Browse Learn
How does a new drug get FDA approval?
Before a new medicine reaches patients, scientists must answer three questions: Is it reasonably safe to test? Does it work? And do its benefits outweigh its risks? Here is how drug testing moves from the laboratory to FDA approval—and what happens afterward.
1. The drug development roadmap
Drug development usually moves from preclinical research into three phases of human testing. The company then sends the results to the US Food and Drug Administration (FDA), which reviews the evidence before deciding whether the medicine can be marketed. Safety monitoring continues after approval. [1]
2. Preclinical testing: Before the first patient
Researchers begin in the laboratory, often using cells, biological systems and animal studies. They ask whether the drug affects its intended target, how it moves through the body and whether it causes harmful effects. These tests help select a starting dose for human studies. [2]
Preclinical results do not prove that a medicine will help people. They help decide whether it is reasonably safe to begin a closely monitored human trial.
Does the drug act on the target scientists expect?
How much drug reaches the body, and for how long?
What side effects appear at different doses?
3. Phase 1: Is it safe enough to keep testing?
Phase 1 is often the first time a new drug is given to people. Studies commonly involve 20–100 participants, frequently healthy volunteers, although some treatments—such as cancer medicines—are first tested in patients. Researchers learn how the body absorbs and removes the drug, what side effects occur and which doses are reasonable to study next. [3]
The main question is usually how to give the medicine safely, not whether it has been proven to treat the disease. Small early signals of benefit can still help researchers plan Phase 2.
4. Phase 2: Does it show signs of working?
Phase 2 trials generally study people who have the disease. They look for preliminary evidence of effectiveness, continue evaluating safety and help choose the dose to take forward. Studies commonly enroll dozens to several hundred patients. [3]
For example, a medicine for high blood pressure might lower blood pressure more than a placebo (a treatment without the active drug). That is encouraging, but a small study may give an uncertain estimate of the true benefit. A larger trial may produce a different result.
Is there a promising signal?
Do patients receiving the medicine improve? Is there a clear dose response?
Will the result hold up?
Was the observed difference real, large enough to matter and reproducible in more patients?
5. Phase 3: Confirm the benefits and risks
Phase 3 studies are usually larger and designed to provide the main evidence needed to judge whether a medicine works for its proposed use. They often involve hundreds to thousands of patients and compare the investigational medicine with a placebo or another treatment. Scientists closely track benefits, side effects and the reliability of the findings. [3]
A good trial decides in advance who will participate, what result counts as success and how the results will be analyzed. Random assignment and blinding, when appropriate, help prevent misleading comparisons. Larger size alone does not make a poorly designed trial trustworthy.
The central FDA requirement
6. What does “substantial evidence of effectiveness” mean?
For a medicine to be approved, the FDA must have adequate evidence that it works for the use described on its label. US law calls this “substantial evidence”. The definition begins with:
In plain English, the FDA expects evidence from studies designed well enough for qualified experts to conclude that the drug really has the claimed effect. Researchers must distinguish a true benefit from chance, bias and other explanations. The evidence must also support the specific disease, patients and conditions described in the proposed label. [4]
Can researchers tell what happened because of the medicine rather than other factors?
Are the study's measurements and analyses strong enough to support a conclusion?
Does the evidence support what the company claims the medicine treats?
Does the law require two successful Phase 3 trials? No. Multiple adequate and well-controlled trials have traditionally been a common path, but the statute allows one adequate and well-controlled clinical investigation plus confirmatory evidence when FDA determines that the package is sufficient. Whether evidence qualifies depends on its quality and the scientific context, not simply the trial's phase number. [4] [5]
FDA published a revised draft guidance on substantial evidence in June 2026. It is draft guidance, not a change to the underlying legal standard. [5]
7. FDA review: Can the drug be marketed?
After testing, the company submits its studies and other information to the FDA. Reviewers examine effectiveness, safety, the proposed label and manufacturing quality. A medicine may show a benefit but still face problems if the risks are too high, the evidence is not reliable or its manufacturing controls are inadequate. [6]
Approval means the FDA judges that the available evidence supports the medicine's benefits outweighing its risks for the specified use—not that the medicine is risk-free or that every unanswered question has been resolved.
8. After approval: Research continues
Once a drug is marketed, it may be used by more people and over longer periods than were studied before approval. Rare side effects, drug interactions or problems in particular patient groups may only become clear later. The FDA monitors safety reports, and companies may conduct required or voluntary post-approval studies. [1]
These studies can answer new questions about long-term safety, real-world use or other medical uses. Approval marks the beginning of broader use—not the end of scientific evaluation.
Sources and further reading
- FDA — The Drug Development Process.
- FDA — Drug Development and Review Definitions (preclinical research, IND and NDA).
- FDA — Step 3: Clinical Research (Phase 1–3).
- 21 U.S.C. § 355(d) — Substantial evidence of effectiveness (statutory definition).
- FDA — Demonstrating Substantial Evidence of Effectiveness for Human Drug and Biological Products (revised draft guidance, June 2026).
- FDA — Drug Review Process: Ensuring Drugs Are Safe and Effective.
Educational overview focused on the usual US pathway for new medicines. Trial designs and regulatory evidence packages differ by disease and product.
Continue learning
Approval is one stage in a longer chain. These guides cover what comes before and after it.
Educational content, not investment advice. Terms