The cynomolgus monkey (Macaca fascicularis) non-human primate (NHP) is widely used for filovirus vaccine testing. To use limited BSL-4 resources efficiently and minimize NHP usage, Simon's two-stage design was adapted to screen candidate Ebola virus (EBOV) vaccines in up to six NHPs with two (optimal), three, or four NHPs in Stage 1. Using the optimal design, two NHPs were tested in Stage 1. If neither survived, the candidate was rejected. Otherwise, it was eligible for Stage 2 testing in four NHPs. Candidates advanced if four or more NHPs were protected over both stages. An 80% efficacious candidate vaccine had 88.5% probability of advancing, and a 40% efficacious candidate vaccine had 83% probability of rejection. Simon's two-stage design was used to screen 27 EBOV vaccine candidates in 43 candidate regimens that varied in dose, adjuvant, formulation, or schedule. Of the 30 candidate regimens tested using two NHPs in Stage 1, 15 were rejected, nine were withdrawn, and six were tested in Stage 2. All six tested in Stage 2 qualified to advance in the product development pipeline. Multiple regimens for the EBOV vaccines approved by the European Medicines Agency (EMA) and the US Food and Drug Administration (FDA) in 2019 were tested in this program. This approach may also prove useful for screening Sudan virus (SUDV) and Marburg virus (MARV) vaccine candidates.
Keywords: Ebola virus (EBOV); Marburg virus (MARV); Simon’s two-stage; Sudan virus (SUDV); vaccine screening.