Rationale: Daily rifapentine plus isoniazid-pyrazinamide in mice infected with Mycobacterium tuberculosis produces cure in 3 months. Whether cure corresponds to latent infection contained by host immunity or true tissue sterilization is unknown.
Objectives: To determine the length of treatment with rifapentine-isoniazid-pyrazinamide or rifampin-isoniazid-pyrazinamide needed to prevent relapse in immune-deficient mice.
Methods: Aerosol-infected BALB/c and nude mice were treated 5 days per week with either 2 months of the rifapentine-based regimen followed by rifapentine-isoniazid up to 12 months or the same regimen with rifampin instead of rifapentine. Cultures of lung homogenates were performed during the first 3 months and then every 3 months. Relapse rates were assessed after 3, 6, 9, and 12 months of treatment in BALB/c (± 1 mo of cortisone) and nude mice.
Measurements and main results: All rifapentine-treated mice were lung culture-negative at 3 months but 13% of BALB/c that received cortisone and 73% of nude mice relapsed. After 6, 9, and 12 months of treatment no mouse relapsed. Rifampin-treated BALB/c mice remained culture positive at 3 months. All were culture negative at 6, 9, and 12 months. None, including those receiving cortisone, relapsed. Rifampin-treated nude mice harbored more than 4 log(10) lung cfu at Month 2 and approximately 6 log(10) cfu with isoniazid resistance at Month 3. A supplementary experiment demonstrated that 7 days a week treatment did not prevent isoniazid resistance, whereas addition of ethambutol did.
Conclusions: In nude mice, sterilization of tuberculosis is obtained with rifapentine-containing treatment, whereas failure with development of isoniazid resistance is obtained with rifampin-containing treatment.