A primary cause of death for persons with cystic fibrosis (CF) is respiratory failure resulting from damage by chronic lung infections and inflammation. More than 75% of CF patients will develop chronic P. aeruginosa (Pa) lung infections, and Pa infection is highly correlated with worse patient outcomes. Therefore, there has been a strong emphasis on early treatment and eradication of Pa infections in order to delay chronic infection and preserve lung function. Pa lung infections are aggressively treated with inhaled antibiotics aimed at bacterial eradication, regardless of symptomatology, which has led to small, but significant decreases in the prevalence of Pa chronic CF lung infections in all age groups. The standard of care for diagnosing lung infection etiology is sputum culture, but many patients (especially children) cannot spontaneously produce sputum. The most commonly-used alternative to sputum culture, upper-airway samples (e.g., oropharyngeal swabs), have poor sensitivity (34 ? 71%) in diagnosing Pa lower-airway infections, and the collection of bronchoalveolar lavage fluid is invasive. Therefore, Pa lung infections can go undiagnosed for months, delaying treatment and contributing to patient morbidity and mortality. Our long-term goal is to substantially shorten the time-to-diagnosis through the development of sensitive and specific breath-based diagnostics for infection etiology in the polymicrobial CF lung milieu. The overall objectives of this application is to focus on the most broadly critical pathogen, Pa, 1) validating and expanding our putative volatile biomarker panel in expectorating subjects, and 2) evaluating breath volatile variability in non- expectorating subjects in order to develop a validation study in that patient cohort. Our central hypothesis is that our panel of volatile biomarkers found in breath are specific and sensitive for detecting Pa lung infections, independent of current co-infections (symptomatic or asymptomatic) and infection history. This hypothesis has been formulated from our own preliminary data from patients and a murine model. We will test our central hypothesis through pursuit of two Specific Aims, 1: Refine and validate volatile biomarkers in breath of expectorating adult and pediatric CF patients for detecting established Pa lung infections. We expect the breath biomarkers to have a higher sensitivity (> 71%) and equivalent specificity (? 95%) to OP swab cultures. And, 2: Quantify intra-subject breath variability in the second target (non-expectorating) pediatric population. Expected outcomes: These data will enable our primary goal: to validate breath biomarkers for the detection of Pa infection in expectorating adults and pediatrics with CF. We expect the breath biomarkers to have a higher sensitivity and equivalent specificity to OP swab cultures. This will enable direct translation to the clinic through the use of one-dimensional GC-MS, which is common to many clinical labs and has no translational barrier from our GCGC-TOFMS system. Our secondary goal is to obtain the preliminary data needed to translate this work to non-expectorating children with CF in order to detect new-onset Pa infections.
The proposed research is relevant to public health because it addresses a specific, high-impact need of persons with cystic fibrosis, which is to have a diagnostic that can detect the bacterial pathogen Pseudomonas aeruginosa earlier. Using a state-of-the-art analytical system to validate and further refine the diagnostic panel of small volatile molecules in exhaled breath, this research promises to eventually translate into a system that will provide a diagnosis of respiratory infection etiology in minutes. This will benefit the patient because the correct treatment depends on the correct diagnosis and thus, this work is relevant to the NIH core mission to protect and improve health as well as the mission of the NIAID to advance diagnostic options for infectious diseases.
