John M. Hickey

Background/Objectives: Recombinant poliovirus (PV) virus-like particle (VLP) antigens mimic the conformation of the surface proteins in native PVs (i.e., serotype-specific D-antigen epitopes). Since they lack genomes and are non-infectious, PV-VLPs offer the promise of a safer, next-generation polio vaccine compared to traditional inactivated (IPV) or attenuated live (OPV) vaccines. Sandwich D-antigen ELISA formats are commonly used to measure the in vitro potency values (relative D-antigen content, DU/mL) of unadjuvanted trivalent IPV antigens. If IPV is formulated with aluminum-salt adjuvants, however, a pretreatment step (i.e., adjuvant dissolution or antigen desorption) is required, which may compromise antigen integrity during sample handling. Methods: This work describes the development of three competitive ELISAs to measure the relative D-antigen content of aluminum-salt adjuvanted PV-VLPs (Types 1, 2, 3) without the need for pretreatment. Results: First, key assay parameters were established, including specificity, accuracy, precision, linearity, limit of quantification, and stability-indication. Next, preformulation characterization studies were performed with these methods including (1) rank-ordering the inherent thermal stability profiles of the PV-VLPs (Types 1 > 3 > 2) in-solution and adsorbed to an aluminum phosphate adjuvant (AdjuPhos™, AP) and (2) determining the effect of formulation variables on the thermal stability profiles of AP-adsorbed PV-VLPs including antimicrobial preservatives (thimerosal, 2-PE) and five different antigens present in pediatric combination vaccines (D, T, wP, Hib, Hep B). Conclusions: The development and application of three competitive D-antigen ELISAs were demonstrated, and future use in formulation and storage stability studies with the AP-adjuvanted, trivalent PV-VLPs (Types 1, 2, 3) is discussed with the long-term goal to develop a stable, efficacious, multi-dose, hexavalent combination vaccine presentation.

During the multi-dose formulation development of recombinant vaccine candidates, protein antigens can be destabilized by antimicrobial preservatives (APs). The degradation mechanisms are often poorly understood since available analytical tools are limited due to low protein concentrations and the presence of adjuvants. In this work, we evaluate different analytical approaches to monitor the structural integrity of HPV16 VLPs adsorbed to Alhydrogel™ (AH) in the presence and absence of APs (i.e., destabilizing m-cresol, MC, or non-destabilizing chlorobutanol, CB) under accelerated conditions (pH 7.4, 50 °C). First, in vitro potency losses displayed only modest correlations with the results from two commonly used methods of protein analysis (SDS-PAGE, DSC). Next, results from two alternative analytical approaches provided a better understanding of physicochemical events occurring under these same conditions: (1) competitive ELISA immunoassays with a panel of mAbs against conformational and linear epitopes on HPV16 VLPs and (2) LC-MS peptide mapping to evaluate the accessibility/redox state of the 12 cysteine residues within each L1 protein comprising the HPV16 VLP (i.e., with 360 L1 proteins per VLP, there are 4320 Cys residues per VLP). These methods expand the limited analytical toolset currently available to characterize AH-adsorbed antigens and provide additional insights into the molecular mechanism(s) of AP-induced destabilization of vaccine antigens.