Véronique Barban

Preclinical and clinical development of YFV 17D-based chimeric vaccines against dengue, West Nile and Japanese encephalitis viruses

Dengue viruses (DENV), West Nile virus (WNV) and Japanese encephalitis virus (JEV) are major global health and growing medical problems. While a live-attenuated vaccine exists since decades against the prototype flavivirus, yellow fever virus (YFV), there is an urgent need for vaccines against dengue or West Nile diseases, and for improved vaccines against Japanese encephalitis. Live-attenuated chimeric viruses were constructed by replacing the genes coding for Premembrane (prM) and Envelope (E) proteins from YFV 17D vaccine strain with those of heterologous flaviviruses (ChimeriVax technology). This technology has been used to produce vaccine candidates for humans, for construction of a horse vaccine for West Nile fever, and as diagnostic reagents for dengue, Japanese encephalitis, West Nile and St. Louis encephalitis infections. This review focuses on human vaccines and their characterization from the early stages of research through to clinical development. Phenotypic and genetic properties and stability were examined, preclinical evaluation through in vitro or animal models, and clinical testing were carried out. Theoretical environmental concerns linked to the live and genetically modified nature of these vaccines have been carefully addressed. Results of the extensive characterizations are in accordance with the immunogenicity and excellent safety profile of the ChimeriVax-based vaccine candidates, and support their development towards large-scale efficacy trials and registration.

Standardized quantitative RT-PCR assays for quantitation of yellow fever and chimeric yellow fever–dengue vaccines

Yellow fever-dengue chimeras (CYDs) are being developed currently as live tetravalent dengue vaccine candidates. Specific quantitative assays are needed to evaluate the viral load of each serotype in vaccine batches and biological samples. A quantitative real-time RT-PCR (qRT-PCR) system was developed comprising five one-step qRT-PCRs targeting the E/NS1 junction of each chimera, or the NS5 gene in the yellow fever backbone. Each assay was standardized using in vitro transcribed RNA qualified according to its size and purity, and precisely quantified. A non RNA-extracted virus sample was introduced as external quality control (EQC), as well as 2 extraction controls consisting of 2 doses, 40 and 4,000 GEQ (genomic equivalents), of this EQC extracted in parallel to the samples. Between 6 and 10 GEQ/reaction were reproducibly measured with all assays and similar titers were obtained with the two methods when chimeric virus samples were quantified with the E/NS1- or the NS5-specific assays. Reproducibility of RNA extraction was ensured by automation of the process (yield>or=50%), and infectious virus was isolated in >or=80% of PCR-positive sera from immune monkeys.

Broad neutralization of wild-type dengue virus isolates following immunization in monkeys with a tetravalent dengue vaccine based on chimeric Yellow Fever 17D/Dengue viruses

The objective of the study was to evaluate if the antibodies elicited after immunization with a tetravalent dengue vaccine, based on chimeric yellow fever 17D/dengue viruses, can neutralize a large range of dengue viruses (DENV). A panel of 82 DENVs was developed from viruses collected primarily during the last decade in 30 countries and included the four serotypes and the majority of existing genotypes. Viruses were isolated and minimally amplified before evaluation against a tetravalent polyclonal serum generated during vaccine preclinical evaluation in monkey, a model in which protection efficacy of this vaccine has been previously demonstrated (Guirakhoo et al., 2004). Neutralization was observed across all the DENV serotypes, genotypes, geographical origins and isolation years. These data indicate that antibodies elicited after immunization with this dengue vaccine candidate should widely protect against infection with contemporary DENV lineages circulating in endemic countries.

Genetic stability of a dengue vaccine based on chimeric yellow fever/dengue viruses

A tetravalent dengue vaccine based on four live, attenuated, chimeric viruses (CYD1-4), constructed by replacing the genes coding for premembrane (prM) and envelope (E) proteins of the yellow fever (YF)-17D vaccine strain with those of the four serotypes of dengue virus, is in clinical phase III evaluation. We assessed the vaccines genetic stability by fully sequencing each vaccine virus throughout the development and manufacturing process. The four viruses displayed complete genetic stability, with no change from premaster seed lots to bulk lots. When pursuing the virus growth beyond bulk lots, a few genetic variations were observed. Usually both the initial nucleotide and the new one persisted, and mutations appeared after a relatively high number of virus duplication cycles (65-200, depending on position). Variations were concentrated in the prM-E and non-structural (NS)4B regions. PrM-E variations had no impact on lysis-plaque size or neurovirulence in mice. None of the variations located in the YF-17D-derived genes corresponded with reversion to the wild-type Yellow Fever sequence. Variations in NS4B likely reflect virus adaptation to Vero cells growth. A low to undetectable viremia has been reported previously [1-3] in vaccinated non-human and human primates. Combined with the data reported here about the genetic stability of the vaccine strains, the probability of in vivo emergence of mutant viruses appears very low.

