Background: External controls are evolving as a valuable approach in regulatory decision-making regarding drug development, particularly when single-arm trials (SATs) are warranted in cases where it is not feasible to conduct randomized controlled trials (RCTs). However, their broader acceptance requires an ongoing accumulation of evidence to demonstrate their reliability, applicability across diverse therapeutic areas, and alignment with regulatory standards. This dissertation research advances the understanding of external control study methodology, offering practical recommendations for researchers, regulatory bodies, and healthcare decision-makers. It contributes to the growing body of evidence on real-world data applications and addresses gaps in the use of patient-reported outcomes in external control studies, reinforcing the importance of methodological rigor in external control studies to maximize their reliability and impact in drug development research and regulatory approvals.

Objective: The overarching aim of this dissertation research is to evaluate the utility of using existing real-world data and historical trial data as external controls in target trial emulation studies assessing treatment effect on objective and patient-reported outcomes (PRO) in the context of cardiovascular therapies.

Methods: A series of target trial emulation external control studies were conducted to replicate RCT findings of empagliflozin in heart failure adult populations and contextualize SAT findings of dabigatran in a pediatric population with venous thromboembolism (VTE). Contemporaneous external control data was derived from a historical trial, EMPEROR-Preserved (Aim 2), and real-world database, IQVIA PharMetrics® Plus for Academics (Aims 1 and 3). Imbalance in the measured baseline covariates between treatment groups were adjusted for using propensity score methods. For Aim 2, the PRO endpoint of change from baseline to Week 52 in the clinical summary score (CSS) of the Kansas City Cardiomyopathy Questionnaire (KCCQ) and 95% confidence interval (CI) was estimated using an inverse probability of treatment weighting (IPTW)-adjusted mixed model for repeated measures. Objective endpoints (e.g., hospitalization for heart failure & all-cause hospitalization (Aim 1), VTE recurrence & bleeding events (Aim 3), were analyzed using time-to-event analyses, with hazard ratios estimated using IPTW-Cox models, and cumulative incidence curves (Aim 1) and Kaplan-Meier curves (Aim 3) reported. Bias analyses were conducted to assess the sensitivity of the findings to the effect of unmeasured confounders.

Results: The IPTW-adjusted effect estimates for the PRO and objective endpoints were generally consistent with the target trials, with overlapping CIs and no statistically significant differences between the target trials and the ECA studies, leading to the same conclusion. Sensitivity analyses supported the robustness of the findings.

Conclusion: Replication of the evaluation of a PRO endpoint and objective outcomes was successful in the series of target trial emulation external control studies conducted. Challenges and lessons learned are used to formulate recommendations and propose best practices for external control applications. In general, continuous methodological refinement is essential to address biases and ensure that external control studies consistently produce valid and generalizable insights, ultimately strengthening their role in clinical research and regulatory decision-making.