Table 1. Key gaps identified in the WHO Technical Consultation and suggested solutions [8].

Identified gaps	Proposed solutions	Additional comments
How can we bridge preclinical data into clinical trials and identify the PK-PD measurements that correlate best with bactericidal efficacy and toxicity in vivo?	Phase II and Phase III trials provide opportunities to collect rich and informative data on drug exposures and microbiological response over time. These trials should all include PK sampling, so that PK-PD assessments, linking drug exposures to bactericidal activity and ultimate treatment outcomes, can be performed; these analyses should account for other factors likely to affect outcomes, including disease severity and treatment adherence.	Development and validation of novel biomarkers should be integrated in all PK-PD activities to allow for rapid assessment of the biomarkers and properties of future potential surrogate for bacterial load.
Do we have the best tools currently to identify relevant drug combinations to transition from Phase II to Phase III trials?	Phase IIB/C studies, with arms testing different doses and duration and with collection of treatment outcomes, are likely to strengthen the process for identifying candidate regimens likely to succeed in Phase III.	More quantitative, longitudinal, and time-to-event measures (time-to-positivity on liquid media, time-to-stable culture conversion) are now in common use and are endorsed for broad uptake as viable alternatives to single time-point dichotomous endpoints. Adaptive approaches offer potential reductions in sample size.
How can we overcome the long duration, cost, and constraints of Phase III trials and simplify them without hampering validity and wider drug development?	Both noninferiority and superiority designs are relevant for studies of new TB regimens; their use depends on the indication (drug-susceptible or drug-resistant TB) and on the intended use and value proposition of the new regimen—e.g., better efficacy or shorter duration. New adaptive designs can accelerate Phase II and III trials and improve our ability to select regimens for further investigation.	Innovative, efficient designs (e.g., adaptive strategy designs) need to be further explored for TB drug and regimen development. Many have the potential to accelerate and enhance ability to learn.
What is the role of treatment adherence in development of new TB therapeutics?	Adherence remains an under-valued but important determinant of treatment success. More attention to this domain can help to address the global challenge of treatment default. High-quality data are needed to establish the efficacy and reliability of new methods to measure and sustain adherence.	Ensuring and measuring adherence in clinical trials are essential to correctly interpret results of the trials. Both explanatory and pragmatic trials are needed to answer questions about efficacy and safety and about expected effectiveness in programmatic conditions that includes assessment of adherence.
How can we include key populations, such as children, pregnant women, and people with HIV infection, in clinical trials from the outset, rather than as an afterthought?	More attention is needed to assure the provision of evidence relevant to key subgroups, including pregnant and breastfeeding women, young children, and persons with critical comorbidities such as HIV infection. Novel designs and approaches to integrated substudies would be useful.	The limited evidence base for the prevention and treatment of TB in pregnant women should be emphasized. More PK studies of first-line, second-line, and new anti-TB drugs in pregnant women are needed. Appropriate formulations of drugs for infants and young children should be developed during the early phases of regimen development and testing, whenever feasible. Drug–drug interaction studies between anti-TB and ARV drugs should be conducted as early as feasible within regimen development. Careful joint management of HIV and TB care is essential. In accordance with WHO guidelines, ART should be initiated as soon as possible for all HIV-infected participants with TB in clinical trials, and definitely within the first 8 weeks of TB treatment.

ART, antiretroviral therapy; ARV, antiretrovirals; PK-PD, pharmacokinetics and pharmacodynamics; TB, tuberculosis.