Azithromycin to Modify Bronchiectasis Exacerbation Risk (AMBER)

The Azithromycin to Modify Bronchiectasis Exacerbation Risk (AMBER) trial is a prospective, randomized, double-blind, placebo-controlled, parallel-group superiority clinical trial conducted within the Assiut University bronchiectasis translational research platform.

The trial enrolls adults with clinically and radiologically confirmed non-cystic fibrosis bronchiectasis (NCFB). Participants will be randomized at the individual-patient level in a 1:1 allocation ratio to one of two treatment groups:

1. Standard care plus matching placebo once daily for 12 months.
2. Standard care plus azithromycin 250 mg orally once daily for 12 months.

Standard bronchiectasis care may include mucolytics, bronchodilators, chest physiotherapy, patient education, airway-clearance support, and treatment of acute exacerbations with antibiotics with or without systemic corticosteroids according to clinical indication.

The primary objective is to determine whether azithromycin added to standard care reduces the occurrence of at least one bronchiectasis exacerbation during 12-month follow-up compared with matching placebo added to standard care.

The primary endpoint is occurrence of at least one bronchiectasis exacerbation during 12-month follow-up. A qualifying bronchiectasis exacerbation requires both clinically significant worsening of bronchiectasis-related respiratory symptoms and initiation or escalation of acute therapy. Acute therapy may include antibiotics, systemic corticosteroids where clinically indicated, intensified airway-clearance treatment, emergency assessment, or hospital admission.

Secondary clinical outcomes include total number of bronchiectasis exacerbations during follow-up, time to first bronchiectasis exacerbation, severe bronchiectasis exacerbation requiring hospitalization, bronchiectasis-related hospital admission, longitudinal change in forced expiratory volume in one second percent predicted (FEV1% predicted), resting room-air oxygen saturation measured by pulse oximetry (SpO2), C-reactive protein (CRP), neutrophil-to-lymphocyte ratio (NLR), Modified Medical Research Council dyspnea scale (mMRC), Bronchiectasis Health Questionnaire (BHQ) where systematically collected, and safety outcomes.

Safety outcomes include adverse events (AEs), serious adverse events (SAEs), gastrointestinal intolerance, cardiac symptoms or cardiac safety findings, QT prolongation or arrhythmia-related concern where assessed, treatment interruption, treatment discontinuation, clinically significant laboratory or symptom-based safety concerns, microbiological or resistance-related concerns where systematically collected, and death where applicable.

The trial uses a double-blind placebo-controlled design. Participants, treating clinicians, investigators involved in follow-up, outcome assessors, endpoint adjudicators where used, and investigators involved in outcome assessment will remain unaware of participant-level randomized treatment assignment until database lock unless emergency unblinding is required for participant safety. Study medication will be prepared in sequentially numbered containers according to the concealed randomization sequence.

The formal sample-size calculation is based on the primary binary endpoint. The calculation assumes a control-arm exacerbation occurrence of 60%, azithromycin-arm exacerbation occurrence of 40%, two-sided alpha level of 0.05, 90% statistical power, 1:1 allocation, and 10% allowance for attrition or incomplete endpoint ascertainment. The minimum attrition-adjusted requirement is 145 participants per arm. The planned AMBER randomized sample is at least 500 complete analyzable participants, with 250 participants per arm. If operationally feasible, randomized enrollment may extend to approximately 520 complete analyzable participants while preserving the 1:1 allocation ratio to improve precision for secondary and translational analyses.

The AMBER trial is scientifically linked to the Bronchiectasis Assessment of Severity and Exacerbations (BASE) framework and the Bronchiectasis Phenotype Identification Model (BPIM), which are separate locked methodological tools within the same translational platform. BASE and BPIM outputs will not be used for randomization stratification. If the relevant tool versions are locked and published before AMBER outcome analysis begins, BASE and BPIM may be applied for prespecified characterization, supportive adjustment, subgroup analysis, treatment-response interpretation, dynamic state occupancy analysis, and multistate transition analysis. No BASE or BPIM coefficient, threshold, class definition, score, model structure, or labeling rule will be recalculated, refitted, re-estimated, reweighted, recalibrated, or threshold-retuned using AMBER randomized treatment data.

Prespecified secondary translational analyses include longitudinal mixed-effects modeling, time-to-event analysis, exacerbation burden modeling, point-change analysis, mediation analysis, subgroup treatment-response heterogeneity analysis, dynamic BASE/BPIM state occupancy analysis, multistate and dynamic state transition analysis, correlation and regression analyses, joint longitudinal-survival modeling, sensitivity analyses, per-protocol supportive analysis, and safety analysis. These analyses are prespecified secondary translational analyses. They are not primary endpoints, are not interpreted as definitive primary treatment-effect tests, and will not replace the primary intention-to-treat (ITT) randomized comparison.

An independent Data Monitoring Committee (DMC) will monitor participant safety, serious adverse events, cardiac safety concerns, treatment discontinuation, recruitment progress, retention, and trial conduct. The DMC will operate independently from investigators responsible for participant recruitment, treatment delivery, endpoint assessment, and primary statistical analysis.

The locked Version 1.0 methodological disclosure document, protocol and statistical analysis plan (SAP), primary sample-size source code, and endpoint-level sample-size support matrix are archived in Zenodo: https://doi.org/10.5281/zenodo.20178963.

In addition, the AMBER public preregistration is available through the Open Science Framework under DOI 10.17605/OSF.IO/RE54V.