To evaluate the long-term comparative efficacy of inebilizumab (INEB) over N-MOmentum open label period (OLP) vs azathioprine and other immunosuppressants (AZA/IST) and vs historical-placebo (PBO) in participants with neuromyelitis optica spectrum disorder (NMOSD).

Two historical comparator groups (HCGs) of participants who received AZA/IST (N=132) or PBO only (N=106) were derived using data from published NMOSD studies and were used to evaluate the comparative efficacy of INEB (N=208) over the OLP. Hazard ratios (HR) for INEB vs HCGs were estimated using a Cox proportional hazards (PH) regression. Time to NMOSD attack was analyzed using parametric and flexible survival (spline) models that were fit to INEB and HCGs. 

The HR (95% CI) of time to NMOSD attack for the N-MOmentum PBO group compared to historical-PBO groups was 1.15(0.67–1.91); P=0.58. The HR (95% CI) for time to NMOSD attack for INEB vs AZA/IST and PBO groups were 0.29(0.17,0.42); P<0.001 and 0.15(0.10,0.21); P<0.001, respectively. A time-varying spline with two internal knots and normal linear predictor provided the best fit.  At 4 years, the model estimated an attack-free probability (95% CI) of 77%(71,83) for INEB, 36%(27,46) for AZA/IST, and 12%(7,20) for PBO. Results indicate a greater difference in efficacy for INEB vs PBO compared to AZA/IST vs PBO, suggesting a significant reduction in risk of attack for INEB vs both AZA/IST and PBO.

INEB was associated with a statistically significant reduction in risk of an NMOSD attack and provided a long-term attack-free probability over the OLP compared to the relative short-term benefit observed with AZA/IST.