In randomized clinical trials, the accumulated data are often monitored through interim monitoring. Predicting the future course of the trial based on the accumulated data from trial itself may provide a useful tool for monitoring the trial. This roundtable will provide an overview of approaches for predicting various outcomes including: time to reach landmark events, treatment effect, optimal combination of enrollment period and follow-up length, conditional power and predictive power etc. We seek to discuss the barriers of using prediction in the real clinical trials.