Behaviour of five commercial measles vaccines in an accelerated stability test

Between one and seven lots of five different commercial measles vaccines were investigated in an accelerated stability test conducted at 37 ± 0·5°C. Two of the vaccines were found to be very sensitive to heat since they lost about one log 10 TCID 50 per dose per week at that temperature and should be considered as belonging to the first generation of vaccines. The other three vaccines were more stable having apparent 37°C degradation constants of 0·252, 0·231 and 0·125 log 10 TCID 50 per dose per week respectively as calculated from ‘best fit’ regression lines for loss of titre (expressed in log 10 ) against time. They can be considered as ‘second generation’ vaccines. However, the linearity hypothesis cannot be accepted for two of these vaccines because of a more rapid initial fall in titre. There are two phases in the degradation of the vaccines: an initial rapid fall followed by a slower exponential decrease in titre. If only results after at least two weeks' storage are considered, linear regression lines can be fitted to the data on all three ‘second generation’ vaccines. The degradation constants are then very similar, at around 0·1 log 10 TCID 50 per dose per week, for all three vaccines. On the other hand, the initial losses in titre as calculated by the difference between the initial titre and the zero time intercept of the latter regression lines were different being 0·95, 0·71 and 0·23 log 10 TCID 50 per dose respectively. Therefore, in order to characterize fully the stability properties of a measles vaccine both the magnitude of the initial rapid loss and the degradation constant of the exponential phase of the fall in ritre with time should be specified. If this is done significant differences in the stability of so-called ‘second generation’ vaccines can be detected.