Research · 5 min read
What Phase 1, Phase 2 and Phase 3 Actually Mean
The three phases are defined in one short regulation. It describes what each phase is for and roughly how many people are in it — and it says the phases may overlap, which is the part marketing copy tends to leave out.
Key takeaways
- The three phases are defined at 21 CFR 312.21, which states that although generally sequential, they may overlap.
- Phase 1 is the initial introduction into humans, generally 20 to 80 subjects; phase 2 is controlled studies in usually no more than several hundred patients; phase 3 is expanded controlled and uncontrolled trials in several hundred to several thousand.
- The regulation's own phase 3 definition includes uncontrolled trials, so a phase 3 label is not proof that a comparison was made.
- FDA's draft weight-reduction guidance recommends far larger numbers for this class: 3,000 subjects on drug and no fewer than 1,500 on placebo for at least a year at the maintenance dosage.
- The federal registry statute excludes phase 1 studies from its definition of an applicable drug clinical trial, borrowing the meaning of phase 1 directly from 21 CFR 312.21.
- Sponsors must report dropouts, deaths and enrollment by age group, gender and race in the annual report required by 21 CFR 312.33.
Answer first: the phases are a regulation, not a marketing ladder
The words phase 1, phase 2 and phase 3 are not industry slang. They are defined at 21 CFR 312.21, a section headed "Phases of an investigation," which opens by saying that the clinical investigation of a previously untested drug is generally divided into three phases, and adds immediately that although in general the phases are conducted sequentially, they may overlap.
That second clause matters. A company can be running phase 2 work on one question while phase 3 work on another is already underway. A statement that a drug is "in phase 3" describes a stage of one investigation, not a position in a queue that everything else has cleared.
The same part of the regulations explains why the phases exist as separate concepts at all. Under 21 CFR 312.22, FDA's primary objectives in reviewing an investigational new drug application are, in all phases, to assure the safety and rights of subjects, and in phases 2 and 3, to help assure that the quality of the scientific evaluation is adequate to permit an evaluation of the drug's effectiveness and safety. Reviews of phase 1 submissions focus on assessing safety. Reviews of phase 2 and 3 submissions also assess scientific quality and the likelihood the investigations will yield data capable of meeting statutory standards for marketing approval.
What each phase is for, in the regulation's own words
Phase 1 includes the initial introduction of an investigational new drug into humans. The regulation says these studies are typically closely monitored and may be conducted in patients or in normal volunteer subjects, and that they are designed to determine the metabolism and pharmacologic actions of the drug in humans, the side effects associated with increasing doses, and, if possible, to gain early evidence on effectiveness. The stated purpose is to obtain enough information about pharmacokinetics and pharmacological effects to permit the design of well-controlled, scientifically valid phase 2 studies. Phase 1 also covers studies of drug metabolism, structure-activity relationships and mechanism of action.
Phase 2 is described as 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. The regulation says phase 2 studies are typically well controlled, closely monitored, and conducted in a relatively small number of patients.
Phase 3 is where the wording changes in a way worth noticing. Phase 3 studies are "expanded controlled and uncontrolled trials." They are performed after preliminary evidence suggesting effectiveness has been obtained, and are intended to gather the additional information about effectiveness and safety needed to evaluate the overall benefit-risk relationship and to provide an adequate basis for physician labeling. So the phase 3 bucket is not synonymous with rigorous comparison. It contains both.
The numbers in the regulation are ranges, and they are old
Section 312.21 attaches rough sizes to each phase. The total number of subjects and patients in phase 1 varies with the drug but is generally in the range of 20 to 80. Phase 2 usually involves no more than several hundred subjects. Phase 3 usually includes from several hundred to several thousand subjects.
Those are descriptive ranges, not requirements, and a specific development program can sit well outside them. FDA's draft guidance on developing weight-reduction drugs sets out a much larger expectation for this drug class: its general recommendation is a safety database of 3,000 subjects randomized to the investigational drug within the to-be-recommended dosage range, and no fewer than 1,500 subjects randomized to placebo, for at least one year of treatment at the maintenance dosage.
The guidance explains the reasoning rather than leaving it implicit. It says the size of the safety database for a weight-reduction program should exceed the exposures in the international guidance that normally governs long-term treatment of non-life-threatening conditions, because substantial patient exposure can be expected after approval and because patients with obesity or overweight have a significant background rate of morbidity and mortality. It adds that the recommended sample size would provide 80 percent power to detect, with 95 percent confidence, an approximately 50 percent increase in the incidence of an adverse event that occurs at a rate of 3 percent in the placebo group.
The last sentence is the useful one for reading any trial. A number of participants is chosen to make a specific size of difference detectable. It is not a claim that everything smaller has been ruled out.
Where the phase label carries legal weight
The phase label is not only descriptive. It decides at least two things a reader can check.
The first is public registration. The federal trial registry statute at 42 U.S.C. 282(j) defines an "applicable drug clinical trial" as a controlled clinical investigation, other than a phase 1 clinical investigation, of a drug subject to the drug approval provisions or the biologics provisions. The statute then says that for this purpose the term phase 1 has the meaning given in section 312.21 of title 21 of the Code of Federal Regulations. So the regulation above is the operative definition, and the practical consequence is that a phase 1 study can be entirely absent from the public registry without anything being wrong.
