| Tuberculosis (TB) is the second-leading cause of death by infection. Novel treatment strategies are critically needed to shorten the duration of TB therapy and treat drug-resistant TB. However, several issues hamper the evaluation of new TB regimens and, thus, the TB drug development effort. First, clinical trials evaluating treatment success without relapse are necessarily lengthy, complex, and expensive. Appropriate trial design, trial sequencing, and dosing decisions are not straightforward and depend on the proposed role of the experimental drug or drugs in a multi-drug TB regimen. Second, there are few trial-level surrogate markers of TB treatment success; those in current use have important limitations. Third, statistical methodologies commonly employed in TB trials are not maximally efficient or informative.;This thesis presents original research in the field of TB with a specific focus on the evaluation of new TB therapeutics. First, a phase I clinical trial of two new TB drugs in healthy volunteers is presented to demonstrate how pharmacokinetic (PK) and pharmacodynamic (PD) data can be used to inform and guide subsequent clinical trial design and dosing choices. Second, a longitudinal cohort study nested in a Phase II TB treatment trial is described. In this study, a new diagnostic test that may serve as a surrogate marker for TB treatment success without relapse is evaluated. The third study is a retrospective cohort study evaluating differences in TB treatment outcomes between patients with and without diabetes mellitus. The study illustrates the use of two different statistical techniques to evaluate sputum culture conversion, a common surrogate for relapse.;In summary, this thesis presents one clinical trial, one longitudinal cohort study nested in a clinical trial, and a retrospective cohort study, which have been published or submitted for publication. These studies address methodological challenges related to the evaluation of new TB therapeutics, with a focus on using PK/PD data to enhance drug evaluation and study design; the evaluation and use of new surrogate markers of treatment efficacy; and using informative and efficient statistical techniques. |
