Radiation therapy is an important modality in cancer treatment. To find a good treatment plan, optimization models and methods are typically used. Within the optimization models, several conflicting objectives such as sparing of healthy Organs at Risk (OAR) and eradicating the tumor, are pursued simultaneously. Besides the inherent complexity of some of the clinically-important planning criteria, particularly, the so-called dose-volume requirements for the OAR, the treatment optimization process is further complicated by the presence of uncertainties. In this talk we will briefly survey the optimization approaches that can handle the uncertainties by robustifying the underlying model, and discuss several alternatives to approximate the exact dose-volume requirements in a computationally-tractable fashion.