Novel methods of detection and treatment of Pseudomonas aeruginosa infection in cystic fibrosis

Abstract

Although advances in the management of cystic fibrosis (CF) have led to a significant extension of life expectancy, the disease remains life-limiting. Successive pulmonary infections are the primary cause of morbidity, leading eventually to respiratory failure. Pseudomonas aeruginosa (Pa) is the dominant respiratory pathogen but difficulties in detection and increasing failure of conventional treatments to combat established infection mean that an urgent need exists for non-invasive techniques that screen for Pa, particularly in non-expectorating patients, and for novel antimicrobial therapies. The first part of this thesis aimed to determine whether markers in exhaled breath, measured using selected ion-flow tube mass spectrometry (SIFT-MS), could distinguish between CF patients with and without chronic Pa infection. In what is the largest study to date of well-phenotyped CF patients, I demonstrated that a fingerprint that differentiates according to Pa status does appear to be present in exhaled breath, although the technique is not sufficiently sensitive at present to be used on an individual patient basis. The second part of this thesis studies the potential of using bacteriophage in the treatment of Pa infection. I have shown that novel bacteriophage cocktails are efficacious in vitro against laboratory and clinical Pa strains isolated from CF patients. In a murine model of acute infection, I demonstrate that bacteriophage not only hasten clearance of Pa from the lung but also diminish the associated inflammatory response under certain conditions. Efficacy was demonstrated when bacteriophage was given prophylactically and after established infection, indicating potential utility in a clinical scenario. That bacteriophage remain viable following nebulisation, the preferred route of administration for future human trials, was also ascertained. Whilst the results are promising, with potential to improve patient outcomes, both strands of work only demonstrate proof-of-concept for the studied strategies. A significant amount of additional work, some already underway, is necessary before either approach can be used in a clinical setting.