Phase 1, Phase 2, and Phase 3 clinical trials: what FDA defines and what it requires to advance

"Phase 1," "Phase 2," and "Phase 3" are not marketing labels. They are regulatory categories defined in the U.S. Code of Federal Regulations, and each one carries a different objective, a different population size, and a different set of things FDA expects a sponsor to have in hand before the next phase can begin. This overview sets out the statutory definitions, then walks through what FDA has actually required, procedurally and substantively, to move a program from one phase to the next.

The three phases as FDA defines them

The phase definitions live in 21 CFR 312.21, "Phases of an investigation." The regulation is explicit that these phases can overlap and are a general framework rather than rigid gates, but the objectives and typical sizes are set out clearly ¹.

Phase 1 covers "the initial introduction of an investigational new drug into humans." These studies are typically closely monitored and may enroll patients or normal (healthy) volunteers. They are designed to determine the metabolism and pharmacologic actions of the drug in humans, characterize the side effects associated with increasing doses, and, if possible, gain early evidence of effectiveness. A Phase 1 program should generate enough pharmacokinetic and pharmacologic information to design well-controlled Phase 2 studies. The total enrollment is generally in the range of 20 to 80 subjects ¹.

Phase 2 covers "the controlled clinical studies conducted to evaluate the effectiveness of the drug for a particular indication or indications in patients with the disease or condition under study and to determine the common short-term side effects and risks associated with the drug." These studies are typically well controlled, closely monitored, and conducted in a relatively small number of patients, usually no more than several hundred ¹.

Phase 3 covers the "expanded controlled and uncontrolled trials," performed after preliminary evidence of effectiveness has been obtained. Phase 3 gathers the additional information on effectiveness and safety needed to evaluate the overall benefit-risk relationship of the drug and to provide an adequate basis for physician labeling. Phase 3 studies usually run from several hundred to several thousand subjects ¹.

It is worth noting that the international framework describes the same work by objective rather than by number. ICH E8 classifies clinical studies as human pharmacology, therapeutic exploratory, therapeutic confirmatory, and therapeutic use, and treats these as a categorization by objective across the product lifecycle rather than a strict sequence ¹³⁵. Human pharmacology corresponds broadly to Phase 1, therapeutic exploratory to Phase 2, therapeutic confirmatory to Phase 3, and therapeutic use to Phase 4 ¹³²¹⁴⁹¹⁵¹¹⁵³. The distinction matters most at Phase 3: ICH E9 defines a confirmatory trial as an adequately controlled trial in which the key hypothesis is stated in advance and evaluated, intended to provide firm evidence of efficacy or safety, whereas exploratory trials use more flexible designs and cannot by themselves serve as formal proof of efficacy ¹⁰⁸.

Before Phase 1: the IND must be in effect

No investigational drug may be given to a human subject in a Phase 1 study until the Investigational New Drug application (IND) is in effect ¹⁰⁷. An IND goes into effect either 30 days after FDA receives it, unless FDA notifies the sponsor that the studies are subject to a clinical hold, or earlier if FDA affirmatively tells the sponsor that clinical investigations may begin ¹⁰⁷. In parallel, the sponsor must have submitted the protocol to FDA and obtained Institutional Review Board (IRB) approval; the two can be satisfied in either order, but clinical investigation cannot start until both are met ¹⁴⁶.

What the IND must contain is graded to the phase of investigation. Under 21 CFR 312.23, the application must include an introductory statement and general investigational plan, protocols, chemistry/manufacturing/controls (CMC) information, and pharmacology and toxicology data adequate to justify that the proposed studies are reasonably safe ⁶⁵⁷⁹. FDA builds phase-appropriateness directly into these requirements:

Protocols. For Phase 1, protocols may be less detailed and more flexible, outlining the investigation and specifying the critical safety elements. Detailed protocols are required for Phase 2 and Phase 3 ⁵.
CMC. The information must be sufficient for each phase to assure identity, quality, purity, and strength. In initial Phase 1 the emphasis is on identification and control of raw materials and the drug substance; final specifications are not expected until the end of development, and the sponsor submits expanded CMC information by amendment as development proceeds ⁴.
Nonclinical safety. The pharmacology and toxicology package must be adequate to support the specific studies proposed, and additional safety information is submitted by amendment as the program advances ⁷⁹.

