Modulating immune responses with monoclonal antibodies (mAbs) that target immune molecules has become a promising therapeutic strategy and is under investigation for the treatment of cancer and (auto)-immune diseases. A major hurdle to the development and early clinical investigation of many immunomodulatory mAbs is the inherent risk of adverse immune-mediated drug reactions in humans, such as cytokine storms, autoimmunity, and immunosuppression. Dose selection for first-in-human (FIH) clinical trials involving immunomodulatory mAbs, and mAbs in general, is based on specifically designed preclinical safety studies, primarily in nonhuman primates (NHPs), and on mechanistic ex vivo investigations. Dose selection in such trials is challenging for a number of reasons related to safety. In this context, safety-relevant differences between NHP and human immune systems, species selection/qualification and preclinical study design considerations, the receptor occupancy model and its calculation, the minimal anticipated biological effect level (MABEL) and its use in the selection of a safe starting dose in humans, microdosing and the impact of immunogenicity on safety assessment of mAbs, and safety-relevant formulation properties of therapeutic mAbs are critically reviewed. In addition, the current regulatory requirements are presented and discussed to demonstrate how the TeGenero TGN1412 case is leading to increased regulatory scrutiny regarding dose selection for FIH clinical trials.