Blood Transcriptomic and Inflammatory Protein Biomarkers Associated with Imminent Pulmonary Exacerbation Risk in Cystic Fibrosis

<b>Rationale:</b> The factors that lead to poor pulmonary exacerbation (PEx) outcomes in individuals with cystic fibrosis (CF) are still being investigated; however, delayed diagnosis and treatment are likely contributory. Identifying individuals at imminent risk of PEx could enable closer monitoring and/or earlier initiation of therapies to improve outcomes. <b>Objectives:</b> The goal of this study was to develop blood-based biomarkers that associate with imminent PEx risk in individuals with CF. <b>Methods:</b> We examined the whole-blood transcriptome and 55 inflammatory proteins from plasma and serum on 72 blood samples from 53 individuals with CF. Biomarker candidate genes and proteins were selected from 14 individuals with CF with paired stable and PEx visits (cohort 1). The biomarker candidates were then estimated and tested to classify individuals with CF who would experience a PEx within 4 months of a stable clinic visit or not (cohort 2). <b>Results:</b> A 16-gene panel and 9-protein panel were identified that could distinguish paired stable and PEx visits (area under the receiver operating characteristic curve [AUC] ± standard error = 0.83 ± 0.28 and 0.92 ± 0.18, respectively). These two panels also demonstrated strong performance in classifying individuals with CF who would experience a PEx within 4 months of a clinically stable visit or not (16-gene panel: AUC = 0.88; 9-protein panel: AUC = 0.83). In comparison, serum calprotectin and clinical variables (i.e., sex, precent predicted forced expiratory volume in 1 s, and the number of IV antibiotics in the preceding year) had AUCs of 0.75 and 0.71, respectively. <b>Conclusions:</b> Blood-based mRNA and protein biomarkers demonstrated strong performance in classifying individuals with CF at risk of imminent PEx. If the findings from this study can be validated, there is the potential to use blood biomarkers to enable more personalized disease activity monitoring in CF.