Calibration of the IQVIA Core Diabetes Model to the stroke outcomes from the SUSTAIN 6 cardiovascular outcomes trial of once-weekly semaglutide

Aims In the SUSTAIN 6 cardiovascular outcomes trial, once-weekly semaglutide was associated with a statistically significant reduction in major adverse cardiovascular events compared with placebo. To date, no studies have assessed how accurately existing diabetes models predict the outcomes observed in SUSTAIN 6. The aims of this analysis were to investigate the performance of the IQVIA Core Diabetes Model when used to predict the SUSTAIN 6 trial outcomes, to calibrate the model such that projected outcomes reflected observed outcomes, and to examine the impact of calibration on the cost-effectiveness of once-weekly semaglutide from a UK healthcare payer perspective. Methods The IQVIA Core Diabetes Model was calibrated to ensure that the projected non-fatal stroke event rates reflected the non-fatal stroke event rates observed in SUSTAIN 6 over a two-year time horizon. Cost-effectiveness analyses of once-weekly semaglutide versus placebo plus standard of care were conducted over a lifetime horizon using the uncalibrated and calibrated models to assess the impact on cost-effectiveness outcomes. Results To replicate the non-fatal stroke event rate in SUSTAIN 6, calibration of the model through the application of relative risks for stroke of 1.07 and 1.65 with once-weekly semaglutide and placebo, respectively, was required. In the long-term cost-effectiveness analysis, the uncalibrated model projected an incremental cost-effectiveness ratio for once-weekly semaglutide versus placebo plus standard of care of GBP 22,262 per quality-adjusted life year (QALY) gained, which fell to GBP 17,594 per QALY gained when the calibrated model was used. Conclusions The requirement for calibration to replicate the outcomes observed in SUSTAIN 6 suggests that the reductions in risk of cardiovascular complications observed with once-weekly semaglutide cannot be solely explained by differences in conventional risk factors. Accurate estimation of the risk of diabetes-related complications using methods such as calibration is important to ensure accurate cost-effectiveness analyses are conducted.