Search

A collaborative study on the evaluation of an alternative functional assay, the Vero cell method, to the Ph. Eur. in vivo challenge procedures for potency determination of diphtheria toxoid in 6 different combined vaccines was initiated in January 2001. The study was an extension of a previous study for the validation of serological methods for potency testing of tetanus toxoid vaccines for human use. To allow interim evaluation of test results and to monitor study progress, the project was divided into three consecutive phases. The results of Phase I and II studies are presented in this report. Pre-validation (Phase I) study, performed in two laboratories, indicated that comparable diphtheria potency estimates were obtained in the Ph. Eur. direct intradermal challenge assay in guinea pigs, in Vero cell assay and in indirect ELISA for five vaccines of different potencies (range of estimates: ca. 20-200 IU/ml). The correlation coefficients between the challenge assay and the Vero cell assay corresponded to those between the challenge assay and ELISA, confirming that the antibodies play an important role in protection and that predominantly protective/neutralising antibodies are present in guinea pigs, at the time point investigated. It was observed, for Vero cell assays, that about 16-35 (9-28 in Phase II study) fold lower titre of individual serum samples were obtained when using equine, rather than guinea pig reference serum. The study also provided preliminary information that sera from the same guinea pigs may be used for potency determination of both diphtheria and tetanus toxoid components of vaccines. In Phase II, another five laboratories analysed a subset of the vaccines included in Phase I study plus an additional vaccine. Four laboratories performed the lethal challenge assay and one laboratory carried out the intradermal challenge assay. All laboratories also performed the Vero cell assay and both ELISA for diphtheria antitoxin and ELISA for tetanus antitoxin. One laboratory also performed the tetanus ToBI assay. The correlation coefficient (r) between Vero cell assay and ELISA for diphtheria antitoxin ranged from 0.76 to 0.91 in the different laboratories. The correlation between diphtheria serological assays and challenge assays were confirmed satisfactory as ca. 90 per cent of serum-estimates lead to correct prediction of mortality. All laboratories had identical rankings of the vaccines in all serological assays and in the valid challenge assays. The ranking order was identical to assumed/provided potency for the highest and the lowest vaccine. Two of the vaccines had an inversion in some assays and laboratories. As these two vaccines have almost identical potencies in all assays, these inversions are not significant. As the vaccine doses were optimised for the diphtheria component, serum anti-tetanus toxoid/toxin activities varied widely between the vaccines, making it questionable to apply a parallel line model to calculate exact potencies. However, the dose levels used showed a clear regression and good linearity in general. DTaP vaccines containing the IPV component did not always meet the present Ph. Eur. requirements in the serological assays. It should be further investigated in the Phase III study if this is a general feature of such combined vaccines. Preliminary investigations on samples from two laboratories indicate that the neutralising activity of type 1, 2 and 3 polioviruses can also be detected, in a dose-dependent way. Further studies are in progress with serum samples from other laboratories. In the light of results obtained in the first two phases, it is recommended to proceed with Phase III study to investigate reliability of the in vitro assays. In Phase III it will also be further investigated whether the serological assays for D and T components are suitable for the control of the multi-component vaccines currently marketed in Europe.

Phase I of BSP034 collaborative study was extended in two laboratories to include correlation of serology with in vivo toxin neutralisation test (TNT) using 2 separate sets of 20 serum pools, produced in-house. The study investigated the extent to which the in vitro methods for diphtheria antibodies, Vero cell assay and diphtheria enzyme-linked immunosorbent assay for diphtheria antitoxin (D-ELISA), can detect neutralising antibodies by comparison with TNT in guinea pigs. The study was also performed to compare the antibody neutralising potency obtained in relation to guinea pig (GP) or equine (DI) antitoxin standard. In addition, the study provided an opportunity to compare ELISA for tetanus antitoxin (T-ELISA) and TNT assay for detection of anti-tetanus antibodies, from the same set of serum pools. The data obtained show that antitoxin potency obtained by Vero cell assay, D-ELISA and T-ELISA using the same GP standard, highly correlated with neutralising potency as determined in respective TNT assays. Vero cell assay with DI provided estimates that also correlated with neutralising potency, but were of significantly lower titre. Since reference to DI standard is widely used in serodiagnosis, as well as in clinical studies where diphtheria antitoxin titres obtained in the Vero cell method are taken as surrogate markers for vaccine efficacy, it should be investigated if a similar difference is also observed for human serology.