# How a Drug Goes from Lab to Pharmacy
Out of roughly 10,000 molecules that a company screens at the start, only one, on average, reaches a pharmacy shelf. The journey typically takes 10 to 15 years and, by the most-cited industry estimate (from the Tufts Center for the Study of Drug Development), costs well over a billion dollars per approved drug once you count all the failures along the way.
That single fact explains most of pharma's behavior: the high prices, the patent obsession, the mergers, the caution. Let's walk the value chain and see where the money and the risk actually sit.
Everything starts with a target: a biological mechanism (often a protein) that a disease depends on. Scientists then hunt for a molecule that hits that target.
Two broad approaches:
Teams screen thousands of candidates, then optimize the promising ones. This phase can take 3 to 6 years. No patients yet. Just chemistry, biology, and a lot of dead ends.
The stake: picking the wrong target wastes years. Most failures later in the chain trace back to a target that never really mattered to the disease.
Before any human touches the drug, it is tested in the lab and in animals to check two things: does it look effective, and is it safe enough to try in people?
In the US, the company then files an IND (Investigational New Drug)
The stake: many toxic or useless candidates die here, which is good. Killing a bad drug early is cheap. Killing it in Phase 3 is catastrophic.
Human testing runs in three phases. Each is bigger, costlier, and more revealing than the last. The FDA's own overview is a clear, free primer.
A small group (typically 20 to 100 healthy volunteers) receives the drug. The question is tolerability and dosing, not whether it cures anything. What dose is safe? How does the body process it?
A few hundred patients who actually have the disease. Here the company looks for a signal of effectiveness and watches side effects closely. Many drugs fail here because the promising lab result simply does not hold up in real patients.
Hundreds to thousands of patients, often across many countries and hospitals. These trials are usually randomized and controlled: patients are randomly assigned to the new drug or to a comparison (a placebo or the current standard treatment), so the results are trustworthy.
Phase 3 is where budgets explode. A single large Phase 3 program can cost hundreds of millions of dollars.
The stake: this is the industry's cliff. A drug can pass Phase 1 and Phase 2, then fail Phase 3, wiping out a decade of investment. Historically, only around 10% of drugs that enter Phase 1 ever get approved. That attrition is why one winner has to pay for many losers.
If the trials succeed, the company submits the full evidence package to regulators:
Reviewers scrutinize the data, the manufacturing plans, and the proposed label. They can approve, reject, or ask for more studies. The FDA also convenes independent advisory committees for tough cases; these panels vote publicly, though the FDA makes the final call.
Approval is not global. A drug cleared by the FDA still needs EMA approval to sell in Europe, and separate approvals elsewhere (Japan's PMDA, China's NMPA, and so on). Each regulator can reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → a different conclusion.
The stake: timing. Every month of delay is a month of patent life burning while the drug earns nothing. Regulators offer accelerated pathways (the FDA's Breakthrough Therapy and Fast Track designations, for example) for drugs targeting serious unmet needs.
Knowledge check
1. Why does the extremely low success rate—roughly one approved drug per 10,000 screened molecules—shape so much of pharma's business behavior?
2. A team is deciding between developing a small molecule versus a biologic. Which statement best captures the core trade-off?
3. Why does the lesson describe killing a bad drug in preclinical as 'cheap' but killing it in Phase 3 as 'catastrophic'?
4. What does choosing the 'target' at the discovery stage fundamentally represent, and why is it so consequential?
5. Select ALL correct answers about the preclinical stage and the IND application.
Select all the correct answers.
6. Select ALL correct answers about the drug development value chain as described.
Select all the correct answers.
Approval means you are allowed to sell the drug. Now you have to actually make it, at scale, at consistent quality, every single batch.
This is harder than it sounds, especially for biologics grown in living cells, where tiny changes in temperature or process can alter the product. Facilities must follow GMP (Good Manufacturing Practice), enforceable quality rules that regulators inspect. A failed inspection can shut down supply.
Two realities shape this stage:
The stake: you can have a brilliant drug and still fail commercially if you cannot supply it reliably.
Now the drug reaches doctors and patients. This involves:
Meanwhile the clock is ticking on the patent, typically 20 years from filing, but much of that is consumed during development. A drug may reachreachThe number of unique people exposed to your message in a given period. Unlike impressions, reach counts each person once, no matter how often they see it.View full definition → the market with only 8 to 12 years of exclusivity left.
When patents expire, cheap generics (for small molecules) or biosimilars (for biologics) enter, and revenue can drop sharply. This moment is called the patent cliff, and it drives much of the industry's urgency to keep discovering the next drug.
The stake: the commercial window is short. Companies must recoup the entire cost of development, including all the failures, before exclusivity ends.
Now the pattern makes sense:
None of this is an endorsement or a criticism. It is the logic that a 10 to 15 year, high-attrition, heavily regulated pipelinepipelineAll active sales opportunities across the stages of the sales process, together with their combined potential value and probability of closing.View full definition → produces.