The four phases of FDA approval are Phase 1 (safety testing), Phase 2 (efficacy testing), Phase 3 (large-scale comparison), and Phase 4 (post-market surveillance after the drug is already available to the public). The entire process from Phase 1 through approval takes most drugs over eight years, and only about 13.8% of drugs that enter Phase 1 ever reach the market.
Before any of these phases begin, a drug must go through preclinical development, where researchers test the compound in laboratory and animal studies to confirm it’s reasonably safe to try in humans. The company then submits an Investigational New Drug application to the FDA, which includes data on the drug’s toxicity and pharmacological activity. Only after the FDA reviews this application can human testing begin.
Phase 1: Safety and Dosage
Phase 1 is the first time a drug is tested in people. The goal isn’t to see if the drug works yet. Instead, researchers are trying to answer basic questions: Is it safe? How does the body absorb and process it? What dose can humans tolerate without serious problems?
These trials are small, typically involving 20 to 100 participants. In many cases, those participants are healthy volunteers rather than people with the disease the drug is designed to treat. The study lasts several months, though the average total duration of Phase 1 (including design, enrollment, and analysis) is about 22 months. Roughly 66% of drugs that enter Phase 1 move on to Phase 2.
Phase 2: Efficacy and Side Effects
Phase 2 is where researchers first test whether the drug actually does what it’s supposed to do. Participants at this stage are people who have the disease or condition the drug targets, and the trials typically enroll a few hundred patients. The average duration is about 34 months.
Researchers measure effectiveness differently depending on the disease. In cancer trials, for example, they often look at whether tumors shrink or how long patients go without their disease getting worse. In other fields, the endpoint might be a specific lab value or symptom score. Phase 2 also continues gathering safety data, building a fuller picture of side effects at therapeutic doses.
This phase is one of the biggest hurdles in drug development. Only about 58% of drugs that complete Phase 2 advance to Phase 3. Many drugs show promise in small groups but fail to demonstrate enough benefit to justify the enormous cost of large-scale testing.
Phase 3: Large-Scale Comparison
Phase 3 trials are the definitive test. These are large, randomized, controlled studies conducted across multiple medical centers, enrolling anywhere from 300 to 3,000 or more patients depending on the condition. The drug is compared head-to-head against the current best available treatment (or a placebo, if no standard treatment exists). Average duration is about 45 months.
The FDA typically expects at least two successful Phase 3 trials demonstrating both safety and efficacy before it will approve a drug, though recent regulatory updates have shifted toward accepting a single well-controlled trial paired with confirmatory evidence in some cases. That confirmatory evidence can include data on the drug’s biological mechanism, results in related conditions, or real-world evidence.
About 59% of drugs that enter Phase 3 ultimately gain approval. The combination of all three phases means the overall odds are steep: only about 14 out of every 100 drugs that start human testing ever reach patients. Success rates vary dramatically by field. Vaccines for infectious diseases have the highest overall approval rate at around 33%, while oncology drugs have the lowest at roughly 3.4%.
Phase 4: Post-Market Surveillance
Phase 4 begins after the drug is approved and available to the public. Pre-approval studies involve only several hundred to several thousand patients, which means rare side effects, long-term complications, and interactions with other medications often don’t show up until millions of people are taking the drug in the real world.
The FDA runs several systems to catch these problems. The FDA Adverse Event Reporting System (FAERS) is a database that collects safety reports for all approved drugs and biologics. The MedWatch program allows both healthcare professionals and the general public to voluntarily report serious reactions. Drug manufacturers are also required by law to submit adverse event reports and to flag any manufacturing deviations or quality problems.
Beyond collecting reports, the FDA conducts periodic, unannounced inspections of manufacturing facilities to verify that companies are following proper production standards and adhering to the terms of their approval. A separate program reviews medication error reports to identify patterns, like confusingly similar drug names or packaging that leads to mix-ups. If a drug turns out to cause problems that weren’t visible during clinical trials, the FDA can require label changes, add new warnings, restrict how the drug is prescribed, or pull it from the market entirely.
How Long the Full Process Takes
Adding up the average durations of Phase 1 (22 months), Phase 2 (34 months), and Phase 3 (45 months) gives roughly 8.4 years of clinical testing alone. That doesn’t include the years of preclinical research before human trials begin or the FDA’s review period after Phase 3 data is submitted. From initial discovery to pharmacy shelves, the total timeline for a new drug commonly stretches beyond a decade.
Expedited Pathways That Shorten the Timeline
Not every drug follows the standard timeline. The FDA has several programs that can speed things up for drugs that address serious conditions with few or no existing treatments.
- Fast Track allows more frequent communication with the FDA during development and the option to submit portions of the application on a rolling basis rather than all at once.
- Breakthrough Therapy is for drugs that show substantial improvement over what’s currently available. It includes everything Fast Track offers plus more intensive FDA guidance starting as early as Phase 1.
- Accelerated Approval lets the FDA approve a drug based on a surrogate endpoint, like tumor shrinkage, rather than waiting for long-term outcomes like survival. The company must then conduct post-approval studies to confirm the drug delivers real clinical benefit.
- Priority Review shortens the FDA’s review clock from the standard timeline to six months for drugs that offer significant advances in treatment.
Recent Changes to the Process
Several policy updates issued in early 2026 are reshaping how the FDA evaluates evidence across the drug development process. The most notable shift is the move toward a single-trial standard: one adequate and well-controlled study, supported by confirmatory evidence, is now the default requirement for new drug approval. This is a meaningful departure from the traditional expectation of two successful Phase 3 trials.
The FDA has also introduced a “plausible mechanism” framework for ultra-rare diseases, where patient populations are too small for traditional randomized trials. This pathway is designed primarily for gene editing and RNA-based therapies and allows companies to build an approval case from biological mechanism data rather than conventional clinical endpoints. Additionally, new guidance now supports alternatives to animal testing in preclinical development, including lab-grown organ models, organs-on-chips, and computer simulations, along with clearer rules for using advanced statistical methods that can incorporate real-world data into trial designs.