Lot-to-lot consistency of live attenuated SA 14-14-2 Japanese encephalitis vaccine manufactured in a good manufacturing practice facility and non-inferiority with respect to an earlier product.

We conducted a four-arm, double-blind, randomized controlled trial among 818 Bangladeshi infants between 10 and 12 months of age to establish equivalence among three lots of live attenuated SA 14-14-2 JE vaccine manufactured by the China National Biotec Group's Chengdu Institute of Biological Products (CDIBP) in a new Good Manufacturing Practice (GMP) facility and to evaluate non-inferiority of the product with a lot of the same vaccine manufactured in CDIBP's original facility. The study took place in two sites in Bangladesh, rural Matlab and Mirpur in urban Dhaka. We collected pre-vaccination (Day 0) and post-vaccination Day 28 (-4 to +14 days) blood samples to assess neutralizing anti-JE virus antibody titers in serum by plaque reduction neutralization tests (PRNT). Seroprotection following vaccination was defined as a PRNT titer ≥1:10 at Day 28 in participants non-immune at baseline. Follow-up for reactogenicity and safety was conducted through home visits at Day 7 and monitoring for serious adverse events through Day 28. Seroprotection rates ranged from 80.2% to 86.3% for all four lots of vaccine. Equivalence of the seroprotection rates between pairs of vaccine lots produced in the new GMP facility was satisfied at the pre-specified 10% margin of the 95% confidence interval (CI) for two of the three pairwise comparisons, but not for the third (-4.3% observed difference with 95% CI of -11.9 to 3.3%). Nevertheless, the aggregate seroprotection rate for all three vaccine lots manufactured in the GMP facility was calculated and found to be within the non-inferiority margin (within 10%) to the vaccine lot produced in the original facility. All four lots of vaccine were safe and well tolerated. These study results should facilitate the use of SA 14-14-2 JE vaccine as a routine component of immunization programs in Asian countries.