In this paper, we discuss the opportunities that Nonlinear Model Predictive Control (NMPC) offers in the area of pharmacological therapy as an effective way for achieving optimization goals such as drug usage reduction while respecting safety and therapeutic constraints imposed by physicians. We first review a general formulation of NMPC and discuss some issues associated with its application to biomedical processes. Next, we present two simulated biomedical applications of NMPC: the first one deals with the optimization of anti-retroviral therapy in HIV and the second one discusses the optimization of multiple injection insulin therapy in Type 1 diabetics. Both these examples serve to highlight the main advantages of NMPC.