One of the primary objectives of early clinical drug development is to determine the optimal dose and schedule of administration of the drug under study. If this objective is accomplished, subsequent clinical trials can truly evaluate the antitumor efficacy, tolerability, and safety of the new drug, and the appropriate assessment of benefit/risk can be made. A full understanding of the mechanisms not only of antitumor activity but also of side effects and toxicity is critical to select the optimal schedule of administration. In this regard, preclinical (in vitro and in vivo) studies are often helpful before clinical studies are initiated. However, no preclinical model is fully predictive of the outcome of human clinical trials. Therefore, while preclinical studies can point to potentially fruitful directions in clinical investigation, only after fairly substantial clinical experience does the medical community reach agreement and understanding of the optimal dose and schedule of administration of an agent. For some agents, these conclusions are reached early in the development of the drug (eg, cyclophosphamide, docetaxel). For others (eg, fluorouracil, cytosine arabinoside) the definition of optimal dose and schedule of administration is a never-ending story. The optimal duration of administration for trastuzumab is not known and is currently under active investigation.