Wound infections are the main cause of sepsis in patients with burns and increase burn-related morbidity and mortality. Bacteriophages, natural bacterial viruses, are being considered as an alternative therapy to treat infections caused by multidrug-resistant bacteria. We aimed to compare the efficacy and tolerability of a cocktail of lytic anti-Pseudomonas aeruginosa bacteriophages with standard of care for patients with burns.
In this randomised phase 1/2 trial, patients with a confirmed burn wound infection were recruited from nine burn centres in hospitals in France and Belgium. Patients were eligible if they were aged 18 years or older and had a burn wound clinically infected with P aeruginosa. Eligible participants were randomly assigned (1:1) by use of an interactive web response system to a cocktail of 12 natural lytic anti-P aeruginosa bacteriophages (PP1131; 1 × 10 ⁶ plaque-forming units [PFU] per mL) or standard of care (1% sulfadiazine silver emulsion cream), both given as a daily topical treatment for 7 days, with 14 days of follow-up. Masking of treatment from clinicians was not possible because of the appearance of the two treatments (standard of care a thick cream, PP1131 a clear liquid applied via a dressing), but assignments were masked from microbiologists who analysed the samples and patients (treatment applied while patients were under general anaesthetic). The primary endpoint was median time to sustained reduction in bacterial burden by at least two quadrants via a four-quadrant method, assessed by use of daily swabs in all participants with a microbiologically documented infection at day 0 who were given at least one sulfadiazine silver or phage dressing (modified intention-to-treat population). Safety was assessed in all participants who received at least one dressing according to protocol. Ancillary studies were done in the per-protocol population (all PP1131 participants who completed 7 days of treatment) to assess the reasons for success or failure of phage therapy. This trial is registered with the European Clinical Trials database, number 2014-000714-65, and ClinicalTrials.gov, number NCT02116010, and is now closed.
Between July 22, 2015, and Jan 2, 2017, across two recruitment periods spanning 13 months, 27 patients were recruited and randomly assigned to receive phage therapy (n=13) or standard of care (n=14). One patient in the standard of care group was not exposed to treatment, giving a safety population of 26 patients (PP1131 n=13, standard of care n=13), and one patient in the PP1131 group did not have an infection at day 0, giving an efficacy population of 25 patients (PP1131 n=12, standard of care n=13). The trial was stopped on Jan 2, 2017, because of the insufficient efficacy of PP1131. The primary endpoint was reached in a median of 144 h (95% CI 48-not reached) in the PP1131 group versus a median of 47 h (23-122) in the standard of care group (hazard ratio 0·29, 95% CI 0·10-0·79; p=0·018). In the PP1131 group, six (50%) of 12 analysable participants had a maximal bacterial burden versus two (15%) of 13 in the standard of care group. PP1131 titre decreased after manufacturing and participants were given a lower concentration of phages than expected (1 × 10 ² PFU/mL per daily dose). In the PP1131 group, three (23%) of 13 analysable participants had adverse events versus seven (54%) of 13 in the standard of care group. One participant in each group died after follow-up and the deaths were determined to not be related to treatment. The ancillary study showed that the bacteria isolated from patients with failed PP1131 treatment were resistant to low phage doses.
At very low concentrations, PP1131 decreased bacterial burden in burn wounds at a slower pace than standard of care. Further studies using increased phage concentrations and phagograms in a larger sample of participants are warranted.
European Commission: Framework Programme 7.