Skip to content
All resources

Reading the evidence

Phase 1, 2 and 3 Trials Explained: What Each Clinical Trial Phase Answers

AR

By

· 11 min read

A summary of published research, with citations. It states no amount, schedule or route, and makes no claim about what this compound does in a person. Supplied for laboratory research use only.

On this page · 9 sections
  1. What comes before phase 1?
  2. What is a phase 1 trial designed to answer?
  3. What does a phase 2 trial test?
  4. What does phase 3 add?
  5. How many compounds get through?
  6. What is accelerated approval?
  7. What is a phase 4 trial?
  8. What does a phase label not tell you?
  9. Frequently asked questions

Phase 1 trials ask whether a compound is tolerable in people and how the body handles it, usually in a few dozen participants. Phase 2 trials ask whether it does anything measurable and at what exposure, in tens to a few hundred. Phase 3 trials ask whether that effect holds up in hundreds or thousands of people against a control, and they are what regulators approve on. Phase 4 covers studies after approval. Most compounds never reach phase 1 at all.

Three stacks of paper documents of increasing height on a white desk
Illustration: each trial phase is larger, and answers a different question.

What comes before phase 1?

Everything that happens before a compound is given to a person in a registered trial is preclinical: binding assays, cell culture, tissue explants and animal models. The general limits of that evidence, and why animal findings so often fail to carry over, are set out in how to read a peptide study. This page starts where that one stops.

Most compounds discussed on this site never left the preclinical stage. KPV has been examined in chemically induced colitis in mice and has no controlled human trial. 5-Amino-1MQ is cell and rodent work. Neither is approved anywhere.

BPC-157 is a useful edge case, because it does have published human data. A 2026 systematic review by Poolton and colleagues searched four databases and ClinicalTrials.gov and found three human studies with 30 participants between them. All three were uncontrolled: two retrospective and one prospective pilot of two people. The reviewers rated the certainty of the evidence as very low. Human data that exists is not the same thing as a phase 1 programme. BPC-157 is not approved by Health Canada, the FDA or the EMA. Its animal literature is covered in BPC-157 and connective tissue.

What is a phase 1 trial designed to answer?

Safety, tolerability and pharmacokinetics: what the body does to the drug, how fast it is cleared, and which adverse events appear as exposure rises. Efficacy is at most a secondary question. Groups are small, and participants may be healthy volunteers or, as is common for metabolic drugs, people with the target condition.

Retatrutide's first published trial in people with type 2 diabetes shows the shape. Urva and colleagues enrolled 72 participants at four US centres for 12 weeks in a phase 1b, double-blind, placebo-controlled, multiple-ascending-dose design. The primary outcome was safety and tolerability. Pharmacokinetics and glucose and weight changes were secondary. The half-life was approximately six days. Adverse events were mostly gastrointestinal, and 29 of the 72 participants stopped the study early. The authors concluded that the findings supported moving to phase 2, which is all a phase 1 trial is built to conclude.

A phase 1 trial that reports a striking weight or glucose change has still only produced a hypothesis. It was too small and too short to estimate either with any precision. The design was also not powered to detect them.

What does a phase 2 trial test?

Whether the drug moves the endpoint it is meant to move, and how the effect and the side effects change across exposure levels. Phase 2 is where dose-response is mapped and where most drug programmes end.

Retatrutide's phase 2 obesity trial randomised 338 adults to placebo or one of several dose groups for 48 weeks. Its primary endpoint was the percentage change in body weight at 24 weeks. At 48 weeks the highest dose group had lost 24.2% of body weight on average, against 2.1% on placebo. Gastrointestinal events were dose-related. A trial like this is designed to choose which exposures phase 3 should test, and it did that. It was not designed to establish safety in a broad population or to count heart attacks.

Elamipretide shows the other common outcome. Its small early trials in mitochondrial myopathy produced encouraging signals. The phase 3 trial, MMPOWER-3, randomised 218 adults and missed both primary endpoints: the difference in six-minute walk distance against placebo was −3.2 metres. The full arc is in what is SS-31 (elamipretide). A positive phase 2 result is the entry ticket to phase 3, not a preview of its answer.

What does phase 3 add?

Scale, a prespecified primary endpoint, a control group, and usually replication. Phase 3 trials are often called pivotal or registrational because they are the evidence a regulator reviews. They are large enough to detect less common adverse events and to estimate the effect with a confidence interval narrow enough to act on.

