Narahari S. Pujar

Pre-clinical evaluation of a 15-valent pneumococcal conjugate vaccine (PCV15-CRM197) in an infant-rhesus monkey immunogenicity model

The incidence of invasive pneumococcal disease (IPD), caused by the approximately 91 serotypes of Streptococcus pneumoniae (PN), varies geographically and temporally as a result of changing epidemiology and vaccination patterns as well as due to regional measurement differences. Prevnar(®) (Pfizer), the first licensed pneumococcal conjugate vaccine (PCV), comprises polysaccharides (PS) from 7 serotypes conjugated to the mutant diphtheria toxin carrier protein, CRM197. In the United States and elsewhere, this vaccine has been highly efficacious in reducing the incidence of IPD caused by vaccine serotypes, however, the incidence of non-vaccine serotypes (e.g., 19A, 22F, and 33F) has increased, resulting in the need for vaccines with higher valencies. In response, 10- and 13-valent PCVs have recently been licensed. To further increase serotype coverage, we have developed a 15-valent PCV containing PS from serotypes 1, 3, 4, 5, 6A, 6B, 7F, 9V, 14, 18C, 19A, 19F, 22F, 23F and 33F conjugated to CRM197 and formulated on aluminum phosphate adjuvant. Vaccine immunogenicity was evaluated in infant rhesus monkeys since they, like human infants, respond poorly to unconjugated PN PS. Infant (2-3 month old) rhesus monkeys were vaccinated three times with PCV-15 or Prevnar(®) at 2 month intervals, and serotype-specific IgG antibodies were measured using a multiarray electrochemiluminescence (ECL) assay. The results indicate that antibody responses to PCV-15 and Prevnar(®) were comparable for the 7 common serotypes and that post-vaccination responses to PCV-15 were >10-fold higher than baseline for the 8 additional serotypes.

Study of detergent-mediated liberation of hepatitis B virus-like particles from S. cerevisiae homogenate: identifying a framework for the design of future-generation lipoprotein vaccine processes.

Virus‐like particles (VLPs) are expressed intracellularly in Saccharomyces cerevisiae and the recovery process involves the use of a detergent, which facilitates the release of VLP from host cell components. The detergent‐mediated liberation of VLPs is a critical step in primary recovery and is responsible for setting the backdrop for subsequent purification in terms of product yield and characteristics of the process stream. In this paper the use of Triton X‐100 detergent for the recovery of lipid envelope VLPs, using the hepatitis B surface antigen (HBsAg) as the VLP model, was investigated. To develop a framework that can be adopted in process design for future generation VLP vaccine candidates, the impact of Triton X‐100 was characterized via different response factors: (i) recovery and activity of the HBsAg; (ii) level of protein and lipid contamination from the host cell; and (iii) indirect impact on the performance of an ultrafiltration step following primary recovery. Our studies identified that an increase in detergent concentration favors recovery of HBsAg only to a specific threshold, 0.5% v/v Triton X‐100. Further increase in detergent results in delipidation of HBsAg leading to loss in antigenic activity. The level of contamination due to host protein and lipid co‐liberation is in proportion with the amount of detergent employed. Greater membrane resistance during ultrafiltration was observed for samples generated using higher concentrations of detergent due to the increase in membrane fouling by the contaminants. Based on this study, Triton X‐100 concentrations in the range of 0.2–0.5% v/v appears to be most suitable for recovery of native HBsAg. Choosing between 0.2–0.5% v/v would involve identifying a suitable tradeoff between desired product yield and the level of contamination that can be tolerated by downstream operations.

Exploiting the intracellular compartmentalization characteristics of the S. cerevisiae host cell for enhancing primary purification of lipid-envelope virus-like particles

This article demonstrates how the intracellular compartmentalization of the S. cerevisiae host cell can be exploited to impart selectivity during the primary purification of lipid‐envelope virus‐like particles (VLPs). The hepatitis B surface antigen (HBsAg) was used as the VLP model in this study. Expressed HBsAg remain localized on the endoplasmic reticulum and the recovery process involves treating cell homogenate with a detergent for HBsAg liberation. In our proposed strategy, a centrifugation step is introduced immediately following cell disruption but prior to the addition of detergent to allow the elimination of bulk cytosolic contaminants in the supernatant, achieving ∼70% reduction of contaminating yeast proteins, lipids, and nucleic acids. Recovery and subsequent treatment of the solids fraction with detergent then releases the HBsAg into a significantly enriched product stream with a yield of ∼80%. The selectivity of this approach is further enhanced by operating under moderate homogenization pressure conditions (∼400 bar). Observed improvements in the recovery of active HBsAg and reduction of contaminating host lipids were attributed to the low‐shear conditions experienced by the HBsAg product and reduced cell fragmentation, which led to lower coextraction of lipids during the detergent step. As a result of the cleaner process stream, the level of product capture during the loading stage of a downstream hydrophobic interaction chromatography stage increased by two‐fold leading to a concomitant increase in the chromatography step yield. The lower level of exposure to contaminants is also expected to improve column integrity and lifespan.

Safety and immunogenicity of 15-valent pneumococcal conjugate vaccine (PCV15) in healthy infants

BACKGROUND Pediatric use of pneumococcal conjugate vaccines (PCV) has been associated with significant decrease in disease burden. However, disease caused by non-vaccine serotypes has increased. Safety and immunogenicity of 15-valent PCV (PCV15) containing serotypes included in 13-valent PCV (PCV13) plus serotypes 22F and 33F were evaluated in infants (NCT01215188). METHODS Infants received adjuvanted PCV15, nonadjuvanted PCV15, or PCV13 at 2, 4, 6, and 12-15 months of age. Safety was monitored for 14 days after each dose. Serotype-specific IgG geometric mean concentrations (GMCs) and opsonophagocytic activity (OPA) geometric mean titers (GMTs) were measured at postdose-3, predose-4, and postdose-4. RESULTS Safety profiles were comparable across vaccination groups. At postdose-3, both PCV15 formulations were non-inferior to PCV13 for 10 of 13 shared serotypes but failed non-inferiority for 3 serotypes (6A, 6B, and 19A) based on proportion of subjects achieving IgG GMC ≥0.35 µg/mL. Adjuvanted PCV15 and nonadjuvanted PCV15 were non-inferior to PCV13 for 11 and 8 shared serotypes, respectively, based on postdose 3 comparisons of GMC ratios. PCV15 induced higher antibodies to serotypes 3, 22F, and 33F than PCV13. CONCLUSIONS PCV15 displayed acceptable safety profile and induced IgG and OPA to all 15 vaccine serotypes at levels comparable to PCV13 for 10 of 13 shared serotypes. Study identification: V114-003. CLINICALTRIALS.GOV identifier: NCT01215188.