Whole-Genome Sequencing and Machine Learning Reveal Key Drivers of Delayed Sputum Conversion in Rifampicin-Resistant Tuberculosis

Strategic Insights -

Researchers at First Affiliated Hospital in Zhejiang, People's Republic of China, have published a new study on rifampicin-resistant tuberculosis (RR-TB). The study integrated whole-genome sequencing and machine learning to identify clinical and genomic determinants of sputum culture conversion (SCC) failure in 150 RR-TB patients between 2019 and 2023. The research revealed high rates of isoniazid resistance (74.0%) and rpoB mutations (97.3%), predominantly Ser450Leu, with 90% of strains belonging to Lineage 2 (Beijing family).

Key Takeaways:

  • The study identified smear positivity, levofloxacin resistance, and specific resistance mutations as critical drivers of SCC failure in RR-TB patients.
  • Among the 150 RR-TB patients, 64.7% achieved 2-month SCC, but 18.0% remained culture-positive at 6 months.
  • A Random Forest model achieved robust prediction of SCC failure (AUC: 0.86 ± 0.06 at 2 months; 0.76 ± 0.10 at 6 months).
  • Levofloxacin resistance was identified as the top predictor of SCC failure, followed by embB_p. Met306Ile and smear positivity.
  • The study underscores the importance of targeted RR-TB treatment strategies, including the use of whole-genome sequencing and machine learning.

Statistics:

  • 74.0% of RR-TB patients had isoniazid resistance.
  • 97.3% of RR-TB patients had rpoB mutations, predominantly Ser450Leu.
  • 90% of RR-TB strains belonged to Lineage 2 (Beijing family).
  • 64.7% of patients achieved 2-month SCC.
  • 18.0% of patients remained culture-positive at 6 months.
  • The Random Forest model achieved an AUC of 0.86 ± 0.06 at 2 months and 0.76 ± 0.10 at 6 months.

Sources:

  • "Whole-genome sequencing and machine learning reveal key drivers of delayed sputum conversion in rifampicin-resistant tuberculosis." Frontiers in Cellular and Infection Microbiology, 2025,15.
  • Frontiers Media S.A. (Publisher)
  • Qing Fang, Departments of Pulmonary Medicine, First Affiliated Hospital of Ningbo University, Ningbo, Zhejiang, People's Republic of China.
  • Additional authors: Xiangchen Li, Yewei Lu, Junshun Gao, Yvette Wu, Yi Chen, Yang Che.