Predicting individual outcomes from a range of intravenous antibiotics for sputum-producing exacerbations of chronic lung disease, using microbial genomics and artificial intelligence

Prospective observational study to predict outcomes from a range of intravenous antibiotics for sputum-producing exacerbations of chronic lung disease, using microbial genomics and artificial intelligence

Status

Recruiting

Phases

Unknown

Study type

Observational

Source

ISRCTN

Registry ID

ISRCTN14487574

Enrollment

600

Registered

2025-01-22

Start date

2025-01-01

Completion date

Unknown

Last updated

2025-02-03

For informational purposes only — not medical advice. Sourced from public registries and may not reflect the latest updates. Terms

Conditions

Acute exacerbations of chronic lung disease productive of purulent sputum Respiratory

Interventions

Subjective (e.g. questionnaires) and objective (e.g. time till next exacerbation) assessment of response to therapy before starting treatment and every seven days while on treatment, concurrent with s

Eligibility

Sex/Gender

All

Inclusion criteria

Inclusion criteria: 1. Age: Subjects must be 18 years or older 2. Consent: Subjects must be able and willing to consent 3. Attending the Cambridge Centre for Lung Infection to commence a clinician-directed course of intravenous antibiotics abx for the treatment of clinically diagnosed acute pulmonary exacerbation of chronic lung disease 4. Able to produce a mucopurulent or purulent sputum sample at recruitment

Exclusion criteria

Exclusion criteria: Patients unable to produce purulent or mucopurulent sputum at recruitment

Design outcomes

Primary

Measure	Time frame
The generation of a mechanistic model to explain the relationship between clinical data, conventional sputum microbiology and antibiotic sensitivity results, next-generation sequencing of sputum microbial nucleic acids (DNA and RNA) and clinical outcomes, including: 1. Improvement in symptom scores measured using questionnaires including Chronic Airways Assessment Test (CAAT) and Visual Analogue Scales (VAS) measured every seven days till completion of treatment and two weeks after treatment ends 2. Time till the next exacerbation measured using patient medical notes at one timepoint 3. Changes to antimicrobial resistance, predicted using whole sputum genomics and artificial intelligence	—

Measure

Time frame

The generation of a mechanistic model to explain the relationship between clinical data, conventional sputum microbiology and antibiotic sensitivity results, next-generation sequencing of sputum microbial nucleic acids (DNA and RNA) and clinical outcomes, including: 1. Improvement in symptom scores measured using questionnaires including Chronic Airways Assessment Test (CAAT) and Visual Analogue Scales (VAS) measured every seven days till completion of treatment and two weeks after treatment ends 2. Time till the next exacerbation measured using patient medical notes at one timepoint 3. Changes to antimicrobial resistance, predicted using whole sputum genomics and artificial intelligence

—

Secondary

Measure	Time frame
The generation of an AI algorithm that can predict the outcome of intravenous antibiotics treatment of sputum-producing acute exacerbation of chronic lung disease. Prediction accuracy will be measured against withheld test data using standard tools from the machine learning toolbox including ROC AUC for classification and the r2 metric for regression to continuous scores.	—

Countries

England, United Kingdom

Contacts

Public ContactDavid Abelson

david.abelson@nhs.net+44 (0)7514612874

Outcome results

None listed

Source: ISRCTN (via WHO ICTRP) · Data processed: Feb 4, 2026