Developing a new drug takes roughly 10 to 15 years from the earliest laboratory work to pharmacy shelves, with a median of 8.3 years spent in clinical testing alone. The process moves through distinct stages: discovery, preclinical testing, three phases of human trials, regulatory review, and ongoing monitoring after approval. Most candidates fail along the way. Only about 3 to 10 percent of drugs that enter the first phase of human testing ever reach approval, depending on the disease area. The average cost, factoring in those failures and the cost of capital, runs close to $879 million per drug.
Finding a Target and Building a Drug
Every drug starts with a biological target, usually a protein or gene involved in a disease. Researchers identify these targets by mining large datasets of gene expression, genetic associations, and protein activity, looking for molecules that correlate with how a disease starts or worsens. A genetic link between a specific mutation and higher disease risk, for instance, can point to a protein worth targeting. Newer approaches use chemical genomics, where researchers expose cells to thousands of small molecules and watch how the genome responds, revealing which biological pathways matter most.
Once a target is identified, the next step is finding a compound that interacts with it. High-throughput screening tests millions of chemical compounds against the target to find “hits,” which are then refined into “lead” compounds. From there, chemists systematically tweak the molecule’s structure, measuring how each change affects its potency, selectivity, and behavior in the body. Techniques like X-ray crystallography let researchers see exactly how a compound fits into the target protein, guiding more precise modifications. The goal is a molecule that hits the right target strongly, ignores everything else, and can survive the journey through the human body long enough to work.
Before any compound reaches a person, it goes through extensive preclinical testing. This includes lab studies on how the drug is absorbed, distributed, metabolized, and eliminated, along with animal studies to assess toxicity and determine safe starting doses for humans. Only compounds that clear these hurdles move forward.
Phase 1: First Tests in Humans
Phase 1 trials are the first time a drug enters a human body. These studies typically enroll 20 to 100 participants, often healthy volunteers, though people with the target disease sometimes participate when the drug carries known risks (as with cancer treatments). The primary question is safety: how does the drug behave in the body, what side effects appear at increasing doses, and how much can a person tolerate? Researchers start with very low doses, far below what animal data suggest would be effective, and gradually increase them while closely monitoring participants. These trials also generate early data on how the drug is absorbed and cleared, which shapes dosing plans for later studies.
Phase 2: Does It Actually Work?
Phase 2 trials shift the focus from safety to effectiveness. These studies enroll patients who have the disease the drug is intended to treat, typically several hundred people. Researchers test different doses to find the right balance between benefit and side effects. Many Phase 2 trials use a control group that receives either a placebo or an existing treatment, allowing a direct comparison. This is where the majority of drugs fail. The compound may not produce meaningful improvement, or its side effects at effective doses may be too severe. For cancer drugs specifically, only about 5.7 percent of programs that enter Phase 2 ultimately reach approval.
Phase 3: Large-Scale Proof
Drugs that show promise in Phase 2 advance to Phase 3, which involves hundreds to thousands of patients across multiple locations. These trials are designed to confirm effectiveness in a larger, more diverse population and to detect rarer side effects that smaller studies might miss. Phase 3 trials are almost always randomized and controlled, meaning participants are randomly assigned to receive either the new drug or a comparator, and neither the patient nor the doctor typically knows which one they’re getting.
These trials generate the statistical evidence regulators need to make an approval decision. Even at this late stage, about one in four drugs fail. A Phase 3 trial can take several years to complete, particularly for diseases where outcomes develop slowly. For cancer drugs, about 24 percent of programs that reach Phase 3 go on to win approval.
Regulatory Review
After successful Phase 3 results, the drug’s sponsor submits a New Drug Application (NDA) to the FDA. This package includes all animal and human data, analyses, manufacturing details, and proposed labeling. The FDA has 60 days to decide whether the application is complete enough to review. Incomplete submissions can be rejected outright.
For standard drugs, the FDA aims to complete its review within 10 months. Drugs designated as priorities get a six-month review timeline. During review, FDA scientists independently evaluate the clinical data, inspect manufacturing facilities, and assess whether the drug’s benefits justify its risks. An advisory committee of outside experts may weigh in on particularly complex or novel drugs.
Faster Paths for Serious Diseases
Not every drug follows the standard timeline. The FDA offers several programs to speed development for drugs that address serious conditions with limited treatment options.
- Fast Track facilitates more frequent communication between the drug maker and the FDA throughout development, and allows the company to submit sections of its application on a rolling basis rather than all at once.
- Breakthrough Therapy applies to drugs that show substantial improvement over existing treatments in early clinical evidence. It includes all Fast Track benefits plus more intensive FDA guidance on trial design.
- Accelerated Approval allows drugs to be approved based on a surrogate measure (like tumor shrinkage) rather than waiting for long-term outcomes (like survival), on the condition that the company runs confirmatory studies afterward.
- Priority Review shortens the FDA’s review clock from 10 months to 6 months.
These designations can overlap. A single drug might receive both Breakthrough Therapy and Priority Review, shaving years off the path to patients.
Monitoring After Approval
Approval is not the end of the process. Phase 4 studies, sometimes required as a condition of approval, monitor the drug’s performance in the real world. Clinical trials, even large Phase 3 studies, can only detect side effects that occur relatively frequently or within the trial’s timeframe. Rare adverse effects, interactions with other medications, and long-term outcomes only become visible once tens of thousands of people are taking the drug over years. Phase 4 studies may track overall safety in broad patient populations, evaluate long-term efficacy, or investigate the drug’s effects in specific groups that were underrepresented in earlier trials.
What the Numbers Look Like
The full picture of drug development is defined by high failure rates and enormous costs. A study of cancer drugs developed between 2000 and 2015 found that only 3.4 percent of programs that entered Phase 1 ultimately reached approval, despite oncology accounting for 42 percent of all drug development programs. Success rates vary by disease, ranging from near zero to about 10 percent.
A JAMA Network Open analysis of drug development costs from 2000 to 2018 estimated that the direct cost of developing a single successful drug averaged $172.7 million. But that figure doesn’t account for all the drugs that failed. When the cost of failed programs is included, the average rises to $515.8 million. Factor in the cost of capital (the money tied up for a decade or more that could have been invested elsewhere), and the figure reaches $879.3 million per approved drug. Costs varied widely by therapeutic area, from about $72.5 million for genitourinary drugs to $297.2 million for pain and anesthesia treatments, before accounting for failures.
How AI Is Changing the Process
Artificial intelligence is reshaping the earliest stages of drug development. The most visible breakthrough has been AlphaFold, a model that accurately predicts the three-dimensional structures of proteins. Understanding a protein’s shape is essential for designing a drug that fits into it precisely, and this prediction work used to take months or years of laboratory effort. AI now produces reliable predictions in hours.
Beyond structure prediction, AI systems can design entirely new molecules optimized for a specific target, predict how those molecules will behave in the body, and flag potential safety issues before a single lab experiment is run. This “de novo design” approach compresses what used to be years of iterative chemistry into weeks. The potential savings are substantial: by filtering out poor candidates earlier and designing better molecules from the start, AI-driven discovery could reduce both the cost and timeline of the preclinical phase while improving the odds that a drug survives human testing.