The second is how a sponsor must report. Under 21 CFR 312.33, a sponsor must submit an annual report within 60 days of the anniversary date the application went into effect. That report has to include, for each study, the total number of subjects initially planned, the number entered to date tabulated by age group, gender and race, the number who completed as planned, and the number who dropped out for any reason. It must also include a list of subjects who died during participation with the cause of death for each, and a list of subjects who dropped out in association with any adverse experience whether or not thought to be drug related.
Those reporting duties are why an accurate picture of a development program exists somewhere even before any results are published. They are also a reminder that dropouts are a tracked quantity in the regulatory record, not an afterthought.
How to read a phase claim on a commercial page
A phase label describes the design and stage of a study. It does not describe a conclusion. A drug can be in phase 3 and fail. A drug can complete phase 2 with an encouraging result and never reach a marketing application. The regulation makes no promise about outcome at any stage, and neither should a page quoting it.
There is one more structural point. Under 21 CFR 312.20 a sponsor must submit an investigational new drug application before beginning a covered clinical investigation, and may not begin until that application is in effect. Everything described on this page therefore belongs to the world of investigational drugs under a federal application. A compounded preparation dispensed by a pharmacy is not an approved drug and is not being developed through this process, so phase language borrowed onto a compounded product's marketing is describing something that did not happen to that product.
The questions worth asking when a phase appears in an advertisement are narrow and answerable. Which phase, for which indication, in which population, and is there a registry record. The first three are what the regulation is about. The fourth is checkable in a public database for any trial that is not phase 1.
Sources
- 21 CFR 312.21 — Phases of an investigationThat the clinical investigation of a previously untested drug is generally divided into three phases which may overlap; the description and stated purpose of phase 1, phase 2 and phase 3; the subject-number ranges of 20 to 80, no more than several hundred, and several hundred to several thousand; and the phrase describing phase 3 as expanded controlled and uncontrolled trials.
- 21 CFR 312.20 — Requirement for an IND, and 21 CFR 312.22 — General principles of the IND submissionThat a sponsor may not begin a covered clinical investigation until an investigational new drug application is in effect, and that FDA's review objectives are safety and subject rights in all phases plus, in phases 2 and 3, an assessment of scientific quality and of the likelihood the investigations will yield data capable of meeting statutory standards for marketing approval.
- 21 CFR 312.33 — Annual reportsThe 60-day annual reporting deadline and the required contents, including planned and entered subject numbers tabulated by age group, gender and race, the number completing as planned, the number dropping out for any reason, the list of subjects who died with cause of death, and the list of subjects who dropped out in association with an adverse experience.
- 42 U.S.C. 282 — Director of National Institutes of Health, subsection (j), Expanded clinical trial registry data bankThe statutory definition of an applicable drug clinical trial as a controlled clinical investigation other than a phase 1 clinical investigation, and the statute's adoption of the meaning of phase 1 given in 21 CFR 312.21.
- Obesity and Overweight: Developing Drugs and Biological Products for Weight Reduction — Guidance for Industry (Draft Guidance)The recommended safety database of 3,000 subjects randomized to the investigational drug and no fewer than 1,500 randomized to placebo for at least one year at the maintenance dosage, the reasoning that expected postapproval exposure and background morbidity make the ordinary standard inapplicable, and the statement that this sample size gives 80 percent power to detect, with 95 percent confidence, an approximately 50 percent increase in an adverse event occurring at 3 percent in the placebo group.
Frequently asked questions
Where do the three phases come from?
From 21 CFR 312.21, a section headed "Phases of an investigation." It says the clinical investigation of a previously untested drug is generally divided into three phases and that, although generally sequential, they may overlap. The same part sets out FDA's review objectives for each phase at 21 CFR 312.22.
How many people are in each phase?
The regulation gives ranges rather than requirements. Phase 1 is generally in the range of 20 to 80 subjects. Phase 2 usually involves no more than several hundred patients. Phase 3 usually includes from several hundred to several thousand subjects. A specific program can sit well outside these ranges — FDA's draft weight-reduction guidance recommends a safety database of 3,000 subjects on drug and no fewer than 1,500 on placebo for at least a year.
Are all phase 3 trials controlled?
No. The regulation describes phase 3 studies as expanded controlled and uncontrolled trials. That is the regulation's own wording. A trial being in phase 3 therefore does not by itself establish that it compared the drug against anything.
Why can a phase 1 study be missing from the public registry?
Because the statute excludes it. The definition of an applicable drug clinical trial at 42 U.S.C. 282(j) covers a controlled clinical investigation other than a phase 1 clinical investigation, and it borrows the meaning of phase 1 from 21 CFR 312.21. A phase 1 study can be absent from the registry without any rule being broken.
Does the phase say anything about whether a drug works?
No. A phase describes the stage and design of an investigation, not a result. Under 21 CFR 312.22, FDA's review of phase 2 and 3 submissions assesses whether the investigations are likely to yield data capable of meeting the statutory standards for approval — which is a statement about the quality of the question being asked, not about the answer.
Do compounded products go through these phases?
No. This entire framework applies to investigational new drugs under a federal application, which 21 CFR 312.20 requires before a covered clinical investigation may begin. A compounded preparation is not an FDA-approved drug and is not developed through this process, so phase language attached to a compounded product's marketing is not describing that product's own development.