Endogenous viral elements constitute a complementary source of antigens for personalized cancer vaccines

The Gram-negative opportunistic bacterial pathogen, Pseudomonas aeruginosa is considered by WHO as a “priority pathogen” for which new antibacterial strategies are urgently needed due to antimicrobial resistance development. In addition, P. aeruginosa is a cause of difficult to treat chronic infections due to its ability to form biofilms. Therefore, pseudomonal vaccines have been proposed as alternative strategies to combat these infections for the last 50 years, however, no vaccines are available on the market for human use. The aim of this study was to investigate the capacity of a vaccine composed of seven antigens, identified using EDEN™ (Efficacy Discriminative Educated Network) - a proteome-wide in silico antigen prediction model within AI-Immunology™ platform - in improving outcomes in a murine model of chronic P. aeruginosa lung infection. The primary endpoint was quantitative bacteriology (Colony forming units - CFU) in the lungs of immunized animals compared to control animals. The secondary endpoints were clinical signs (a clinical score), body temperature and weight loss. Mice immunized with the heptavalent combination vaccine had a significantly 1.2 log10 lower lung CFU compared to the control group. Furthermore, the vaccinated mice presented significantly fewer clinical signs of infection, had less reduction in body temperature and weight loss as, compared to control mice. There was a statistically significant correlation between the lung bacteriology and secondary endpoints. Antibodies against all seven antigens were measured by ELISA confirming their immunogenicity. The encouraging results obtained in this, and previous studies provide a proof-of-concept that EDEN™ is a useful tool in identifying vaccine antigens against P. aeruginosa and possibly other problematic pathogens.