Bronchiectasis–Chronic Obstructive Pulmonary Disease Overlap: A ROSE by Any Other Name?

The coexistence of bronchiectasis (BE) with chronic obstructive pulmonary disease (COPD) is well described in a subset of patients, but overlapping symptoms and spirometric findings can present challenges in clinical diagnosis and management. Smaller studies have suggested worse outcomes when both BE and COPD are present, but the prognostic significance has not been prospectively investigated in large cohorts (1). There is also substantial overlap in risk factors for both conditions, and pathophysiologic mechanisms for BE–COPD have not been well characterized.

In this issue of the Journal, Polverino and colleagues (pp. 119–127) report findings from the European Multicentre Bronchiectasis Audit and Research Collaboration (EMBARC) registry, a large multinational cohort of 16,730 patients with BE and 5 years of longitudinal follow-up, examining clinical outcomes in BE–COPD and investigating the value of standardized diagnostic criteria for diagnosing this overlap condition (2). Their findings confirmed that the presence of COPD in patients with BE is associated with worse clinical outcomes. Using the recently developed ROSE (radiographic BE, airflow obstruction, symptoms, and smoking exposure) criteria (3), they found that patients with BE are frequently mislabeled as having COPD. Despite this high rate of misdiagnosis, the association between COPD and worse outcomes was present for both clinician-assigned and ROSE-defined BE–COPD. As the largest prospective study of clinical outcomes in BE–COPD to date, this represents an important step toward characterizing the condition and provides a foundation for additional studies to elucidate its underlying pathophysiology.

When interpreting the authors’ findings, it is important to consider the ROSE definition of BE–COPD in the context of the inclusion criteria for the European Multicentre Bronchiectasis Audit and Research Collaboration registry. All participants in the cohort had radiographic BE and respiratory symptoms, so the distinction between BE and BE–COPD was made on the basis of spirometry (ROSE uses a fixed FEV₁:FVC ratio of <0.7 to define obstruction) and smoking history (>10 pack-years). As 35.1% of the BE-only (i.e., non-COPD) cohort were current or former smokers, COPD was defined on the basis of spirometric obstruction alone in a substantial proportion of patients with BE–COPD. This raises a key question as to whether the observed associations between BE–COPD and increased mortality and exacerbations reflect the presence of airflow obstruction in patients with more advanced BE who are thus at higher risk for poor outcomes or if BE–COPD represents a distinct endotype with unique pathophysiology. COPD diagnosis was associated with higher bronchiectasis severity index, but this relationship is difficult to disentangle, as most BE severity index criteria (4) are also associated with increasingly severe COPD (age, body mass index, FEV₁, exacerbations/hospitalization, severity of dyspnea, colonization with Pseudomonas aeruginosa and other bacteria), in addition to radiographic severity of BE. Interestingly, rates of under- and overdiagnosis of BE–COPD when comparing the objective ROSE definition with clinician diagnosis varied by country. Geographic differences in healthcare access or environmental exposures other than cigarette smoke (e.g., biomass fuels, air pollution, passive tobacco smoke exposure) could account for some of these discrepancies and may be important considerations for further studies of BE–COPD. The observation of increased exacerbations and mortality for both clinician-assigned and objectively defined BE–COPD suggests that despite the observed discrepancies between these approaches, clinicians are already quite good at identifying patients at high risk for adverse outcomes. However, we agree with the authors that the ROSE definition is more appropriate for future clinical studies.

Importantly, the authors’ findings should facilitate additional studies to identify phenotypes and specific risk factors. In addition to similarities in symptoms and clinical presentation, there are several potential mechanisms that could underlie the coexistence of BE and COPD. Emerging evidence supports the potential roles of CFTR (cystic fibrosis transmembrane regulator) dysfunction (5), chronic bacterial inflammation (6), adaptive immune deficiencies (7, 8), and insults occurring during early lung development in the pathogenesis and progression of COPD, in addition to their established roles in BE. Further studies of patients with BE–COPD in these areas will lead to greater understanding of the nature of this overlap condition.

The ultimate aim, however, must be the characterization of traits or phenotypes that can be targeted for therapeutic intervention. The implementation of a standardized definition of BE–COPD could facilitate efforts to identify biomarkers that predict treatment response. A recent example that followed this translational arc can be found in the shift in emphasis from recognizing and defining asthma–COPD overlap as a unique syndrome (9) to the discovery that type 2, or eosinophilic inflammation predicts treatment response to inhaled corticosteroids and biologic agents such as dupilumab (10). Studies in BE–COPD should similarly be directed toward diagnostic and treatment approaches to mitigate the high morbidity and mortality in BE–COPD that are highlighted by this study.