The predictions are based on an informative prior, called surrogate prior, obtained from the data of past trials and external evidence on one or several surrogate endpoints. This prior could be combined with data on the clinical endpoint of interest, if available, in a classical Bayesian framework. Two methods are proposed to address a potential discordance between the surrogate prior and the data on the final endpoint. We investigated the patterns of behaviour of the predictions in a comprehensive simulation study, and we present an application to a drug development in Multiple Sclerosis. The proposed methodology is expected to support decision-making in many different situations, since the use of predictive markers is critical to accelerate drug developments and to select promising drug candidates, better and earlier.