Usage Examples
- Both Phase 3 trials hit the primary endpoint, so we are filing on two adequate and well-controlled studies rather than one plus confirmatory evidence.
- The Phase 3 protocol and the statistical analysis plan were locked before unblinding; nothing in the primary analysis moved after that.
- That subgroup result came out of a Phase 3 trial, but it was not prespecified, so it stays out of the label discussion.
What is Phase 3 Clinical Trial?
Phase 3 clinical trials are expanded controlled studies run after preliminary evidence of effectiveness, sized at several hundred to several thousand subjects to establish the benefit-risk relationship and supply the substantial evidence supporting approval and physician labeling.
Phase 3 clinical trials exist because a Phase 2 signal is not proof. The evidentiary problem FDA has to solve is separating a drug's effect from spontaneous change in the course of the disease, the placebo effect, and biased observation. Phase 3 trials solve it at scale, producing the effectiveness and safety data needed to judge the overall benefit-risk relationship and write physician labeling.
Phase 3 clinical trials cover expanded controlled and uncontrolled studies run after preliminary evidence of effectiveness exists, usually enrolling several hundred to several thousand subjects. The boundary cuts on both sides: dose-ranging and first efficacy evidence belong to Phase 2, postmarketing questions to Phase 4. Phase 3 is also not a synonym for pivotal, which names regulatory role rather than development stage.
Phase 3 clinical trials are run against a protocol and analysis plan fixed before unblinding, because a comparison designed after the data are visible is not a valid comparison. Phase 3 conduct carries live safety duties throughout, including the 7-calendar-day clock for unexpected fatal or life-threatening suspected adverse reactions. The resulting clinical study reports then form the efficacy core of Module 5 in the eCTD.
Not to be confused with
- Phase 2 trial
- Phase 2 evaluates effectiveness in a relatively small population, usually no more than several hundred subjects, and is where dose is selected. Phase 3 may only begin once that preliminary evidence of effectiveness exists.
- Pivotal trial
- pivotal is a regulatory role, meaning the trial the approval decision rests on. A well-controlled Phase 2 study can be pivotal, and a Phase 3 study can be merely supportive.
- Adequate and well-controlled study
- a legal evidence standard set by 21 CFR 314.126 that a trial must meet to count toward approval. Running a study in Phase 3 does not make it adequate and well-controlled.
- Phase 4 study
- conducted after approval, typically as a postmarketing requirement or commitment. Phase 4 refines what the label already claims; Phase 3 is what earns the label.
Phase 3 obligations come from the IND regulations, the approval standard, and the statute behind it.
What you must do
- 1Begin Phase 3 only after preliminary evidence suggesting effectiveness has been obtained, and design the trials to gather the additional effectiveness and safety information needed to evaluate the overall benefit-risk relationship and provide an adequate basis for physician labeling21 CFR 312.21(c)
- 2State the objectives of the investigation and the proposed methods of analysis in the protocol before the study runs, and carry the same statement into the report of results21 CFR 314.126(b)(1)
- 3Build each trial on a design that permits a valid comparison with a control, so the drug effect is quantified rather than inferred21 CFR 314.126(b)(2)
- 4Support the marketing application with substantial evidence from adequate and well-controlled investigations, recognizing that one such investigation plus confirmatory evidence obtained before or after it can meet the standard21 U.S.C. 355(d)
- 5Notify FDA of any unexpected fatal or life-threatening suspected adverse reaction no later than 7 calendar days after initial receipt of the information, and file written IND safety reports for serious and unexpected suspected adverse reactions within 15 calendar days21 CFR 312.32(c)
Common mistakes
Treating the Phase 3 label as proof the trial is adequate and well-controlled
the phase describes where the program is; 21 CFR 314.126 sets a separate evidence standard. A 2,000-patient trial with a soft primary endpoint or an unblinded assessor can be discounted at review, and the cost of that discovery is measured in years, not in amendments.
Reopening the statistical analysis plan after the data are visible
redefining the primary endpoint, the analysis population, or the multiplicity strategy post hoc converts a confirmatory result into a hypothesis-generating one. The reviewer's question is not whether the number is real, it is whether the comparison was specified before anyone could see which specification would win.
Assuming two replicate Phase 3 trials are legally mandatory
the statute allows one adequate and well-controlled investigation plus confirmatory evidence. Sponsors who never test that route with the review division fund a second pivotal trial, often the largest line in the development budget, that the agency might not have required. The inverse error is worse: planning a single-trial filing without agency alignment and finding out at the pre-NDA meeting.
When This Matters
- Both Phase 3 trials hit the primary endpoint, so we are filing on two adequate and well-controlled studies rather than one plus confirmatory evidence.
- The Phase 3 protocol and the statistical analysis plan were locked before unblinding; nothing in the primary analysis moved after that.
- That subgroup result came out of a Phase 3 trial, but it was not prespecified, so it stays out of the label discussion.
Frequently Asked Questions
Phase 3 trials usually enroll from several hundred to several thousand subjects under 21 CFR 312.21(c). No fixed minimum exists. The number is driven by the statistical power required to detect the expected effect on the primary endpoint, and by the size of the safety database the review division expects at filing.
Related Use Cases
- Pharma Use Cases
Cut NDA and sNDA prep time by 60% with AI-assisted drafting and automated readiness checks
- Biotech Use Cases
Compress IND prep from 8-12 weeks to under 3 weeks with AI-assisted drafting and validation
- Clinical Workflows
Track clinical trial regulations across global markets
- Regulatory Affairs Workflows
Cut regulatory intelligence tracking from 10+ hours/week to automated, real-time alerts
