FDA's Early-Phase Solid Tumor Guidance Narrows The First-In-Human Objective
By Morgan Kohler, Executive Editor

The FDA has issued draft guidance telling sponsors how it wants early-phase oncology protocols written, and the document is more prescriptive about what a first-in-human trial is for than many running protocols assume. The Federal Register notice of availability published Nov. 12, 2026, under docket FDA-2026-D-13432. Comments close Jan. 11, 2027.
The guidance, Early-Phase IND Protocol Recommendations for Solid Tumors, was prepared by the FDA’s Oncology Center of Excellence with the Center for Drug Evaluation and Research and the Center for Biologics Evaluation and Research. It covers first-in-human protocols and protocols for drugs and biologics with limited human experience. It is specific to solid tumors and excludes hematologic malignancies.
On dose escalation, the FDA states that the primary objective of a first-in-human escalation trial should be safety and tolerability. Everything else (dose identification, pharmacokinetics, pharmacodynamics, immunogenicity, biomarker analyses, dose-exposure-response relationships, food effects, drug-drug interactions, and preliminary activity measures such as response rates) is framed as secondary. On eligibility, the agency warns that overly restrictive criteria may impede recruitment, limit access to early-phase trials, and produce results that do not apply widely to the intended patient population. Protocols should generally avoid upper age limits, and a proposed limit requires sufficient justification.
Two provisions reach further back into the pipeline than their placement in a clinical guidance suggests. On assessing activity, the FDA states that single-arm trials do not adequately characterize overall survival or progression-free survival, and that a randomized study is generally necessary to evaluate time-to-event endpoints. On patient selection, the agency notes that an in vitro companion diagnostic may need to be developed and validated if biomarker-based selection is essential for safe and effective use, and advises sponsors to consult the Center for Devices and Radiological Health (CDRH) early.
The guidance is part of Operation TrialBlazer, a U.S. Department of Health and Human Services initiative launched in mid-June as a pilot program to speed drug development from the IND stage through late-stage trials, with the stated aim of reversing the shift of clinical development away from the United States.
What This Means For Drug Discovery
The companion diagnostic language is the part discovery teams should be sure to pay attention to. A biomarker hypothesis that originates in target validation and travels into the clinic as a patient-selection criterion is, by the FDA’s description here, no longer an exploratory readout. It is a diagnostic that may require analytical validation on a CDRH timeline that does not compress to fit an IND schedule.
That changes what a target validation package has to settle. The question is not only whether a biomarker associates with response in preclinical models. It is whether the biomarker is intended to select patients or to characterize them, because those two intentions carry different evidence standards and different downstream obligations. Nomination committees that treat the distinction informally, or defer it to translational medicine after candidate selection, are deferring a decision that now has a regulatory clock attached.
The randomization language compounds the effect. If single-arm early-phase data will not characterize survival endpoints, randomization moves earlier in development, and cost moves earlier with it. The practical consequence is that the preclinical evidence of differentiation has to be stronger at the point of candidate nomination, because there is less room to advance a program on the expectation that a signal will emerge during dose expansion. The FDA reinforces this from the other direction in its expansion cohort recommendations, telling sponsors to minimize the number of subjects exposed to potentially ineffective or significantly toxic drugs. It also asks sponsors to justify the maximum sample size in the statistical analysis plan.
There is a tension in the document worth naming. The FDA is discouraging restrictive eligibility criteria in the same guidance in which it describes the conditions under which biomarker-based selection requires a validated companion diagnostic. Both positions are coherent — the agency is distinguishing between criteria that narrow a population without scientific justification and criteria that are essential to safe and effective use — but the burden of drawing that line falls on the sponsor, and it falls early. A tightly enriched population improves the odds of detecting activity in the small cohorts the FDA wants and commits the program to diagnostic development. A broad population avoids that commitment and dilutes the signal.
None of this is a departure from where the FDA has been heading. The agency’s preference for randomized dose-finding and for rational dose optimization in oncology has been explicit since the 2023 dose optimization draft guidance, which finalized in 2024. What is new is the consolidation of those expectations into a single early-phase document, with a comment period that closes in January. It arrives as the rest of the agency's output is pushing model-based and human-relevant evidence earlier into the same programs: The June draft guidance on MABEL dose selection using quantitative systems pharmacology and the September rule recognizing new approach methodologies as nonclinical tests.
Read together, those documents describe a development path in which the first-in-human dose comes from a mechanistic model rather than animal toxicology, the nonclinical package may rest partly on nonanimal methods, and the first clinical trial is expected to answer a narrower question than sponsors have historically asked of it. Each piece reduces one kind of uncertainty and relocates another. The uncertainty accumulates, in every case, at the discovery end.
The Discovery Question
If your first clinical trial can only tell you whether the drug is tolerable, what does your preclinical package have to prove before you are willing to spend it?
Related Drug Discovery Online Editorials
- Are NAMs Validation Centers Really Needed To Drive Greater NAMs Adoption? Scientists Are Wondering
- Years Before Clinical Trials: The Decisions That Shape The Future Of Academic Cancer Discoveries
- Why Immunotherapy Has Had Limited Success In Epithelial Ovarian Cancer
About The Editor
Morgan Kohler is the executive editor of guest expert articles for Drug Discovery Online, working with industry experts to develop insightful and informative pieces about drug discovery and preclinical drug development science, regulatory strategy, trends, analyses, and commentary.
Morgan holds a master’s degree in professional writing and web content development and has worked in the publishing industry since 2018. Before working for Drug Discovery Online, she guided and edited industry-expert-bylined content on topics related to radiology and economics.