Intravenous immunoglobulin as first-line acute treatment in adults with autoimmune encephalitis caused by antibodies to NMDAR, LGI1 and CASPR2

Abstract

Background and objectives: Corticosteroids or plasma exchange are recommended for acute treatment of autoimmune encephalitis (AE). Intravenous immunoglobulins (IVIG) are suggested as an additional treatment option but data on treatment effect is scarce. The objective of the present study was to investigate the impact of the first-line treatment on the three most common forms of AE, in particular, to evaluate the effect of IVIG therapy in these diseases.

Methods: A total of 1274 patients from the German Network for Autoimmune Encephalitis Research (GENERATE) were analyzed, and 388 patients were included in the study because they had either anti-NMDAR, anti-LGI1 or anti-CASPR2 antibodies and firs-line immunotherapy (ivMP monotherapy, ivMP + IVIG, ivMP + PE or ivMP + IVIG + PE) or no immunotherapy at all. For the statistical analyses, patients were stratified according to antibody type, distinguishing between anti-NMDAR (IgG1) and anti-LGI1 as well as anti-CASPR2 (predominantly IgG4). The primary endpoint was the clinical outcome at discharge, which was assessed using the modified Rankin Scale (mRS). The mRS scores were then compared between the different treatment groups over time, and the factors influencing the reduction in mRS at discharge were analyzed. Furthermore, a specific investigation was conducted to determine the differences in outcomes between patients treated with ivMP + IVIG and ivMP + PE, each split by antibody subtype.

Results: In all treatment groups analyzed, significant improvements were observed at the time of discharge and after 12 months compared to disease onset, regardless of the type of first-line treatment. In untreated patients a significant improvement was not observed. The choice of IVIG or PE as an additional treatment to ivMP for anti-NMDAR encephalitis did not affect the primary outcome. In anti-LGI1 or anti-CASPR2 encephalitis, no influence on the primary outcome was observed when IVIG or PE was administered in addition to ivMP, too. However, a direct comparison of the individual antibody subgroups' mRS reductions, depending on the treatment approach (ivMP + IVIG vs. ivMP + PE), revealed that a more significant mRS reduction was observed with ivMP + PE in anti-NMDAR encephalitis.

Discussion: The retrospective data give evidence that there is no difference in outcome for the use of ivMP + PE over ivMP + IVIG or vice versa in the treatment of encephalitis caused by antibodies against NMDAR, LGI1 or CASPR2. Furthermore, the specific method of plasma exchange, whether plasmapheresis or immunoadsorption, did not affect the mRS at discharge.

Keywords: Autoimmune encephalitis; CASPR2; IVIG; Immunotherapy; LGI-1; NMDA.

Fig. 1

Flow chart of study design. After exclusion, 388 patients were analyzed, divided according to the respective therapy approach and the corresponding antibody. CASPR2 contactin-associated-protein-like-2, FU follow-up, IVIG intravenous immunoglobulins, ivMP intravenous methylprednisolone, LGI1 leucine-rich-glioma-inactivated-1, mRS modified Rankin Scale, NMDAR N-methyl-D-aspartate receptor, PE plasma exchange

Fig. 2

A, B Illustration of the distribution of mRS values of patients with anti-NMDAR encephalitis (A) and anti-LGI1/CASPR2 encephalitis (B) in the respective treatment groups in the course from peak of disease to time of discharge The statistical analysis was conducted using the Mann–Whitney test, with a significance level of p < 0.05 (*), p < 0.01 (**), p < 0.001 (***), and p = 0.000 (****).CASPR2 contactin-associated-protein-like-2, IVIG intravenous immunoglobulins, ivMP intravenous methylprednisolone, LGI1 leucine-rich-glioma-inactivated-1, mRS modified Rankin Scale, NMDAR N-methyl-D-aspartate receptor, PE plasma exchange

Fig. 3

A, B Distribution of the ΔmRS between the peak of disease and time of discharge in the individual treatment approaches in the anti-NMDAR (A) and the anti-LGI1/CASPR2 group (B) including mean with 95% CI (solid line) and median (dashed line). The statistical analysis was conducted using the Kruskal–Wallis test. CI confidence intervall, CASPR2 contactin-associated-protein-like-2, FU follow-up, IVIG intravenous immunoglobulins, ivMP intravenous methylprednisolone, LGI1 leucine-rich-glioma-inactivated-1, mRS modified Rankin Scale, NMDAR N-methyl-D-aspartate receptor, PE plasma exchange

Fig. 4

A, B Improvement in the modified Rankin Scale (mRS) score resulting from the application of individual therapeutic approaches. A Illustrates the ratios between non-responders and those who improved in the mRS from peak of disease to discharge in patients with anti-NMDAR encephalitis. B Illustrates the proportion of non-responders at discharge in patients with anti-LGI1/CASPR2 encephalitis. CASPR2 contactin-associated-protein-like-2, IVIG intravenous immunoglobulins, ivMP intravenous methylprednisolone, LGI1 leucine-rich-glioma-inactivated-1, mRS modified Rankin Scale, NMDAR N-methyl-D-aspartate receptor, PE plasma exchange

Fig. 5

Comparisons of mRS differences within the respective therapy groups ivMP + IVIG and ivMP + PE between the detected antibodies against NMDAR vs. LGI1/CASPR2. The mRS difference is significantly higher in the ivMP + PE group compared to patients with antibodies against LGI1/CASPR2 (p = 0.032) (B). This difference cannot be detected in the ivMP + IVIG group (A). In addition, the subdivision was made according to antibody detection (anti-NMDAR vs. anti-LGI1/CASPR2). The subgroups were compared according to therapy (ivMP + IVIG vs ivMP + PE). There were no differences by therapy neither for anti-NMDAR nor anti-LGI1/CASPR2 (C, D). The Mann–Whitney test was used for the analysis. (p < 0.05 *, p < 0.01 **, p < 0.001 ***, p = 0.000 ****). CASPR2 contactin-associated-protein-like-2, IVIG intravenous immunoglobulins, ivMP intravenous methylprednisolone, LGI1 leucine-rich-glioma-inactivated-1, mRS modified Rankin Scale, NMDAR N-methyl-D-aspartate receptor, PE plasma exchange

Fig. 6

A, B The following illustrations depict mRS values at 12 months post-disease onset, according to the NEOS score, for patients with anti-NMDAR encephalitis who received ivMP + IVIG (6A) and those who received ivMP + PE (6B). In both treatment groups, the outcomes were exclusively favourable, with mRS of 0–2 for NEOS values of 0–1. Even with higher NEOS values of 2–4, the outcome was predominantly favourable, with mRS 0–2. IVIG Intravenous immunoglobulins, ivMP intravenous methylprednisolone, mRS modified Rankin Scale, NEOS anti-NMDAR encephalitis One-Year Functional Status, NMDAR N-methyl-D-aspartate receptor, PE plasma exchange