A promising compound sits in a freezer in a sponsor’s lab, backed by strong preclinical data and a team eager to see it help patients. Between that moment and a pharmacy shelf, or an operating room, lies one of the most tightly regulated journeys in medicine. Every drug, biologic, and medical device sold in the United States travels through it, and the journey is built around a small number of formally defined stages that determine how much evidence exists before a treatment ever reaches the people it is meant to help.
For a clinical operations lead scoping a new protocol, a CRO team planning site budgets, or a medtech program manager preparing an FDA submission, these stage names carry real operational weight: how many participants to recruit, how long a study will run, what regulators expect to see before the next stage opens, and how a device program’s path departs entirely from the pharmaceutical model. This guide walks through the clinical trial phases in the order a treatment actually experiences them, then breaks out where the medical device clinical trial phases part ways with the drug framework.
The phases of a clinical trial move from small, closely monitored studies toward large, real-world evidence-gathering. Clinical trial phases 1 through 3 happen before a regulator approves a product for sale; Phase 4 happens after. Each phase answers a different question: is it safe, does it work, does it work better than what already exists, and does it hold up once thousands of people are using it outside a research setting?
The table below is a quick clinical trial phases diagram, summarizing participant numbers, purpose, and typical length for each stage, based on the FDA’s clinical trial phases framework.
Each phase exists to answer a specific question before a study is allowed to grow larger, longer, and more expensive. A treatment cannot skip ahead. It has to earn its way from one stage to the next by producing evidence that satisfies the FDA and the study’s own data safety monitoring board.
Phase 1 is the first time an investigational drug or biologic is given to humans, often as part of first-in-human trials. The question is narrow and specific: is this safe, and at what dose. According to the FDA:
“Study Participants: 20 to 100 healthy volunteers or people with the disease/condition. Length of Study: Several months. Purpose: Safety and dosage.”
Participants, usually healthy volunteers unless the treatment carries meaningful risk (as with many oncology drugs, where patients with the disease are enrolled instead), are monitored closely for adverse events while researchers escalate the dose in controlled steps. Pharmacokinetics, how the body absorbs, distributes, and clears the drug, is a major focus. Roughly 70% of drugs that enter Phase 1 move on to Phase 2, according to FDA data, which reflects how much of the early risk in drug development is safety risk rather than efficacy risk.
Key characteristics of Phase 1:
Also read: Understanding Good Clinical Practice (GCP) in Imaging
Once a treatment has cleared the basic safety bar, Phase 2 asks whether it actually does what it is supposed to do. Researchers enroll people who have the condition the treatment targets, rather than healthy volunteers, and start measuring effectiveness alongside continued safety monitoring. This is also where sponsors typically settle on the dose and regimen that will carry into Phase 3, and where they can request FDA guidance on how to design the larger study that follows.
Key aspects of Phase 2:
Only about a third of drugs that enter Phase 2 advance to Phase 3, which makes this the phase where the largest share of candidates are stopped. A treatment can be safe and still fail here if the effect size is too small or too inconsistent to justify a large, expensive Phase 3 program.
Phase 3 is the confirmatory stage: large, often multicenter, frequently randomized against a placebo or the current standard of care. The FDA describes it this way:
“Study Participants: 300 to 3,000 volunteers who have the disease or condition. Length of Study: 1 to 4 years. Purpose: Efficacy and monitoring of adverse reactions.”
Running a study at this scale, often across dozens of sites and several countries, means coordinating imaging, lab data, and adverse event reporting consistently everywhere the study operates. Well-run prospective imaging trials only produce a defensible efficacy signal if every site is acquiring and reading images the same way, which is a large part of why Phase 3 budgets and timelines run so much higher than earlier phases.
This phase involves:
Roughly 25% to 30% of drugs that enter Phase 3 go on to receive approval. That success rate, higher than the earlier phases despite the scale, reflects the fact that sponsors generally only advance a treatment into Phase 3 once Phase 2 has produced a reasonably convincing efficacy signal.
Approval is not the end of the evidence-gathering process. Phase 4, sometimes called post-marketing surveillance, tracks a treatment once it is available to the general population, a group far larger and more diverse than any pre-approval trial could enroll. This is where rare side effects, ones that only appear in 1 in 10,000 patients or fewer, tend to surface, along with real-world effectiveness data across populations that were underrepresented in the earlier phases.
Key features of Phase 4:
Progression between clinical trial phases is not automatic. At the end of each phase, the sponsor, together with its data safety monitoring board and, often, direct input from the FDA, decides whether the evidence justifies moving forward. The FDA specifically offers sponsors a consultation after Phase 2 to help design a well-powered Phase 3 study, precisely because so much rides on getting that next phase right.
The attrition along the way is significant. FDA figures put the phase-to-phase advancement rate at roughly 70% from Phase 1 to Phase 2, about 33% from Phase 2 to Phase 3, and 25% to 30% from Phase 3 to approval. Multiplied across the full sequence, only a small fraction of treatments that begin human testing ever reach the market, which is why regulators and sponsors alike treat the transition points between clinical trial phases 1 to 4 as real decision gates rather than formalities.
FDA involvement does not start after Phase 3 is finished. Before Phase 1 can begin, a sponsor must submit an Investigational New Drug (IND) application, including preclinical safety data, manufacturing information, and the clinical protocol. The FDA review team has 30 days to respond, either allowing the trial to proceed or placing it on clinical hold if participants would face an unreasonable risk.
