Logistic regression estimates of odds ratios for a cure on Regimen A compared to standard regimen by duration of treatment with Regimen A in a single simulated trial using the proposed trial design. Odds ratio estimates are shown by the solid circles connected with the solid line, while the upper and lower 95% CI bounds for these estimates are shown with solid circles connected by dashed lines. The non-inferiority boundary of 0.63 is shown by the horizontal line at that level on the vertical axis. The vertical axis is presented on the log scale. CI, confidence interval.