FDA guidance reinforces that this is a deliberately graded, flexible standard. The agency's regulations allow "a great deal of flexibility" in the amount of data submitted for initial human studies, and the amount of information needed varies with the phase, the proposed duration of the studies, the dosage form, and the information already available ⁴²⁴⁰. For the earliest work, the Exploratory IND guidance describes studies conducted very early in Phase 1, with very limited human exposure and no therapeutic or diagnostic intent, supported by a reduced preclinical package tailored to the study's scope; such studies are not designed to establish a maximum tolerated dose ³⁹³⁸⁴¹⁴⁴.

Progression is gated by safety, not by a "pass" certificate

FDA does not issue a formal approval to move from Phase 1 to Phase 2 or from Phase 2 to Phase 3. Legally, a sponsor may proceed once the relevant protocols are submitted under the IND and IRB-approved, and provided FDA has not imposed a clinical hold. The clinical hold is the primary mechanism by which FDA stops or prevents progression.

Under 21 CFR 312.42, a clinical hold is an FDA order to delay a proposed study or suspend an ongoing one, and it can apply to one or more studies under an IND ¹⁰. The grounds are tiered by phase:

For Phase 1, FDA may hold a study if subjects would be exposed to unreasonable and significant risk, the investigators are unqualified, the investigator's brochure is misleading or materially incomplete, the IND lacks sufficient information to assess risk, or (for certain life-threatening diseases) subjects of reproductive potential are excluded over reproductive-toxicity concerns without adequate justification ¹⁰.
For Phase 2 or Phase 3, all of the Phase 1 grounds apply, plus an additional ground: that the protocol is clearly deficient in design to meet its stated objectives ¹⁰.

The practical effect is significant. If a proposed study is on hold, the drug may not be administered; if an ongoing study is held, no new subjects may be enrolled and patients already on the drug should be taken off unless FDA specifically permits continuation for safety reasons ¹⁰. A held investigation may resume only after FDA notifies the sponsor it may proceed, and only after the cited deficiencies are corrected. Even a complete response from the sponsor does not authorize restarting until FDA affirmatively lifts the hold ¹².

So the "requirement" to move between phases is best understood negatively: the sponsor must have generated enough safety and design information that FDA has no basis to impose a hold on the next set of studies, and the additional-design ground at Phase 2/3 means the later protocols must be scientifically adequate to meet their objectives, not merely safe.

The milestone meetings: where phase transitions are actually negotiated

In practice, phase transitions are managed through a structured sequence of FDA-sponsor meetings. The regulation at 21 CFR 312.47 encourages these meetings and singles out two by name. The end-of-Phase 2 meeting is used to determine whether it is safe to proceed to Phase 3, to evaluate the Phase 3 plan and protocols, to assess the adequacy of current studies and pediatric plans, and to identify any additional information needed to support a marketing application; FDA states these should occur before major Phase 3 commitments but should not delay the Phase 2-to-Phase 3 transition ². The pre-NDA meeting, held near the end of Phase 3, is used to identify major unresolved problems, confirm the studies intended to support effectiveness, review the status of pediatric studies, and agree on the technical content, statistical methods, and format of the planned application ²³.

For serious or life-threatening illnesses, 21 CFR 312.82 adds earlier formal touchpoints. Pre-IND meetings allow the sponsor to reach agreement on the design of the animal studies needed to start human testing and to discuss Phase 1 scope. End-of-Phase 1 meetings allow agreement on the design of the Phase 2 controlled trials intended to support approvability; the documentation procedures for end-of-Phase 2 conferences apply to these meetings as well ¹³.

FDA's guidance places these into the PDUFA meeting-type framework. The milestone meetings, pre-IND, end-of-Phase 1, end-of-Phase 2, and pre-NDA/pre-BLA, are generally Type B meetings ⁵⁰⁶¹. Type A meetings address stalled programs or important safety issues; Type C is a catch-all for other development or review meetings; and Type D is a focused meeting on a narrow set of issues used to give timely feedback at key decision points ⁶¹⁵¹⁵⁶. The agency's stated purposes track the phase transitions directly:

Pre-IND: align on mechanism of action and possible study designs and open the development dialogue, which FDA notes can help avoid clinical-hold issues and support a complete IND ⁵⁰⁵⁹.
End-of-Phase 1 (EOP1): reach agreement on the design of the Phase 2 controlled trials and discuss the development program, including pediatric plans ⁵⁰.
End-of-Phase 2 (EOP2): reach agreement on Phase 3 design, including benefit-risk considerations, patient populations, endpoints, trial duration, dose-response, and trial size ⁵⁰⁶³.
Pre-NDA/pre-BLA: align on the content and format of the marketing application and, where relevant, early submission of portions of the application ⁶¹⁶².

