Efficacy

If a drug or regimen is efficacious in non-pregnant adults and adequate drug exposures are achieved in pregnancy, then efficacy can be assumed in pregnant and lactating women without additional efficacy trials. 

The following efficacy concerns should, however, be considered:  

1. TB treatment outcomes for pregnant women are not known to be the same as non-pregnant adults. These should be evaluated in all studies. Standard definitions of TB treatment outcomes in adults should be used (36).
2. Time to culture conversion might be relatively more important during pregnancy not only in terms of maternal and infant clinical outcomes but also for prevention of transmission of TB disease to neonates (rapid reduction in infectiousness). Though this may be challenging to measure in the absence of large sample sizes or long durations of follow up, time-to-culture conversion could be used as a surrogate marker. This is particularly relevant given ongoing research focused on developing highly potent, shortened regimens.
3. Congenital TB is so rare that it might not be an outcome for a trial, but should be considered for inclusion in surveillance studies.

Safety

Investigators should use standardized maternal, pregnancy​,​ and infant outcomes so that data may be more easily pooled across studies.

Separate analyses of disaggregated outcomes do not usually capture the overall clinical experience of the mother–infant pair or take into account correlations between clinical outcomes in the mother and the infant, thus the deleterious effects are compounded. To address these considerations, a composite outcome can be used to examine safety and efficacy in the mothers and their infants simultaneously. The composite outcome may rank or score the mother–infant pairs relative to one another based on the overall clinical experience of the pairs. Composite adverse pregnancy outcome also reduces bias when comparing live birth outcomes including pulmonary TB, small-for-gestational age and low birth weight.

Composite maternal–fetus/infant outcomes and/or estimands can account for risk–benefit trade-offs between maternal health versus fetus/infant health, efficacy versus safety; or if a regimen is superior in a plurality of treatment attributes, then there will be better power to evaluate regimens with respect to the composite outcome, than comparing regimens with respect to a single treatment attribute (such as maternal efficacy alone, or maternal safety alone).