Prognostic Value of Persistent CSF Antibodies at 12 Months in Anti-NMDAR Encephalitis

Abstract

Background and objectives: Anti-NMDA receptor (NMDAR) encephalitis is defined by the presence of antibodies (Abs) targeting the NMDAR in the CSF. This study aimed to determine the prognostic value of persistent CSF NMDAR-Abs during follow-up.

Methods: This retrospective observational study included patients diagnosed with anti-NMDAR encephalitis in the French Reference Center for Paraneoplastic Neurological Syndromes and Autoimmune Encephalitis and for whom CSF samples were obtained at diagnosis and >4 months of follow-up to evaluate CSF NMDAR-Ab persistence. Because patients were tested for CSF NMDAR-Abs at different time points, samples were stratified into different periods of follow-up (i.e., 12 months was considered for the 9- to 16-month follow-up period).

Results: Among the 501 patients diagnosed with anti-NMDAR encephalitis between January 2007 and June 2020, 89 (17%) were tested between 4 and 120 months for CSF NMDAR-Abs after clinical improvement and included in the study (75/89 women, 84%; median age 20 years, interquartile range [IQR] 16-26). During follow-up, 21 of 89 (23%) patients had a relapse after a median time of 29 months (IQR 18-47), and 20 of 89 (22%) had a poor outcome (mRS ≥3) after a median last follow-up of 36 months (IQR 19-64). Most patients (69/89, 77%) were tested at the 12-month follow-up period, and 42 of 69 (60%) of them had persistent CSF NMDAR-Abs. When comparing patients with persistent or absent CSF NMDAR-Abs at 12 months, poor outcome at the last follow-up was more frequent in the former (38% vs 8%, p = 0.01), who had relapses more often (23% vs 7%), which also appeared earlier in the course of the disease (90% during the following 4 years of follow-up vs 20%), although no significant difference was observed at long-term follow-up (p = 0.15). In addition, patients with persistent CSF NMDAR-Abs at 12 months had higher titers of CSF NMDAR-Abs at diagnosis.

Discussion: In this study, patients with persistent CSF NMDAR-Abs at 12 months were more likely to have subsequent relapses and a poor long-term outcome. However, these findings should be interpreted with caution because of the variability in the time of sampling of this study. Future prospective studies are required to validate these results in larger cohorts.

Figure 1. Flowchart and Graphical Representation of the Study Design

All the patients who fulfilled the inclusion criteria at 6, 12, 24, 36, 60, and >70 months of follow-up were codified according to the presence (green) or absence (red) of CSF NMDAR-Abs. If the patient was not tested, the time point is shown in gray. Patients who had a relapse during the follow-up are depicted in yellow, and the time of relapse was highlighted with a black frame. NMDAR = NMDA receptor; SC = seroconversion.

Figure 2. Kaplan-Meier Analysis of Relapses in Patients With Persistent and Absent CSF NMDAR-Abs at 12-Month Follow-up

No significant difference was observed in terms of relapses at long-term follow-up (p = 0.15) among patients with persistent (green) or absent (blue) CSF NMDAR-Abs at 12 months. However, patients with persistent CSF NMDAR-Abs at 12 months relapsed earlier: 9 of 10 (90%) relapsed between 12 months and 5 years of follow-up, whereas only 1 of 5 (20%) patients with absent CSF NMDAR-Abs at 12 months relapsed during the same time interval. NMDAR-Abs = NMDA receptor antibodies.

Figure 3. Functional Status of Patients With Persistent or Absent CSF NMDAR-Abs at 12-Month Follow-up

Graphical representation of the mRS of patients with persistent (right) or absent (left) CSF NMDAR-Abs at 12-month follow-up. The vertical axis represents the number of patients included (framed) and the time of follow-up: M0, month 0; M3, month 3; M6, month 6; M9, month 9; M12, month 12; M18, month 18; M24, month 24; and M36, month 36. By contrast, the horizontal axis represents the percentage of patients with a determined mRS. mRS = modified Rankin Scale; NMDAR-Abs = NMDA receptor antibodies.