Immunogenicity, protective efficacy, and non-replicative status of the HSV-2 vaccine candidate HSV529 in mice and guinea pigs.

HSV-2 vaccine is needed to prevent genital disease, latent infection, and virus transmission. A replication-deficient mutant virus (dl5-29) has demonstrated promising efficacy in animal models of genital herpes. However, the immunogenicity, protective efficacy, and non-replicative status of the highly purified clinical vaccine candidate (HSV529) derived from dl5-29 have not been evaluated. Humoral and cellular immune responses were measured in mice and guinea pigs immunized with HSV529. Protection against acute and recurrent genital herpes, mortality, latent infection, and viral shedding after vaginal HSV-2 infection was determined in mice or in naïve and HSV-1 seropositive guinea pigs. HSV529 replication and pathogenicity were investigated in three sensitive models of virus replication: severe combined immunodeficient (SCID/Beige) mice inoculated by the intramuscular route, suckling mice inoculated by the intracranial route, and vaginally-inoculated guinea pigs. HSV529 immunization induced HSV-2-neutralizing antibody production in mice and guinea pigs. In mice, it induced production of specific HSV-2 antibodies and splenocytes secreting IFNγ or IL-5. Immunization effectively prevented HSV-2 infection in all three animal models by reducing mortality, acute genital disease severity and frequency, and viral shedding. It also reduced ganglionic viral latency and recurrent disease in naïve and HSV-1 seropositive guinea pigs. HSV529 replication/propagation was not detected in the muscles of SCID/Beige mice, in the brains of suckling mice, or in vaginal secretions of inoculated guinea pigs. These results confirm the non-replicative status, as well as its immunogenicity and efficacy in mice and guinea pigs, including HSV-1 seropositive guinea pigs. In mice, HSV529 produced Th1/Th2 characteristic immune response thought to be necessary for an effective vaccine. These results further support the clinical investigation of HSV529 in human subjects as a prophylactic vaccine.

Yellow fever vaccine: comparison of the neurovirulence of new 17D-204 Stamaril™ seed lots and RK 168-73 strain.

The neurovirulence of two new candidate 17D-204 Stamaril™ working seed lots and that of two reference preparations were compared. The Stamaril™ working seed lots have been used for more than twenty years for the manufacturing of vaccines of acceptable safety and efficacy. The preparation designated RK 168-73 and provided by the Robert Koch Institute was used as a reference. It was confirmed that RK 168-73 strain was not a good virus control in our study because it has a very low neurovirulence regarding both the clinical and histopathological scores in comparison with Stamaril™ strain and is not representative of a vaccine known to be satisfactory in use. The results were reinforced by the phenotypic characterization by plaque assay demonstrating that RK 168-73 was very different from the Stamaril™ vaccine, and by sequencing results showing 4 mutations between Stamaril™ and RK 168-73 viruses leading to amino acid differences in the NS4B and envelop proteins.

Evaluation of Chimeric Yellow Fever 17D/Dengue Viral Replication in Ticks

Chimeric yellow fever 17D/DENV-1-4 viruses (CYD-1-4) have been developed as a tetravalent dengue vaccine candidate which is currently being evaluated in efficacy trials in Asia and America. While YF 17D and DENV are mosquito-borne flaviviruses, it has been shown that CYD-1-4 do not replicate after oral infection in mosquitoes and are not transmitted to new hosts. To further document the risk of environmental dissemination of these viruses, we evaluated the replication of CYD-1-4 in ticks, the vector of tick-borne encephalitis virus (TBEV), another member of the flavivirus family. Females of two hard tick species, Ixodes ricinus and Rhipicephalus appendiculatus, were inoculated intracoelomically with CYD-1-4 viruses and parent viruses (DENV-1-4 and YF 17D). Virus persistence and replication was assessed 2, 16, and 44 days post-inoculation by plaque titration and qRT-PCR. CYD-1-4 viruses were detected in I. ricinus ticks at early time points post-inoculation, but with infectious titers at least 100-fold lower than those observed in TBEV-infected ticks. Unlike TBEV, complete viral clearance occurred by day 44 in most ticks except for CYD-2, which had a tendency to decline. In addition, while about 70% of TBEV-infected I. ricinus nymphs acquired infection by co-feeding with infected tick females on non-viremic hosts, no co-feeding transmission of CYD-2 virus was detected. Based on these results, we conclude that the risk of dissemination of the candidate vaccine viruses by tick bite is highly unlikely.