With the average age of radiation therapy patients decreasing and the advent of more complex treatment options comes the concern that the incidences of radiation-induced cancer might increase in the future. The carcinogenic effects of radiation are not well understood for the entire dose range experienced in radiation therapy. Longer epidemiologic studies are needed to improve current risk models and reduce uncertainties of current risk model parameters. On the other hand, risk estimations are needed today to judge the risks versus benefits of modern radiation therapy techniques. This paper describes the current state-of-the-art in risk modeling for radiation-induced malignancies in radiation therapy, distinguishing between two volumes: first, the organs within the main radiation field receiving low or intermediate doses (typically between 0.1 and 50 Gy); and second, the organs far away from the treatment volume receiving low doses mainly due to scattered and secondary radiation (typically below 0.1 Gy). The dosimetry as well as the risk model formalisms are outlined. Furthermore, example calculations and results are presented for intensity-modulated photon therapy versus proton therapy.