After Phase 3 concludes, the sponsor files a New Drug Application (NDA) or Biologics License Application (BLA), a comprehensive submission covering every study conducted from preclinical work through Phase 3. FDA's review team, medical officers, statisticians, pharmacologists, and chemists, among them, typically has 6 to 10 months to reach a decision, and FDA inspectors visit clinical trial sites directly to check for data integrity. When questions remain, the FDA can convene an advisory committee of independent experts before making a final call. Sponsors working with imaging endpoints should also be familiar with the FDA guidelines for clinical trial imaging, which shape how imaging data is acquired, read, and submitted as part of that evidence package.
It is tempting to apply the Phase 1 through 4 framework to medical devices as well, but that is not how the FDA structures device development. Medical device clinical trial phases follow a different logic built around three general stages rather than four numbered ones: exploratory, pivotal, and post-market.
The exploratory stage, which includes early feasibility and first-in-human device studies, is typically small in scale and used to evaluate the basic advantages and limitations of a device design before it is finalized. The pivotal stage is the definitive one: it is where a sponsor gathers the valid scientific evidence needed to support the device's primary safety and effectiveness claims for its intended use, functionally similar in purpose to a drug's Phase 3, though built around a device-specific study design. How that evidence is generated depends heavily on device classification. Lower-risk Class I and Class II devices generally move through the 510(k) pathway by demonstrating substantial equivalence to an already-cleared device, while higher-risk Class III devices require Premarket Approval (PMA), which mandates at least one clinical study.
Where imaging is the primary endpoint, whether that is a diagnostic device or an imaging-guided therapeutic, sponsors need to understand how imaging endpoints for medical device approval are evaluated, since imaging-based endpoints carry their own standards for reader blinding, image quality, and reproducibility. Post-market obligations for devices also differ from the drug model. Many Class III devices carry specific post-approval study requirements, and post-market surveillance for medical devices can include real-world registries and adverse event reporting that continue for years after clearance or approval.
Added together, the four clinical trial phases 1 to 4 alone can span three to seven years: several months for Phase 1, several months to two years for Phase 2, one to four years for Phase 3, and an open-ended Phase 4 that continues after approval. That figure does not include the preclinical research that precedes Phase 1, the time needed to recruit eligible participants at each stage, or the 6 to 10-month FDA review window between Phase 3 and approval. In practice, a sponsor's total timeline from IND filing to approval is almost always longer than the sum of the individual phase lengths suggests, and delays in participant recruitment are one of the most common reasons a program's timeline slips.
The outcome of a clinical trial depends on several interconnected factors, and weakness in any one of them can undermine an otherwise well-funded, well-designed study.
Key factors include:
For studies that rely on imaging as an endpoint, one more factor sits alongside these: the consistency of image acquisition and review across every site. A trial can be well designed and well powered on paper and still produce unreliable results if scans are acquired inconsistently between sites or reviewed without a standardized process. That is one of the reasons central review and standardized imaging workflows have become a routine part of Phase 2 and Phase 3 protocol design, particularly in oncology and other therapeutic areas where imaging is the primary or a key secondary endpoint.
That compound in the freezer, if it survives the journey, will have passed through years of increasingly demanding scrutiny by the time it reaches a patient. Each clinical trial phase exists to answer one question well before the next, larger, more expensive question is even asked, and that discipline is what allows a physician to prescribe a new treatment, or a hospital to adopt a new device, with confidence in the evidence behind it.
For the CROs, sponsors, and academic research teams running the imaging-dependent studies inside that framework, especially in Phase 2 and Phase 3, where sample sizes grow and sites multiply, the operational bar rises fast. Consistent acquisition protocols, central and blinded independent review, and a reliable audit trail across every site are what keep an imaging endpoint defensible from Phase 1 through FDA submission and, for devices, through post-market surveillance as well.
Phase 1 tests safety and dosage in 20 to 100 participants. Phase 2 evaluates effectiveness and side effects in up to several hundred participants who have the condition. Phase 3 confirms efficacy and monitors adverse reactions in 300 to 3,000 participants, often against a placebo or standard treatment. Phase 4 monitors long-term safety and real-world effectiveness after regulatory approval, with no fixed endpoint.
Yes. Phase 0 is an optional, exploratory stage that takes place before Phase 1, involving a very small number of participants, often fewer than 15, who receive microdoses of the investigational drug. It is not designed to test safety or efficacy in a therapeutic sense; its purpose is to gather early pharmacokinetic data on how the drug behaves in the human body before committing to a full Phase 1 program.
FDA approval decisions are made after Phase 3, based on the New Drug Application (NDA) or Biologics License Application (BLA) the sponsor files once Phase 3 data is complete. Phase 4 takes place after approval and does not gate market access; it monitors the treatment's safety and effectiveness once it is already available to patients.
No. Medical devices generally follow a three-stage framework instead: exploratory or feasibility studies, a pivotal study that provides the primary evidence of safety and effectiveness, and post-market studies after clearance or approval. The specific pathway, 510(k) or Premarket Approval, depends on the device's risk classification.
Yes. Through the Collective Minds Research platform for CROs and pharma, we support imaging across clinical trial phases 1 through 4. The platform is built for studies where multicenter imaging and imaging-based endpoints are central to evaluating clinical performance, from early feasibility work through post-market follow-up.
Reviewed by: Pilar Flores Gastellu on August 27, 2026