How long do clinical trials take?
Single trials, whole programs and what changes the clock
Matic Broz, PhD13 min
A new drug usually spends about 8 to 10 years in clinical trials before it is approved. For new drugs approved by the FDA from 2010 to 2020, the median time from the first study in people to approval was 8.3 years. An industry benchmark that adds up the time spent in each phase puts the average at 10.5 years.
Each trial within that span is much shorter. In our analysis of 26,680 completed drug and biologic trials, the median industry-sponsored trial ran about 9 months in Phase 1, 2 years in Phase 2 and 2.4 years in Phase 3, from first participant to final visit. Trials run by universities and hospitals, and trials in cancer, took considerably longer.
How long does it take a drug to get through clinical trials?
A drug that reaches approval typically spends 8 to 10 years in human testing and regulatory review. Brown and colleagues measured this for 405 innovative drugs approved by the FDA from 2010 to 2020, counting from the opening of the first human study to marketing approval. The median was 8.3 years, the range ran from under 5 years to more than 20, and the median barely moved from one approval year to the next.[1]
The fastest programs show what is possible under ideal conditions. Osimertinib, a lung cancer drug, went from first human dose to accelerated approval in 984 days. Elexacaftor, part of the cystic fibrosis combination Trikafta, took 1,043 days without any accelerated pathway, helped by its developer's experience with three earlier cystic fibrosis drugs.[1]
The BIO industry benchmark, built from 6,151 successful phase transitions recorded in 2011 to 2020, gives a longer average of 10.5 years from Phase 1 to approval. It splits that time into 2.3 years in Phase 1, 3.6 years in Phase 2, 3.3 years in Phase 3 and 1.3 years at the regulatory stage.[2]
| Stage | Typical participants | FDA's typical length | Mean time in stage (BIO, 2011-2020) |
|---|---|---|---|
| Phase 1 | 20 to 100 | Several months | 2.3 years |
| Phase 2 | Up to several hundred | Several months to 2 years | 3.6 years |
| Phase 3 | 300 to 3,000 | 1 to 4 years | 3.3 years |
| FDA review | - | Goal of 10 months (standard) or 6 months (priority) | 1.3 years |
| Phase 1 to approval | - | - | 10.5 years |
The two estimates measure different things, which explains most of the two-year gap. Brown's figure is a median taken from dates in FDA review documents, and only for drugs that were approved. BIO's is a sum of mean phase durations from companies' public disclosures. Means are pulled upward by a few very long programs, and disclosure dates include the months a company spends analysing results, raising money or deciding whether to continue.[2]
Neither clock starts at the beginning of drug development. Target discovery, chemistry, animal studies and early screens such as ADMET prediction come first and add several more years to the total cost and timeline. Both estimates also describe only the drugs that succeeded. Roughly nine in ten candidates that enter Phase 1 never reach approval, as the clinical trial failure rates show, and many of them stop partway through these phases.
How long does each phase take?
A single industry-sponsored drug trial usually lasts from a few months to about three years. We measured this directly in ClinicalTrials.gov, the trial registry run by the US National Library of Medicine. Among drug and biologic trials completed in 2021 to 2025, the median industry trial took 8.6 months in Phase 1, 23.6 months in Phase 2 and 28.8 months in Phase 3. Trials led by other sponsors, mostly universities, hospitals and government institutes, took 37.1, 46.0 and 36.9 months.[5]
These durations run from the day the first participant enrolls to the last participant's final visit. A single participant is usually involved for much less time, because people join over many months and each follows the protocol's own visit schedule. The registry also records a primary completion date, when the main result has been measured. Counted to that date, the industry medians were 7.7 months in Phase 1, 20.1 in Phase 2 and 23.6 in Phase 3.[5]
| Phase | Trials | All sponsors | Industry | Other sponsors | Cancer | Other diseases |
|---|---|---|---|---|---|---|
| Phase 1 | 7,885 | 12.2 | 8.6 | 37.1 | 50.5 | 8.8 |
| Phase 1/2 | 1,850 | 40.1 | 35.9 | 45.2 | 62.2 | 27.8 |
| Phase 2 | 6,981 | 35.1 | 23.6 | 46.0 | 60.0 | 25.0 |
| Phase 2/3 | 738 | 26.2 | 25.6 | 26.3 | 58.7 | 24.7 |
| Phase 3 | 4,624 | 31.3 | 28.8 | 36.9 | 75.6 | 26.3 |
| Phase 4 | 3,734 | 24.0 | 22.1 | 24.3 | 32.4 | 23.7 |
A trial is not the same as a phase. A drug usually runs several trials in each phase, sometimes in parallel and sometimes one after another, which is why the time a program spends in Phase 2 (3.6 years in the BIO benchmark) exceeds the length of a typical Phase 2 trial.[2]
