Use and Safety of Immunotherapeutic Management of N-Methyl-d-Aspartate Receptor Antibody Encephalitis: A Meta-analysis

Abstract

Importance: Overall, immunotherapy has been shown to improve outcomes and reduce relapses in individuals with N-methyl-d-aspartate receptor (NMDAR) antibody encephalitis (NMDARE); however, the superiority of specific treatments and combinations remains unclear.

Objective: To map the use and safety of immunotherapies in individuals with NMDARE, identify early predictors of poor functional outcome and relapse, evaluate changes in immunotherapy use and disease outcome over the 14 years since first reports of NMDARE, and assess the Anti-NMDAR Encephalitis One-Year Functional Status (NEOS) score.

Data sources: Systematic search in PubMed from inception to January 1, 2019.

Study selection: Published articles including patients with NMDARE with positive NMDAR antibodies and available individual immunotherapy data.

Data extraction and synthesis: Individual patient data on immunotherapies, clinical characteristics at presentation, disease course, and final functional outcome (modified Rankin Scale [mRS] score) were entered into multivariable logistic regression models.

Main outcomes and measures: The planned study outcomes were functional outcome at 12 months from disease onset (good, mRS score of 0 to 2; poor, mRS score greater than 2) and monophasic course (absence of relapse at 24 months or later from onset).

Results: Data from 1550 patients from 652 articles were evaluated. Of these, 1105 of 1508 (73.3%) were female and 707 of 1526 (46.3%) were 18 years or younger at disease onset. Factors at first event that were significantly associated with good functional outcome included adolescent age and first-line treatment with therapeutic apheresis, corticosteroids plus intravenous immunoglobulin (IVIG), or corticosteroids plus IVIG plus therapeutic apheresis. Factors significantly associated with poor functional outcome were age younger than 2 years or age of 65 years or older at onset, intensive care unit admission, extreme delta brush pattern on electroencephalography, lack of immunotherapy within the first 30 days of onset, and maintenance IVIG use for 6 months or more. Factors significantly associated with nonrelapsing disease were rituximab use or maintenance IVIG use for 6 months or more. Adolescent age at onset was significantly associated with relapsing disease. Rituximab use increased from 13.5% (52 of 384; 2007 to 2013) to 28.3% (311 of 1100; 2013 to 2019) (P < .001), concurrent with a falling relapse rate over the same period (22% [12 of 55] in 2008 and earlier; 10.9% [35 of 322] in 2017 and later; P = .006). Modified NEOS score (including 4 of 5 original NEOS items) was associated with probability of poor functional status at 1 year (20.1% [40 of 199] for a score of 0 to 1 points; 43.8% [77 of 176] for a score of 3 to 4 points; P = .05).

Conclusions and relevance: Factors influencing functional outcomes and relapse are different and need to be considered independently in development of evidence-based optimal management guidelines of patients with NMDARE.

Figure 1.. Associations Between Age at Onset and Sex and Modified Rankin Scale (mRS) score

A, Patients with reported age at disease onset are displayed (n = 1517). Of these, sex was reported for 1497 of 1517 patients and tumor status for 1491 of 1517 patients. A total of 1160 of 1517 patients (76.5%) had an age at onset of 30 years or younger. Female sex was most pronounced among teenagers and young adults (age, 13 to 30 years; 634 of 763 [83.1%]), less pronounced in children 12 years and younger (237 of 383 [61.9%]), and reversed in adults 50 years and older (55 of 118 [46.6%]). Overall, 389 of 1524 patients (25.6%) had a tumor (ovarian teratoma or other ovarian tumor in 324 of 1524 [21.3%]), and tumors were more frequent in adults (290 of 802 [36.2%]) than children (93 of 698 [13.3%]; χ² = 102.4; P < .001), and in female patients (347 of 1097 [31.6%]) than male patients (33 of 395 [8.4%]; χ² = 82.9; P < .001). B, Outcome (mRS score at final follow-up) according to age at disease onset in 1269 patients. Patients with onset in infancy (2 years or younger) or older adulthood (65 years or older) tended to have worse outcome.

Figure 2.. Independent Association of Clinical Characteristics and Treatment Factors With Functional Outcome at 12 Months and Relapsing Disease Course at 24 Months and Later

Results from the bootstrapped logistic regression models are displayed. Adjusted odds ratios (ORs), 95% CIs, and P values were derived from the distributions of the regression coefficients over 10 000 folds. CSF indicates cerebrospinal fluid; EEG, electroencephalography; IVIG, intravenous immunoglobulin; MRI, magnetic resonance imaging; mRS, modified Rankin Scale; TA, therapeutic apheresis.

Figure 3.. Changes in Rituximab Use at First N-Methyl-d-Aspartate Receptor Antibody Encephalitis (NMDARE) Event and Changes in Relapse Rate Over Time

Data are displayed over 6 temporal epochs, defined by the year of disease onset, if reported; otherwise, the year of publication was used. A, Proportion of patients receiving rituximab at first event over 6 temporal epochs in the whole cohort (363 of 1484 patients [24.5%]) and in the subset of patients with severe disease (modified Rankin Scale score of 5) at first event (205 of 627 [32.7%]), showing a greater increase in the proportion of rituximab use in patients with severe disease. B, Proportion of patients with reported relapse over 6 temporal epochs (182 of 1380 patients [13.2%]). Error bars represent 95% CIs.

Figure 4.. Association Between Modified Anti-NMDAR Encephalitis One-Year Functional Status (NEOS) Score and 1-Year Functional Status

Probability of good functional status (modified Rankin Scale score of 0 to 2) at 1 year after disease onset according to the modified NEOS score for all patients with available data in the imputed (n = 582) and nonimputed (n = 112) data sets. Error bars represent 95% CIs.