Table 1. . Treatment efficacy outcomes used in clinical pharmacokinetic-pharmacodynamic studies and clinical trials for tuberculosis.

Timing of outcome measurement	Outcome measurement	Strengths	Weaknesses
Late (typically the end of post-treatment follow-up)	Treatment failure or post-treatment relapse	1. Clinically relevant (target PK–PD parameters for TDM would ideally be validated against this outcome) 2. Gold standard end point for stable cure in Phase III clinical trials	1. Studies are very long and expensive to conduct 2. Unless careful molecular tests can be undertaken the end point of relapse may be contaminated by TB re-infection in highly endemic settings, especially with high rates of HIV co-infection
Early (typically 2 months for DS-TB, may be 24 weeks for MDR-TB)	SCC† at a defined end point	1. Results are simple to understand and interpret 2. Only requires sputum sampling at two time-points	1. Only modest correlation with late outcomes 2. Binary data do not use all information (e.g., 2-month results cannot discriminate between patients who are culture negative at 8 weeks, even if one culture converted at 2 weeks and the other at 6 weeks)
	Time to SCC†	1. Results are simple to understand and interpret 2. Provides more discriminatory data than SCC	1. Correlation with late outcomes has not been well validated 2. Weekly sputum sampling required; wider sampling windows reduce the accuracy of the measurement
	Statistical modeling of bacterial elimination rates from serial quantitative bacteriology data‡	1. Provides information on antimicrobial efficacy across the whole sampling time, even on patients who do not convert to negative 2. Allows multiphase patterns of bacterial clearance to be assessed	1. Correlation of summary parameters from mixed effects modeling with late outcomes has not been well validated 2. Data analysis is computationally complex to perform 3. Results are not always simple to understand and interpret

^†SCC or ‘time to SCC’ data can be generated using solid or liquid culture media; culture conversion is often later in liquid culture systems.

^‡Quantitative bacteriology data can be generated from log₁₀CFU/ml counts on solid media to Time to Positive results in liquid culture systems.

DS-TB: Drug-susceptible TB; MDR-TB: Multidrug resistant TB; PK–PD: Pharmacokinetic–pharmacodynamic; SCC: Sputum culture conversion; TDM: Therapeutic drug monitoring.