Our main vaccine platform is based on recombinant vesicular stomatitis virus (rVSVs), a live-attenuate vaccine approach. Over the years we have generated several rVSVs expressing the glycoproteins (GP) of representative isolates of all species of Ebola virus: Sudan ebolavirus (SEBOV), Zaire ebolavirus (ZEBOV), Tai forest ebolavirus (TFEBOV), Bundibugyo ebolavirus (BEBOV) and Reston ebolavirus (REBOV). Additionally, we generated rVSVs expressing the GPs of two isolates of Marburg virus: Lake Victoria marburgvirus isolate Musoke and Angola. All vaccine vectors have been extensively characterized in cell culture and their protective efficacy has been evaluated in animal models (rodents, nonhuman primates) against homologous challenges. In an effort to decipher the mechanism of protection of the rVSV vaccine vectors we used the rVSV/ZEBOVgp as a model. We could demonstrate in nonhuman primates that antibodies specific to the foreign immunogen play a critical role in protection. Recent similar work also confirmed a role of antibodies for the mechanism of protection mediated by the rVSV vaccine vector against MARV. Overall, we postulate that antibodies (total and neutralizing IgG) play a key role for the mechanism of protection for all rVSV-based vaccine candidates. In response to the recent Ebola outbreak in West Africa, the rVSV/ZEBOVgp vaccine candidate was fast-tracked and shown to be safe and immunogenic in humans. Phase III clinical trials with this vaccine candidate were initiated in Guinea, Sierra Leone and Liberia. To support the clinical trials we have shown that the GMP-produced rVSV/ZEBOVgp vaccine lot used in West Africa protects against challenge with a recent local isolate, a proof that had been missing at trial start. A recent preliminary report published in Lancet from the human trial in Guinea reports success of the rVSV/ZEBOVgp vaccine in a ring vaccination approach. This remarkable outcome is supported by another recent study of our group in nonhuman primates looking into the minimum time needed for protection. We could demonstrate complete protection when rVSV/ZEBOVgp was administered at least one week and partial protection when administered as close as three days prior to challenge. Overall, this is a milestone achievement in the development of Ebola countermeasures. Cross-protection among the different Ebola and Marburg virus species is an important consideration, but is thought to be difficult to achieve due to relatively high genetic variability and the general lack of cross-protective antibodies among genera in particular, but also among species within a single genus. In a first attempt to address this issue, we previously used a single-injection protocol with three blended vaccine vectors (rVSV/SEBOVgp, rVSV/ZEBOVgp and rVSV/MARVgp) and demonstrated complete protection against challenge with the three homologous virus species. We have also performed another proof-of-concept study, in which we evaluated cross-protection following immunization with a single vaccine vector (rVSV/ZEBOVgp or rVSV/CIEBOVgp) and demonstrated partial cross-protection against challenge with a heterologous virus species (BEBOV). This demonstrates that monovalent rVSV-based vaccines may be useful against a newly emerging filovirus species; however, heterologous protection across species remains challenging and may depend on enhancing the immune responses either through booster immunizations or through the inclusion of multiple immunogens. Overall, we can conclude that single monovalent rVSV vaccine vectors can provide partial cross-protection in cases of challenge virus species that are genetically more closely related. As mentioned above, one approach to overcome this limitation is the use of blended monovalent rVSV vaccine vectors, which provide broader protection against homologous and partial protection against certain heterologous challenges. Another approach to overcome the limitations in cross-protection is the use of multivalent rVSV vaccine vectors. In a proof-of-concept study in rodent models protection against ZEBOV and Andes virus (ANDV) or ZEBOV and influenza virus (H5N1) challenge was demonstrated using a single rVSV vector expressing both the ZEBOVgp and the ANDV glycoprotein or ZEBOVgp and a H5 hemagglutinin, respectively. This data showed that the use of bivalent rVSV vectors are a feasible approach to vaccination against multiple pathogens. Based on the results described above, we have over the past two fiscal year successfully generated additional bivalent and trivalent rVSV vectors expressing two or three different filovirus GPs, one as a transmembrane protein (replacing the VSV glycoprotein) and one or two as soluble glycoproteins that will be secreted during vector replication. Recovery of these recombinant vaccine viruses turned out to be difficult but has recently been successful. In vitro characterization of these vectors, including viral growth curves and verification of foreign immunogen expression has been completed. Efficacy testing in the Ebola and Marburg hamster models has resulted in promising results; nonhuman primate studies are scheduled.