Clinical trial phases are the conventional stages through which an investigational medicine is tested in humans, each answering a different question and each larger than the one before. Phase 1 establishes safety and dosing, Phase 2 looks for a signal of efficacy in patients, Phase 3 confirms benefit and harm in a population resembling intended users, and Phase 4 continues to study the product once it is in routine use.
The phases are a framework, not a legal sequence. Designs are frequently combined, some pathways compress or skip stages, and trials of devices, vaccines, and procedures follow related but distinct routes. This guide explains what happens at each stage and how the resulting evidence supports a regulatory decision.
Before Phase 1: Preclinical Development
Human testing is preceded by laboratory and animal research intended to characterize the mechanism, establish a plausible therapeutic effect, and assess toxicity. Preclinical work also produces the pharmacology and safety data used to select a starting dose for the first human study. Manufacturing must meet quality standards, since the material given to participants has to be consistent and characterized.
In the United States, a sponsor submits an Investigational New Drug application to the FDA containing preclinical data, manufacturing information, and the proposed clinical protocol; the study may proceed if the agency does not object within the review period. In the European Union, clinical trial authorisation is sought from member states through a coordinated process, with the European Medicines Agency operating the shared portal and database. Independent ethics review is required in both systems.
Phase 1: Safety, Tolerability, and Dose
Phase 1 is the first administration of the candidate to humans. Its objectives are to characterize safety and tolerability, to identify a dose range suitable for further study, and to describe pharmacokinetics — absorption, distribution, metabolism, and excretion — and pharmacodynamics, the biological effect produced.
Participants are typically a small number of healthy volunteers. Where the intervention is expected to be toxic, as with cytotoxic cancer treatments, Phase 1 enrols patients with the condition instead, since exposing healthy people to that risk would not be justifiable.
Dosing is usually escalated in sequential cohorts, with each increment permitted only after review of safety data from the preceding one. Common variations include single ascending dose and multiple ascending dose designs, food-effect studies, and drug-interaction studies. Phase 1 is not designed to demonstrate that a treatment works.
Phase 2: Preliminary Efficacy
Phase 2 enrols patients who have the condition the candidate is intended to treat. It seeks the first evidence that the intervention produces the intended effect, refines the dose and schedule, and broadens the safety picture in the target population, whose physiology and concurrent medications differ from healthy volunteers.
Phase 2 studies are sometimes divided informally: an earlier stage exploring dose and biological effect, and a later stage using randomized, controlled designs with a clinical or well-validated surrogate endpoint. They are larger than Phase 1 but generally not large enough to establish clinical benefit definitively or to detect uncommon harms.
Attrition is heaviest at this stage. Many candidates that looked promising in early testing show no meaningful effect once studied in patients, or reveal a benefit-harm balance that cannot support further development.
Phase 3: Confirmatory Trials
Phase 3 is designed to confirm efficacy and characterize the balance of benefits and harms in a population resembling the intended users. These trials are typically randomized, controlled, blinded where feasible, and conducted across many sites and often many countries.
The comparator may be placebo, where no proven effective treatment exists, or an active treatment representing current standard care. Trials may be designed to show superiority over the comparator, or non-inferiority — that the candidate is not worse by more than a prespecified margin — which is used when a new option offers advantages in safety, convenience, or cost rather than greater efficacy.
Phase 3 provides the statistical power to detect clinically meaningful differences in the primary endpoint and to observe adverse events too infrequent to appear in smaller studies. Regulators commonly expect more than one adequate and well-controlled trial supporting efficacy, though a single compelling trial with confirmatory evidence may suffice in defined circumstances. Independent data monitoring committees review accumulating data and may recommend stopping for harm, clear benefit, or futility.
Regulatory Review and Approval
When the evidence package is complete, the sponsor applies for authorization: a New Drug Application or Biologics License Application to the FDA, or a Marketing Authorisation Application evaluated by the EMA's scientific committees on behalf of the European Commission.
Regulators assess the complete dossier — clinical efficacy and safety, pharmacology, manufacturing quality, and proposed labelling. They may consult advisory committees, inspect trial sites and manufacturing facilities, and request additional analyses. The outcome is not binary: approval may be granted with restrictions on the eligible population, with required risk-management measures, or with an obligation to conduct specified post-authorization studies.
Expedited Pathways
Both agencies operate mechanisms to speed development and review for products addressing serious conditions with unmet need. In the United States these include fast track, breakthrough therapy, priority review, and accelerated approval based on a surrogate endpoint reasonably likely to predict benefit, subject to confirmatory studies. In the European Union, accelerated assessment, conditional marketing authorisation, and the PRIME scheme serve analogous purposes. These pathways change timing and evidentiary sequence; they do not remove the requirement to demonstrate an acceptable benefit-risk balance.
Phase 4: Post-Authorization Studies
Phase 4 studies are conducted after a product is on the market. They examine long-term safety and effectiveness under routine conditions, study populations under-represented in earlier trials such as older adults or people with multiple conditions, compare the product with other established options, and investigate rare adverse effects detectable only at scale.
Phase 4 research is distinct from, though related to, ongoing pharmacovigilance: the continuous collection and assessment of safety reports that can lead to label changes, restrictions, or withdrawal. Regulators can require specific post-authorization safety or efficacy studies as a condition of approval.
Where the Phase Model Does Not Apply Directly
- Medical devices follow risk-class-based pathways. Many are authorized on the basis of demonstrated equivalence to an existing device or on conformity assessment against standards, with clinical investigation required primarily for higher-risk classes.
- Vaccines use the same phase terminology but with distinct endpoints, very large Phase 3 trials to detect uncommon events, and immunogenicity as well as clinical efficacy measures.
- Combined and seamless designs merge stages — Phase 1/2 or Phase 2/3 — under a single protocol with prespecified decision rules.
- Platform and adaptive trials evaluate multiple candidates against a shared control under a master protocol, allowing arms to enter and leave.
- Investigator-initiated and academic trials may study established medicines in new indications or compare existing treatments, without a commercial development programme behind them.
Sources
- U.S. Food and Drug Administration — clinical research phases; Investigational New Drug process; expedited programmes
- European Medicines Agency — clinical trials regulation; marketing authorisation procedures; PRIME and conditional authorisation
- U.S. National Institutes of Health — clinical trial phase definitions
- ClinicalTrials.gov — phase classification in trial records
- International Council for Harmonisation — E6 Good Clinical Practice; E8 general considerations for clinical studies; E9 statistical principles
- World Health Organization — International Clinical Trials Registry Platform
