Chronic rhinosinusitis (CRS) is a debilitating condition affecting around 9% of Australians. Exacerbating CRS are sinonasal bacterial biofilms, which are often resistant to current treatments. Staphylococcus aureus forms biofilms in sinuses of CRS patients, and their presence associated with poorer outcomes. Development of novel therapies is therefore paramount and this study assessed safety and efficacy of a topically applied S. aureus specific bacteriophage cocktail for treatment of S. aureus biofilms in vivo. Utilising a sheep model of sinusitis, a bacteriophage cocktail (CT-SA, 2x106pfu/mL, in 0.9% saline) was applied to frontal sinuses (n=6) and compared to a control treatment (NT, an inactive version of CT-SA, in 0.9% saline). Safety was assessed using scanning electron microscopy (SEM) and histology of sinonosal mucosa following 3 days of treatment. Efficacy was assessed through initial formation of S. aureus biofilm in frontal sinuses and subsequent treatment with either CT-SA or NT for a period of 5 days. Biofilm formations present on sinus mucosa were imaged using a LIVE/DEAD BacLight stain and confocal scanning laser microscopy, and COMSTAT2 software allowed computation of biofilm biomass levels. Results showed no significant differences in inflammation and tissue morphology when comparing CT-SA and NT treatments. In addition, SEM investigation found CT-SA treatment was not toxic or harmful to mucosal cilia. Biofilm biomass quantification highlighted significant (p=0.0043) reduction in biofilm levels in the CT-SA (0.102 ± 0.0197µm³/µm²) group compared with the control (0.608 ± 0.174µm³/µm²). Our results demonstrate this S. aureus specific bacteriophage cocktail to be safe and efficacious for topical sinonasal treatment of S. aureus biofilms. Translation of this treatment to a clinical setting is the next step in assessing CT-SA as a therapeutic against S. aureus biofilms in CRS.