The relevant guidance documents are "Best Practices for Communication Between IND Sponsors and FDA During Drug Development" and "Formal Meetings Between the FDA and Sponsors or Applicants of PDUFA Products" ⁵⁰⁵¹.

The substantive bar for the Phase 2-to-Phase 3 transition

The point where FDA's expectations become most demanding is the move into Phase 3, because Phase 3 trials are meant to be the confirmatory evidence supporting the label. Two threads of FDA guidance define what a sponsor is expected to bring to that transition.

First, dose selection and dose-response. FDA considers dose-response data desirable for almost all new molecular entities, obtained from sound, well-controlled studies ⁷⁷⁸⁰. The agency created a dedicated interaction, the end-of-Phase 2A (EOP2A) meeting, to occur after Phase 1 and the first exposure-response trials in patients but before Phase 2B and Phase 3, specifically to discuss trial design, modeling and simulation, and dose estimation and selection, with the objective of choosing dosing regimens for later trials and designing informative dose-response studies ⁷⁵⁷⁶⁷⁹. FDA's exposure-response guidance similarly frames Phase 1 and Phase 2 exposure-response work as the basis for proof of concept and for defining dosing regimens carried into confirmatory trials ⁸⁹⁹². The governing guidance documents are ICH E4, "Dose-Response Information to Support Drug Registration," "Exposure-Response Relationships," and "End-of-Phase 2A Meetings" ⁷⁷⁸⁹⁷⁵.

Second, the quantity and quality of effectiveness evidence that Phase 3 must be designed to deliver. The statutory standard is substantial evidence from adequate and well-controlled investigations. FDA guidance has historically interpreted this to mean effectiveness should ordinarily be supported by more than one adequate and well-controlled trial conducted by independent investigators, with a single particularly persuasive study sufficient only in unusual circumstances ⁸⁶⁹⁶. More recent guidance recognizes that either two adequate and well-controlled trials, or one adequate and well-controlled trial plus confirmatory evidence, may establish substantial evidence in appropriate contexts ⁹⁵. ICH E9's expectations for a confirmatory trial reinforce what "adequate and well-controlled" means at Phase 3: the design and major statistical aspects must be justified in the protocol, the primary analysis and hypotheses must be prespecified, multiplicity must be addressed in the analysis plan, the trial should address only a limited number of questions, and the results should be robust and generalizable to the intended population ¹⁰⁸¹¹⁷¹⁰⁹.

In short, the end-of-Phase 2 milestone is where FDA and the sponsor confirm that the dose is justified, the endpoints and population are agreed, and the Phase 3 protocols are adequate to generate the confirmatory evidence a marketing application will require.

How to read the phase framework as a whole

The three-phase structure is a graded escalation of both exposure and evidentiary weight. Phase 1 establishes that the drug can be given to humans safely and characterizes its pharmacology in tens of subjects ¹. Phase 2 tests effectiveness for a defined indication and characterizes short-term risk in up to several hundred patients ¹. Phase 3 generates the confirmatory, adequately controlled evidence, in hundreds to thousands of subjects, needed to weigh benefit against risk and write the label ¹.

What FDA "requires" to advance is layered rather than a single sign-off: the IND must be in effect and remain free of a clinical hold ¹⁰⁷¹⁰; the protocols for each new phase must be submitted and IRB-approved ¹⁴; the CMC and nonclinical packages must be expanded to remain phase-appropriate ⁴⁹; and, at each milestone, the sponsor is expected to reach documented agreement with FDA on the design of the next phase, most consequentially at the end-of-Phase 2 meeting that gates entry into confirmatory Phase 3 trials ²⁵⁰⁶³. The regulations set the floor; the milestone meetings are where the substantive expectations for the next phase are actually fixed.