The well-established standard approach to the development of systemic antineoplastic therapy begins with phase 1 trials, when new agents or novel combination drug strategies are initially introduced into the clinical setting. 1 Although the primary purpose of phase 1 trials is the determination of safety, these early clinical investigative efforts are also associated with several important secondary endpoints that are essential to future drug development, including: 1) carefully defining the overall toxicity profile and exploring strategies to mitigate serious side effects, 2) establishing both the dose and schedule for subsequent studies, 3) understanding the agent's pharmacokinetic or pharmacodynamic profile, 4) gaining preliminary evidence of efficacy, and 5) learning of the possible relations between proposed biomarkers and response or toxicity. Depending on the specific circumstances, phase 1 trials will be followed by 1 or more phase 2 efficacy and/or phase 3 randomized comparative studies.Unfortunately, this observer has noted a recent trend in which some oncology investigators have suggested bypassing formal phase 1 testing of presumably theoretically attractive new/novel antineoplastic drug combinations (cytotoxic and noncytotoxic), with the initial evaluation of the regimen being undertaken in the phase 2, or even phase 3, setting. 2 This commentary was written to challenge the need for, and wisdom of, this approach and to highlight the potential danger to research subjects participating in later stage trials that are not supported by rigorous phase 1 data.What follows are several possible justifications for considering early phase studies unnecessary before embarking on phase 2 to 3 efficacy trials, and a brief rebuttal to each suggested rationale for bypassing the conduct of well-designed and conducted phase 1 trials.