Measuring recent in-country TB transmission using a classification model with whole genome sequencing data

Area of Science:

• Epidemiology
• Genomics
• Public Health

Background:

• Molecular typing of Mycobacterium tuberculosis isolates identifies transmission clusters.
• Clustering alone does not confirm recent in-country transmission; infections may be imported or historical.

Purpose of the Study:

• To develop and validate a classification model to differentiate recent TB infections within the Netherlands.
• To accurately estimate recent domestic TB transmission using whole genome sequencing (WGS) data.

Main Methods:

• A two-stage classification model was developed: individual patient labeling followed by cluster labeling.
• The model integrated epidemiological data, WGS, and confirmed epidemiological links from cluster investigations.
• Patients were categorized into 'likely,' 'possibly,' and 'unlikely' recent infection groups.

Main Results:

• In 2018-2023, 28% of TB patients with WGS results had clustered isolates.
• After classification, only 11% to 15% of cases were likely recently infected within the Netherlands.
• The majority of clustered cases represented imported or non-recent infections.

Conclusions:

• The WGS-based classification model is a valuable tool for monitoring TB transmission and progress toward pre-elimination.
• The findings highlight that a significant portion of clustered TB cases are not due to recent domestic transmission.
• Accurate estimation of recent in-country transmission is crucial for targeted TB control strategies.