Validation of an enzyme-linked immunosorbent assay for the quantification of human IgG directed against the repeat region of the circumsporozoite protein of the parasite Plasmodium falciparum

Abstract

Background: Several pre-erythrocytic malaria vaccines based on the circumsporozoite protein (CSP) antigen of Plasmodium falciparum are in clinical development. Vaccine immunogenicity is commonly evaluated by the determination of anti-CSP antibody levels using IgG-based assays, but no standard assay is available to allow comparison of the different vaccines.

Methods: The validation of an anti-CSP repeat region enzyme-linked immunosorbent assay (ELISA) is described. This assay is based on the binding of serum antibodies to R32LR, a recombinant protein composed of the repeat region of P. falciparum CSP. In addition to the original recombinant R32LR, an easy to purify recombinant His-tagged R32LR protein has been constructed to be used as solid phase antigen in the assay. Also, hybridoma cell lines have been generated producing human anti-R32LR monoclonal antibodies to be used as a potential inexhaustible source of anti-CSP repeats standard, instead of a reference serum.

Results: The anti-CSP repeats ELISA was shown to be robust, specific and linear within the analytical range, and adequately fulfilled all validation criteria as defined in the ICH guidelines. Furthermore, the coefficient of variation for repeatability and intermediate precision did not exceed 23%. Non-interference was demonstrated for R32LR-binding sera, and the assay was shown to be stable over time.

Conclusions: This ELISA, specific for antibodies directed against the CSP repeat region, can be used as a standard assay for the determination of humoral immunogenicity in the development of any CSP-based P. falciparum malaria vaccine.

Figure 1

Specificity demonstrated by inhibition of a serially diluted sample.

Figure 2

Comparison of heat inactivated samples with non-inactivated samples. A, Deming regression (ultrastructural model) of non-inactivated serum (untreated‘) as a function of inactivated serum samples (inactivated) where variances are considered as equal. B, Plot of the bias - Deming regression represented with ratios (inactivated/untreated) versus inactivated (ultrastructural model) where variances are considered as equal.

Figure 3

Stability of quality control samples.A, Stability of the Low Titer Quality control sample measured in each plate of the ELISA run (with the lower and upper rejection limit). B, Stability of the High Titer Quality control sample measured in each plate of the ELISA run (with the lower and upper rejection limit).

Figure 4

Comparison of His-R32LR with the original R32LR as coating antigen.A, Deming regression (ultrastructural model) of His-R32LR as a function of R32LR where variances are considered as equal. B, Plot of the bias - Deming regression represented with ratios (His-R32LR/R32LR) versus R32LR (ultrastructural model) where variances are considered as equal.

Figure 5

Optical density of serial dilutions of the original standard serum obtained from an exploratory malaria vaccine recipient, compared to the monoclonal MAL2A antibody and the monoclonal MAL1C antibody.