Objective 1- Optimise XBA devices in combination with molecular assays and establish pilot manufacturing and production for clinical studies
In the first 18 months, Avelo has led optimization of two XBA sampling devices: AveloCollect (blow-tube based) and AveloMask, (facemask sampling device). Having established an advanced breath aerosol simulator and undertaken further optimization based on human factor engineering principles, usability assessments, clinical and simulator data, Avelo established pilot manufacturing of both devices under an ISO 13485 quality management system. Liquid elution, extraction and laboratory workflows for molecular testing from breath samples have been optimized in preparation for clinical studies. Devices were shipped to four clinical study sites for further evaluation of their accuracy and screening potential in clinical studies under Objectives 2, 3 and 4.
Objective 2- Comparison of XBA sampling devices for detecting TB using the Respiratory Aerosol Sampling Chamber (RASC) as a benchmark
Both XBA devices have been shipped to DTHF sites in Cape Town, clinical staff trained in sample collection and molecular testing workflows set up and validated (with support from Avelo). Clinical study protocols were designed and local ethical approval obtained, with patient recruitment starting on 20 May 2024. Recruitment of patients with confirmed pulmonary TB was slower than anticipated due to several factors out of the DTHF control, but after addition of extra sites for recruitment, 96% of the target sample size has been reached (as of 25 July), with anticipated recruitment completed in August 2025. 38/75 patients have had XBA samples tested using the Xpert MTB/RIF Ultra (Xpert) molecular platform. Following interim data analysis showing higher sensitivity using qPCR on XBA compared to Xpert, further samples have been bio banked for qPCR testing. All participants have also provided samples in the RASC.
Objective 3- Evaluate the diagnostic accuracy, acceptability and usability of XBA sampling for TB diagnosis in symptomatic patients
To evaluate the diagnostic accuracy of XBA sampling in symptomatic adults, we are conducting a study in patients presenting to a health facility in Bucharest, Romania, and who are being investigated for presumed TB. Study protocols have been designed and written to compare XBA sampling to a reference standard consisting of sputum samples tested by molecular methods and TB culture; XBA sampling devices have been delivered to the study site in Romania, along with training in sample collection. Study initiation was delayed due to ethical and regulatory approval challenges related to interpretation of new EU In Vitro Diagnostic Regulations (IVDR), but all approvals have now been obtained. Recruitment of patients into the study was started by IPMN with support of UKHD and other partners in May 2025. By 29 July 2025, 19/260 patients have been recruited, and XBA samples stored for molecular testing for TB and drug (rifampicin) resistance. Protocols for evaluating acceptability and usability of the devices using mixed-methods are being finalized, with data collection planned for Q3 2025.
Objective 4- Establish the feasibility and acceptability of XBA sampling for screening of TB disease and respiratory viruses
To investigate feasibility of XBA sampling for screening of respiratory infections, we are undertaking a multi-country, multi-site pragmatic feasibility study in migrants arriving at centers in Milan (Italy) and Heidelberg (Germany). Participants provide one XBA sample and a nasal swab - irrespective of respiratory symptoms - and undergo standard of care testing for TB and viruses. We have written study protocols and obtained ethical and regulatory approvals from relevant committees and organizations in both countries. We undertook a ‘study within a trial’ pilot phase alongside mixed-methods data collection to optimize sampling and testing workflows in April 2025. Early data from focus groups with migrants showed good acceptability of both XBA devices, and no safety concerns during sample collection. By 18 July 2025, 88/1,136 participants have been recruited, and will continue throughout the upcoming winter season to allow for circulating seasonal respiratory viruses. Qualitative data from interviews, focus groups and observations will supplement quantitative data on diagnostic yield, and an explorative assessment of self-sampling will take place in Q1 2026.
Objective 5: Establish the potential impact of implementing XBA sampling for diagnosis and screening of respiratory infections
Models for cost-effectiveness and impact modelling of XBA devices will be conceptualized in Q3 2025, and populated with costing data and empirical data of performance of XBA devices from clinical studies, with results expected in late 2026.
