Within-host population dynamics of extensively drug-resistantrevealed by an over 3-year longitudinal study.

BACKGROUND AND OBJECTIVES: Drug resistance is a major contributor to tuberculosis (TB) deaths worldwide. Understanding the dynamics of in-host evolution of(MTB) drug resistance can help to improve treatment success rates.

METHODOLOGY: The microevolution of drug-resistant MTB was studied in three patients with long-standing, extensively drug-resistant TB (XDR-TB) by analyzing whole genome sequences of serial isolates collected during treatment.

RESULTS: We identified three patterns ofMTB microevolution during long-term, ineffective treatment: (i) new drug-resistant subpopulations emerge and compete with other subpopulations during treatment; (ii) drug resistance profiles remaining stable without the emergence of new drug-resistant subpopulations; and (iii) after a drug is stopped, new drug-resistant subpopulations continue to emerge and compete with existing subpopulations.

CONCLUSIONS AND IMPLICATIONS: The microevolution of drug-resistant MTB within patients on long-term ineffective treatment is complex. Subpopulations with different resistance-conferring mutations can compete with each other and with newly emerged subpopulations. Often, one subpopulation eventually dominates and achieves long-term stability. This work deepens the understanding of MTB microevolution in XDR-TB patients.