Comparison and correlation of in vitro and in vivo approaches for determining Pseudomonas aeruginosa bacteriophages activity

Abstract The rise of infections caused by multidrug-resistant bacteria has driven the search for alternative therapies, with Pseudomonas aeruginosa representing a major target. Among these alternatives, the use of bacteriophages (phage therapy) employs these viruses that specifically infect bacteria to fight infections. Although phage efficacy has been demonstrated both in vitro and in vivo , few studies have directly correlated these two dimensions. This study aimed to evaluate, compare, and correlate in vitro and in vivo phage activity. For this, three strictly lytic bacteriophages (ph9027, ph1461, and ph3678) were isolated and tested against 15 clinical isolates and three reference strains of P. aeruginosa . Their in vitro activity was assessed through Spot-Test, Efficiency of Plating, Local Virulence, and Virulence Index assays, followed by in vivo assays in Galleria mellonella larvae and Balb/C mice in a peritonitis model with bacteremia induction. Virulence Index and animal treatments were performed using a phage cocktail (1:1:1 ratio, 1.0 × 10⁹ PFU/mL). In vitro assays revealed varying phage activity, with the Virulence Index showing the strongest correlation with in vivo results in G. mellonella (rs = 0.7648). Larval treatment achieved up to 100% survival, significantly different ( p < 0.05) from bacteria-only controls. In mice, phage therapy reduced bacterial loads by up to 3 logs in blood, spleen, peritoneal lavage, liver, and lung ( p < 0.05) in comparison with control groups. Therefore, the present study, in addition to reiterating the safety and therapeutic efficacy of phage therapy in animal models, presents evidence that Virulence Index techniques, associated with the survival results of G. mellonella larvae, can be excellent tools for directing preclinical assays due to the strong statistical correlation observed between these data and those obtained in mouse experiments.