid (OR=3.838, 95% CI: 1.103-13.323, P=0.035). No significant difference was observed between the three groups regarding the cumulative dose of corticosteroids administered during the first three months of long-term treatment (P>0.05). Additionally, no significant difference in the cumulative dosage of corticosteroids was found between patients in Group 6-12 months and Group >12 months during the first 6 months after beginning long-term treatment. The mRS scores of the three groups were not statistically significant before and after first-line treatment or at the last follow-up. Bonferroni multiple comparison test indicated that the mRS scores of patients in Group 6-12 months and Group >12 months were not statistically significant at 3 months and 12 months after the start of long-term treatment. During the follow-up, 50 (90.9%) patients achieved satisfactory neurological function (mRS score ≤2). Five patients (9.1%) experienced a first relapse and 2 of them were overlapped with both anti-NMDA receptor and glial antibodies. The incidence of adverse effects was significantly higher in Group >12 mo (17, 70.8%) than in Group 3-6 mo (3, 21.4%) and Group 6-12 mo (5, 29.4%) (P=0.003).
The beneficial effects of oral corticosteroid treatment may do not persist beyond 12 months and may even contribute to an increased incidence of adverse effects. In order to optimize the effectiveness and safety of treatment, we recommend a corticosteroid course of 3-12 months. Patients with reduced levels of consciousness may be more inclined to choose longer courses of corticosteroids for long-term treatment. Patients with an “overlapping syndrome” may require more intense immunotherapy to prevent relapse.