TRIUMPH-1, retatrutide's first published phase 3 obesity trial, randomised 2,339 adults without diabetes for 80 weeks. It had three prespecified primary outcomes: weight change in everyone, a pain score in a nested subgroup of 574 with knee osteoarthritis, and the apnoea–hypopnoea index in 243 with obstructive sleep apnoea. The retatrutide phase 3 article covers the results and the programme around it.

Regulators generally want more than one adequate trial. Bremelanotide's RECONNECT programme ran two identical phase 3 trials in parallel, 1,267 women randomised in total, and both met their co-primary endpoints. That design answers a question a single trial cannot: whether the result reproduces. The regulatory record that followed is in the PT-141 article.

Even phase 3 is narrower than it sounds. Body weight, HbA1c and symptom scores are intermediate measures. The trials that count heart attacks, strokes and deaths are separate outcome trials, often an order of magnitude larger. Retatrutide's, TRIUMPH-Outcomes (NCT06383390), lists an estimated 10,000 participants and is active, not recruiting.

PhaseQuestion it is designed to answerTypical sizeExample on this site
PreclinicalIs there a mechanism and an effect in a model?Cells, tissue, animalsKPV, 5-Amino-1MQ, MOTS-c
Phase 1Is it tolerable, and how is it handled by the body?TensRetatrutide phase 1b: 72 enrolled
Phase 2Does it move the endpoint, and at what exposure?Tens to a few hundredRetatrutide phase 2: 338 randomised
Phase 3Does the effect hold up against control, at scale?Hundreds to thousandsTRIUMPH-1: 2,339; MMPOWER-3: 218
Outcome trialDoes it change hard clinical events?Thousands to tens of thousandsTRIUMPH-Outcomes: about 10,000 estimated
Phase 4What happens after approval, in wider use?Any size4TAZPower: 48 estimated

Rare diseases compress these sizes. A phase 3 trial in an ultra-rare disorder may have fewer participants than a phase 1 trial for a common one, which is one reason the evidence behind rare-disease approvals is often thinner.

How many compounds get through?

Few. Wong, Siah and Lo analysed more than 406,000 trial records covering over 21,000 compounds and estimated success rates by phase and therapeutic area. In oncology, they put the success rate at 3.4% in their sample. Rates differ by field, but the direction does not: most programmes stop. That is why "in clinical trials" describes a stage, not a likely outcome.

What is accelerated approval?

A US pathway that lets the FDA approve a drug for a serious condition on a surrogate or intermediate endpoint judged reasonably likely to predict clinical benefit, before that benefit is shown. The sponsor must then run a confirmatory trial. Whether a marker is a fair stand-in for an outcome is the subject of surrogate endpoints explained.

Confirmation does not always come. Gyawali and colleagues followed 93 cancer indications granted accelerated approval. Confirmatory trials for 19 of them (20%) reported an improvement in overall survival. Another 19 reported improvement in the same surrogate used before approval, and 20 in a different surrogate.

Elamipretide is the example on this site. In September 2025 it received US accelerated approval to improve muscle strength in Barth syndrome, a cardiolipin disorder, for patients weighing at least 30 kg. The randomised, double-blind crossover part of its trial, TAZPOWER, had 12 participants and met neither primary endpoint. The improvements were reported in the open-label extension that followed. Why open-label data carry less weight is the subject of open-label vs placebo-controlled. The confirmatory trial, 4TAZPower (NCT07531251), is registered as phase 3b/4. It is randomised, double-blind and placebo-controlled, with an estimated 48 participants over 72 weeks, and its stated objective is to confirm the efficacy behind the accelerated approval. The Drugs profile of the decision describes it as elamipretide's first approval anywhere, and this site's SS-31 article found no Health Canada authorisation.

What is a phase 4 trial?

Any study of a drug after it is approved. Some are safety surveillance in broader populations than the trials enrolled. Some are confirmatory trials required by a regulator, as with 4TAZPower. Some test an approved drug for a new question entirely. The SAMSON nocebo study of statin side effects is registered as phase 4 because statins are approved, not because it is a late stage of anything. The label describes regulatory timing, not quality.

What does a phase label not tell you?

Whether the trial was any good. A phase 3 trial can be open-label, poorly blinded or analysed selectively. A phase 2 trial can be rigorous. Read the design, not the number.

What the trial found. A registry entry marked "completed" says nothing about the result, and a paper's abstract can lead with a secondary finding. How to read a PubMed abstract covers what to check first.

Whether the compound is approved. Retatrutide has published phase 3 trials and is not approved by Health Canada or any other regulator. A Canadian authorisation carries a Drug Identification Number, searchable in Health Canada's Drug Product Database. The rules for research material are in peptides and Canadian law.

