Abstract / Summary
The management of non-cystic fibrosis bronchiectasis (NCFB) is challenged by clinical heterogeneity and by the absence of head-to-head comparisons between established antimicrobial therapies and novel anti-inflammatory agents. We aimed to compare the relative efficacy and durability of dipeptidyl peptidase-1 (DPP-1) inhibitors, long-term macrolides, and inhaled antibiotics for the prevention of exacerbations in adults with NCFB. We systematically searched PubMed, Embase, Cochrane CENTRAL, Web of Science, and ClinicalTrials.gov from inception to April 2026 for randomised controlled trials (RCTs) comparing long-term macrolides, inhaled antibiotics, or DPP-1 inhibitors with placebo (or with each other) for exacerbation prevention in NCFB. The protocol was prospectively registered in PROSPERO (ID: CRD420261350340). A frequentist random-effects network meta-analysis (NMA) was performed. The primary outcome was the annualised rate of exacerbations (AER, expressed as a rate ratio); the principal secondary outcome was time to first exacerbation (hazard ratio, HR). Serious adverse events (SAE) were summarised descriptively. Treatments were ranked using P-scores as descriptive measures of relative ranking; P-scores were not interpreted as evidence of statistically significant superiority between active treatments. Between-study heterogeneity was quantified using I², τ² and Cochran’s Q. Fifteen RCTs providing 24 active treatment arms were included; 3,726 participants were randomised to active treatment, each with a matched placebo comparator. The evidence network was star-shaped, with placebo as the common comparator. For AER, azithromycin had the highest P-score (0.91), followed by HSK31858 40 mg (0.87) and HSK31858 20 mg (0.73); between-study heterogeneity was moderate and statistically significant (I² = 55.0%, τ² = 0.043, p = 0.018). For time to first exacerbation, azithromycin (P-score 0.94) and HSK31858 40 mg (0.91) again occupied the highest-ranked positions, with lower heterogeneity (I² = 32.5%, p = 0.139). However, the indirect azithromycin-versus-HSK31858 comparisons were not statistically significant for either outcome. Descriptive safety data indicated a favourable serious-adverse-event profile for DPP-1 inhibitors, whereas inhaled aztreonam and inhaled ciprofloxacin were associated with higher discontinuation. DPP-1 inhibitors, particularly HSK31858, showed favourable estimates for exacerbation prevention and may represent non-antibiotic alternatives to long-term macrolides. However, treatment rankings were based predominantly on indirect comparisons and should not be interpreted as evidence of statistical superiority between active treatments. Inhaled antibiotics may have an important targeted role, particularly in patients with chronic Pseudomonas aeruginosa infection and frequent exacerbations. These findings support individualised, risk-based treatment selection.