Abstract
INTRODUCTION: Semaglutide, a glucagon-like peptide 1 receptor agonist (GLP1RA), is available in both parenteral and oral preparations. Studies of injectable preparations have convincingly demonstrated its beneficial effect on major adverse cardiac events (MACE). This predictive analysis was undertaken to forecast early termination of the SOUL trial (oral semaglutide) as well as the primary events. METHODS: SOUL is a multicenter, double-blind, placebo-controlled randomized controlled trial (RCT) evaluating the reduction in MACE associated with oral semaglutide versus placebo in patients with type 2 diabetes (T2D) and cardiovascular (CV) disease. A sample of 9642 participants will be followed for 5 years and 5 months. A random-effects model meta-analysis, pooling hazard ratios from previous RCTs, was conducted using R software to inform the predictive model. The background CV event rates from the placebo arms of previous RCTs with semaglutide were matched with the pre-adjudicated assumptions of the SOUL trial to create the predictive model. The truncated trial duration, MACE, and its individual components in the intervention and placebo arms were estimated. The predicted difference between the two groups was estimated using the chi-squared test. RESULTS: A pooled analysis of 10,013 patients revealed a significant reduction in the number of MACEs associated with semaglutide (HR 0.79, 95% CI 0.69-0.91). Predictive analysis indicated that 1225 events would be achieved by 3.78 years, suggesting premature termination. CONCLUSION: The mathematical model based on the meta-analysis predicts that the SOUL study on oral semaglutide will be terminated early, with oral semaglutide showing benefits in terms of MACE compared to placebo. If the SOUL study corroborates the findings of this model, it may not only form the basis for the calculation of power but also define the duration of such studies, reducing costs and easing the process of designing cardiovascular outcome trials (CVOTs). PROTOCOL REGISTRATION: INPLASY202460061.