To monitor the parameters controlling yellow fever vaccine production,
eight different lots produced without stabilizers were studied. It wa
s found that the freeze-thaw cycle did not have any adverse effect on
virus infectivity and the mean loss in virus titre during lyophilizati
on was 0.51 log10 ml-1. A thermodegradation study after storage of vac
cine at different temperatures showed that the vaccine did not pass th
e accelerated stability test. To stabilize the 17D-204 substrain vacci
ne, four stabilizers with different sugars and amino acids were examin
ed. The optimum time for addition of the stabilizers was found to be d
uring homogenization of infected embryos. An accelerated stability tes
t at 37-degrees-C indicated that stabilizers S1, S2 and S3 kept the va
ccine stable for up to 4 weeks, whereas the vaccine was only stable fo
r up to 3 weeks in stabilizer S4. Stabilizer S3 was found statisticall
y to be the best. Reproducibility in production methodology was establ
ished by preparing and testing more batches of the vaccine using stabi
lizer S3.