Phase 1
A Phase 1 trial usually lasts several months. FDA describes it as a first test in 20 to 100 healthy volunteers or patients, aimed at safety and dosage.[3] The industry median of 8.6 months fits that description, and half of industry Phase 1 trials finished within 3.4 to 20.5 months.[5]
Cancer is the exception. Phase 1 cancer trials enroll patients rather than healthy volunteers, often escalate the dose cohort by cohort and then add expansion groups for several tumor types. Their median was 50.5 months, and BIO likewise found that oncology programs spend longer in Phase 1 than any other disease area, 2.7 years against an average of 2.3.[5][2]
Phase 2
A Phase 2 trial typically takes about two years. FDA puts it at several months to 2 years, with up to several hundred patients, testing whether the drug appears to work and refining the dose.[3] The industry median was 23.6 months, while the middle half of trials ran 13.2 to 40.7 months.[5]
Phase 2 is the longest stage of a program even though its individual trials are not the longest. In the BIO benchmark a drug spent 3.6 years there, often running more than one trial before a company committed to the much larger cost of Phase 3.[2] It is also the stage with the highest attrition: about 71% of programs that enter Phase 2 never reach Phase 3.[2]
Phase 3
A Phase 3 trial usually lasts one to four years. FDA's range is 1 to 4 years for 300 to 3,000 participants, enough to confirm benefit and detect less common side effects.[3] The industry median was 28.8 months, with the middle half between 16.4 and 52.8 months.[5]
Cancer Phase 3 trials had a median of 75.6 months, more than 6 years, but reached their primary completion date at 47.3 months. The difference is follow-up: survival trials keep tracking patients after the main analysis, so the final visit can come years after the result that supports approval.[5]
FDA review
After Phase 3, the sponsor submits a New Drug Application or Biologics License Application. Under the current PDUFA agreement, FDA aims to act on 90% of applications for new molecular entities and original biologics within 10 months of the 60-day filing date for standard review, or 6 months for priority review. That is about 12 and 8 months from submission.[4]
The goal is not the whole story. BIO measured an average of 1.3 years from filing to approval, and Brown and colleagues found that failing the first review cycle was associated with 829 more days of review.[2][1]
Why do some clinical trials take longer than others?
Disease, sponsor and endpoint explain most of the spread. Cho and colleagues analysed 90,366 trials in the Citeline database and found that an oncology indication was the strongest predictor of a long trial. An academic or government sponsor, the trial phase, target enrollment and trial region also mattered.[6]
The disease sets the endpoint, and the endpoint sets the clock. A trial that measures a blood marker can finish soon after the last dose, while a heart-failure or cancer trial must wait for enough deaths, hospitalizations or progressions to occur. In Cho's data, Phase 3 oncology trials averaged 58.6 months, against 30.5 for cardiovascular disease and 14.8 for vaccines.[6] Across whole programs, BIO found a range from 9.2 years in allergy to 12.2 years in urology.[2]
| Disease area | Phase 1 | Phase 2 | Phase 3 | Filing to approval | Total |
|---|---|---|---|---|---|
| Allergy | 1.5 | 3.8 | 2.9 | 1.1 | 9.2 |
| Metabolic | 2.0 | 3.2 | 3.1 | 1.2 | 9.5 |
| Infectious disease | 2.0 | 3.5 | 3.1 | 1.2 | 9.7 |
| Ophthalmology | 2.1 | 2.9 | 3.4 | 1.3 | 9.8 |
| Autoimmune | 2.1 | 3.6 | 3.2 | 1.1 | 10.0 |
| Oncology | 2.7 | 3.7 | 3.1 | 0.8 | 10.3 |
| Respiratory | 2.1 | 3.5 | 3.3 | 1.5 | 10.4 |
| Psychiatry | 2.3 | 3.4 | 2.8 | 1.8 | 10.4 |
| Other | 1.9 | 3.5 | 3.2 | 1.8 | 10.5 |
| All indications | 2.3 | 3.6 | 3.3 | 1.3 | 10.5 |
| Endocrine | 1.8 | 3.4 | 3.7 | 1.8 | 10.7 |
| Hematology | 2.2 | 3.4 | 3.6 | 1.5 | 10.7 |
| Gastroenterology | 1.6 | 3.9 | 3.9 | 1.4 | 10.8 |
| Neurology | 2.1 | 3.7 | 3.7 | 1.6 | 11.1 |
| Cardiovascular | 2.4 | 3.8 | 4.2 | 1.2 | 11.5 |
| Urology | 2.7 | 5.0 | 2.9 | 1.6 | 12.2 |
Oncology programs end close to the average in total despite their long trials. Their regulatory stage is the shortest of any area at 0.8 years, which BIO attributes partly to FDA's oncology office making heavy use of expedited pathways and real-time review.[2]
Recruitment is the delay sponsors control least. Of 2,579 Phase 2 and 3 trials that closed in 2011, 19% either stopped for poor accrual or finished with less than 85% of their planned enrollment. The risk was higher for trials with more eligibility criteria, fewer sites or non-industry funding (25% against 16% for industry).[7] Slow recruitment was also the most common reason given by terminated trials in our analysis of why clinical trials fail.