Anything about material sold outside the trial. Trial results describe the sponsor's investigational product in screened participants. They are not evidence about any research reagent sold under the same name, which is the subject of gray-market retatrutide and the evidence.

Every compound named here is discussed as a research subject. Nothing here is medical, dosing, or therapeutic guidance.

Frequently asked questions

What does "phase 2/3" mean? A seamless design that combines a dose-finding or proof-of-concept stage with a confirmatory stage under one protocol. It saves time, especially in rare diseases, but the two stages still answer different questions. TAZPOWER was registered as phase 2/3 with 12 participants, which shows how loosely the label maps onto size.

Is a phase 3 result enough for approval? Not on its own. Regulators review the whole dossier: usually more than one adequate trial, manufacturing quality, and the balance of benefit and harm for a defined population. Approval is granted for a specific indication and formulation in one jurisdiction, and it does not transfer to other uses, countries or products.

Why do so many drugs fail in phase 3 after a good phase 2? Phase 2 trials are small, so their estimates are noisy, and the programmes that advance tend to be the ones whose estimates came out high by chance. Phase 3 enrols a broader population with a stricter analysis, and the effect often shrinks or vanishes, as it did for elamipretide in mitochondrial myopathy.

Do research peptides go through these phases? A compound sold for laboratory research has, by definition, not been approved for human use. Some, like retatrutide, are investigational drugs in a sponsor's trial programme. Others, like KPV, have never entered phase 1. Either way, trial evidence attaches to the investigational product studied, not to research material.

Selected Research References

  1. Urva S, Coskun T, Loh MT, et al. LY3437943, a novel triple GIP, GLP-1, and glucagon receptor agonist in people with type 2 diabetes: a phase 1b, multicentre, double-blind, placebo-controlled, randomised, multiple-ascending dose trial. The Lancet. 2022;400(10366):1869-1881. pubmed.ncbi.nlm.nih.gov/36354040
  2. Jastreboff AM, Kaplan LM, Frías JP, et al. Triple-Hormone-Receptor Agonist Retatrutide for Obesity — A Phase 2 Trial. New England Journal of Medicine. 2023;389(6):514-526. pubmed.ncbi.nlm.nih.gov/37366315
  3. Jastreboff AM, Kaplan LM, Davies MJ, et al.; TRIUMPH-1 Investigators. Retatrutide, a Triple Hormone Receptor Agonist, for Treatment of Obesity. New England Journal of Medicine. 2026. pubmed.ncbi.nlm.nih.gov/42814954
  4. Karaa A, Bertini E, Carelli V, et al. Efficacy and Safety of Elamipretide in Individuals With Primary Mitochondrial Myopathy: The MMPOWER-3 Randomized Clinical Trial. Neurology. 2023;101(3):e238-e252. pubmed.ncbi.nlm.nih.gov/37268435
  5. Kingsberg SA, Clayton AH, Portman D, et al. Bremelanotide for the Treatment of Hypoactive Sexual Desire Disorder: Two Randomized Phase 3 Trials. Obstetrics & Gynecology. 2019;134(5):899-908. pubmed.ncbi.nlm.nih.gov/31599840
  6. Wong CH, Siah KW, Lo AW. Estimation of clinical trial success rates and related parameters. Biostatistics. 2019;20(2):273-286. pubmed.ncbi.nlm.nih.gov/29394327
  7. Gyawali B, Hey SP, Kesselheim AS. Assessment of the Clinical Benefit of Cancer Drugs Receiving Accelerated Approval. JAMA Internal Medicine. 2019;179(7):906-913. pubmed.ncbi.nlm.nih.gov/31135808
  8. Shirley M. Elamipretide: First Approval. Drugs. 2026;86(3):377-383. pubmed.ncbi.nlm.nih.gov/41335372
  9. Reid Thompson W, Hornby B, Manuel R, et al. A phase 2/3 randomized clinical trial followed by an open-label extension to evaluate the effectiveness of elamipretide in Barth syndrome, a genetic disorder of mitochondrial cardiolipin metabolism. Genetics in Medicine. 2021;23(3):471-478. pubmed.ncbi.nlm.nih.gov/33077895
  10. Poolton JT, Minett J, Kazmi A. Human evidence for BPC-157 in orthopaedic and regenerative medicine: a systematic review of efficacy and safety. Acta Orthopaedica Belgica. 2026;92(3). pubmed.ncbi.nlm.nih.gov/42885452

All products sold by NovaEvum are for research and laboratory use only. They are not intended for human or animal consumption, diagnosis, treatment, cure, or prevention of any disease.