Developing a new medicine takes an average of 12–15 years and costs over £1 billion, yet fewer than one in ten candidate drugs that enter human trials ever reach patients. This rigorous, staged testing process exists because a drug that works must also be safe — and history shows that shortcuts have devastating consequences.
Where do potential new drugs come from?
Drug discovery begins long before any testing on living organisms. Scientists identify a target — usually a protein involved in a disease — and search for molecules that interact with it.
Sources of potential drugs include:
- Plants and microorganisms — penicillin was discovered from the mould Penicillium notatum; aspirin originated from willow bark (salicylic acid). Many cancer drugs derive from natural compounds.
- Computer modelling — scientists use structural data about a target protein to design molecules virtually, predicting binding without synthesising anything yet.
- Combinatorial chemistry — thousands of molecular variants are synthesised and screened rapidly in automated assays.
- Repurposing existing drugs — a drug approved for one condition may be tested against another (e.g. thalidomide, originally a sedative, is now used for certain cancers and leprosy complications under strict controls).
What happens during pre-clinical testing?
Before any human receives a candidate drug, it must pass pre-clinical tests:
- Cell cultures — the drug is applied to isolated human or animal cells in the laboratory to assess basic toxicity and whether it has the desired biological effect.
- Tissue testing — effects on small pieces of organ tissue are studied.
- Animal testing — the drug is given to laboratory animals (typically mice, rats, and sometimes primates) to assess toxicity, dosage range, and side effects in a living system.
Pre-clinical testing filters out the vast majority of candidates — those that are too toxic, unstable, or ineffective do not proceed. However, results in animals do not always predict human responses accurately, which is why human clinical trials are essential.
What are the phases of a clinical trial?
A clinical trial tests a drug in human volunteers. There are four phases:
| Phase | Participants | Purpose |
|---|---|---|
| Phase 1 | Small group (~20–80 healthy volunteers) | Safety: find safe dose range; identify side effects |
| Phase 2 | Larger group (~100–300 patients with the condition) | Efficacy and further safety at therapeutic doses |
| Phase 3 | Large group (~1,000–3,000+ patients) | Compare with existing treatments; large-scale safety data; regulatory approval sought |
| Phase 4 | General population (post-approval) | Long-term and rare side effects monitored after the drug is on the market |
A drug must pass each phase before moving to the next. Regulatory bodies (such as the MHRA in the UK) review the full trial data before granting approval for general prescription.
What is a double-blind trial and why is it used?
A double-blind, randomised controlled trial is the gold standard for Phase 2 and 3 testing. In this design:
- Patients are randomly assigned to either the treatment group (receive the new drug) or the control group (receive a placebo — an identical-looking tablet or injection containing no active ingredient).
- Neither the patients nor the doctors administering the treatment know who is receiving what — both are "blind" — until the trial ends and the code is broken.
This design eliminates two major sources of bias:
- Patient bias (the placebo effect) — patients who believe they are receiving a real treatment often report feeling better simply because of that belief. A placebo control separates genuine drug effects from this expectation effect.
- Observer bias — doctors who know which patients are receiving the drug may (unconsciously) assess their progress more positively. Blinding the doctors prevents this.
Why is peer review important in drug approval?
Before trial results can influence medical practice or regulatory decisions, they must be published in a peer-reviewed scientific journal. Peer review means that independent scientists with relevant expertise assess the methods, data, and conclusions for validity before publication.
Peer review does not guarantee that results are correct — biases, errors, and even fraud can pass through — but it provides a critical quality check. Regulatory bodies such as the MHRA require that multiple independent trials support a drug's efficacy and safety before approval.
What went wrong with thalidomide?
Thalidomide was developed in the 1950s as a sedative and treatment for morning sickness in pregnancy. It was approved and prescribed widely in the UK and Europe without adequate testing in pregnant animals. Tragically, it caused severe limb abnormalities in thousands of babies born to mothers who took it in early pregnancy — it was a teratogen (causes birth defects).
Thalidomide was withdrawn in 1961. This disaster directly led to the much stricter multi-phase testing requirements now in place globally. The case is a warning about the danger of bypassing rigorous pre-clinical and clinical testing, and about the importance of testing drugs specifically in pregnant subjects when the drug might be used in pregnancy.
Frequently asked questions
What is a placebo and why does it matter?
A placebo is an inactive substance (such as a sugar tablet or saline injection) that looks and feels identical to the real treatment. It is given to the control group in a clinical trial so that any improvement in the placebo group can be measured and subtracted from the results. The difference in outcomes between the treatment group and the placebo group then represents the genuine effect of the drug. Without a placebo control, it is impossible to know how much of any reported improvement is due to the drug itself rather than to the patient's expectation of getting better.
Why is it unethical to test drugs only on animals?
Animal testing provides useful data on toxicity and biological effects, but it has ethical costs — it involves causing harm or distress to living creatures. Scientists apply the "3Rs" framework: Replace animal tests with alternatives (cell cultures, computer models) wherever possible; Reduce the number of animals used; Refine procedures to minimise suffering. Animal testing is currently still required for regulatory approval because cell cultures and models cannot yet replicate the full complexity of a living system, but the field is actively developing alternatives.
What does "randomised" mean in a randomised controlled trial?
Randomisation means that each participant is assigned to the treatment or control group by chance (e.g. using a random number generator), not by the choice of the doctor or patient. This ensures that known and unknown factors that might affect the outcome (age, severity of disease, other health conditions) are distributed roughly equally between groups. Without randomisation, the sicker patients might all end up in one group, making any comparison between groups meaningless.
How long does it take for a drug to reach patients?
The full development process — from initial discovery through pre-clinical testing, all four clinical trial phases, regulatory review, and manufacturing — typically takes 12–15 years. Only about 1 in 5,000 to 10,000 initial candidate compounds ever reaches a patient as an approved medicine. In rare emergencies (such as a pandemic), regulators can grant conditional or accelerated approval based on Phase 2–3 data, with ongoing monitoring required.
For Socratic GCSE biology with Professor Darwin — questioning every claim from first principles to clinical evidence — visit aitutors.me.