Rare diseases add another layer. Drugs with orphan designation took about 552 more days in clinical development than comparable drugs, despite smaller trials. The authors point to patients who are hard to find, poorly understood disease courses and endpoints that may have to be developed from scratch.[1]
What can shorten the timeline?
FDA's expedited programs cut the most time. Accelerated approval lets a drug for a serious condition be approved on a surrogate endpoint, such as a laboratory value or tumor shrinkage, before a clinical benefit like longer survival is proven; the sponsor must still run confirmatory studies afterwards.[8] After controlling for other factors, Brown and colleagues found that accelerated approval was associated with 1,100 fewer days of clinical development, about 3 years (95% CI 563 to 1,637 days).[1]
Breakthrough therapy designation is meant to speed the development and review of drugs that may substantially improve on available therapy.[9] It was associated with 479 fewer days, though with a wide interval (5 to 953 days). Priority review shortened the review itself by 103 days, but fast track designation showed no significant effect on development time.[1]
Trial design matters too. Combined Phase 1/2 or Phase 2/3 trials remove the gap between separate studies, and master protocols such as basket, umbrella and platform trials test several drugs or diseases at once, often with a shared control group.[6] Vaccine development uses the same three pre-approval phases, but FDA notes it has no predetermined timeline and that its phases may run in sequence or overlap.[10]
Medical devices follow a different path. Instead of drug-style phases, FDA describes exploratory studies, pivotal studies that support safety and effectiveness, and postmarket studies, and not every device needs all three.[11] Their duration depends on the device's risk class and the evidence its regulatory pathway requires.
Taken together, the numbers describe nested clocks. An individual participant may be in a study for weeks or months; a typical industry trial lasts one to three years; a drug that succeeds spends about 8 to 10 years moving through several such trials and an FDA review. Cancer, rare diseases and academic sponsorship stretch every layer, while accelerated approval has the largest measured effect in the other direction.
Methodology
We queried the ClinicalTrials.gov API v2 on October 24, 2026 (data timestamp October 23, 2026) for interventional studies with an overall status of completed, at least one drug or biological intervention and a study completion date between January 1, 2021 and December 31, 2025. The query returned 30,143 studies. We excluded 3,463 registered as "Not Applicable" for phase, leaving 26,680.[5]
For each study we calculated the days from the actual start date to the actual study completion date, and separately to the actual primary completion date. Month-only dates were read as the first of the month. Industry means a lead sponsor classed as industry; all other classes count as other sponsors. Cancer trials are those with a condition term or ancestor under the MeSH branches Neoplasms by Histologic Type or Neoplasms by Site.
Grouping by completion year captures trials that started at any time but finished in the window, the same approach Our World in Data uses. Its medians for all completed interventional trials finishing in 2024, 408 days in Phase 1, 1,146 in Phase 2 and 994 in Phase 3, sit close to our all-sponsor medians of 371, 1,067 and 952 days.[12] Completed trials exclude those that were terminated or are still running, so the medians describe trials that ran their course. Registry dates are entered by sponsors and are not independently verified.
The analysis script and summary CSV contain the exact query, inclusion rules, trial counts, quartiles and primary-completion medians for every phase and group.
