High Intrapulmonary Rifampicin and Isoniazid Concentrations Are Associated With Rapid Sputum Bacillary Clearance in Patients With Pulmonary Tuberculosis

Abstract

Background: Intrapulmonary pharmacokinetics may better explain response to tuberculosis (TB) treatment than plasma pharmacokinetics. We explored these relationships by modeling bacillary clearance in sputum in adult patients on first-line treatment in Malawi.

Methods: Bacillary elimination rates (BER) were estimated using linear mixed-effects modelling of serial time-to-positivity in mycobacterial growth indicator tubes for sputum collected during the intensive phase of treatment (weeks 0-8) for microbiologically confirmed TB. Population pharmacokinetic models used plasma and intrapulmonary drug levels at 8 and 16 weeks. Pharmacokinetic-pharmacodynamic relationships were investigated using individual-level measures of drug exposure (area-under-the-concentration-time-curve [AUC] and Cmax) for rifampicin, isoniazid, pyrazinamide, and ethambutol, in plasma, epithelial lining fluid, and alveolar cells as covariates in the bacillary elimination models.

Results: Among 157 participants (58% human immunodeficiency virus [HIV] coinfected), drug exposure in plasma or alveolar cells was not associated with sputum bacillary clearance. Higher peak concentrations (Cmax) or exposure (AUC) to rifampicin or isoniazid in epithelial lining fluid was associated with more rapid bacillary elimination and shorter time to sputum negativity. More extensive disease on baseline chest radiograph was associated with slower bacillary elimination. Clinical outcome was captured in 133 participants, with 15 (11%) unfavorable outcomes recorded (recurrent TB, failed treatment, or death). No relationship between BER and late clinical outcome was identified.

Conclusions: Greater intrapulmonary drug exposure to rifampicin or isoniazid in the epithelial lining fluid was associated with more rapid bacillary clearance. Higher doses of rifampicin and isoniazid may result in sustained high intrapulmonary drug exposure, rapid bacillary clearance, shorter treatment duration and better treatment outcomes.

Keywords: antibiotics; antitubercular; pharmacodynamics; pharmacokinetics; tuberculosis.

Figure 1.

Epithelial lining concentrations and bacillary elimination rate. Higher rifampicin (A) and isoniazid (B) ELF AUC were associated with more rapid sputum bacillary elimination (more negative BER) when added to the PK-PD model (Δ objective function value ≥3.84, P = .05, χ² distribution, 1 degree of freedom). Line of best fit shown for illustration. PK data have been log-transformed. Abbreviations: AUC, area under the concentration-time curve; BER, bacillary elimination rate; ELF, epithelial lining fluid.

Figure 2.

Modelled bacillary elimination rate and time on treatment, stratified by 2-month culture conversion and clinical outcome. Serial TTP data from each participant included in the model (n = 125), with each line representing an individual participant and each point the modeled transformed TTP (log₁₀(1/TTP)) as a measure of bacillary load. The rate of reduction in transformed TTP represents the sputum bacillary elimination rate. The dashed horizonal line is the limit of quantification for MGIT data at −1.623 (log₁₀(1/42 days)), with partial likelihood modeling [16, 17] accounting for samples beyond the limit of quantification. In the top panels (A, B), solid lines and points represent samples from those participants that culture converted by 2 months (2MCC), dashed lines and hollow points those participants that failed to culture convert. 
A, Each line represents an individual participant. B, The lines summarize the bacillary elimination rate for sputum culture converters and nonconverters. In the bottom panels (C, D), the solid lines and points represent those participants with favorable clinical outcomes, the dashed lines and hollow points those with unfavorable clinical outcomes. The dashed black horizonal line is the lower limit of quantification for MGIT TTP data at −1.623 (log₁₀(1/42 days)). Abbreviations: MGIT, mycobacteria growth indicator tube; TTP, time to positivity. C, Each line represents an individual participant. D, The lines summarize the bacillary elimination rate for those with favorable and unfavorable outcomes.