In a typical RAR design, initial randomization is often equal, giving each treatment arm an equal chance of being selected. As data on treatment responses accumulate, statistical algorithms update the probabilities of assignment to each treatment arm. These updates are based on interim analyses that assess the relative efficacy and safety of the treatments. The adaptive algorithm may use methods such as the Bayesian approach or frequentist techniques to adjust the allocation ratios.
