The day a drug that stopped at animals
first goes into a person
It worked in mice. So how much do you give a person, to how many people, measuring what?
These twenty lessons take that blueprint apart piece by piece — the 100 µg of phase 0, the 3:1 randomisation of phase I, blinding, escalation rules, adverse-event grades, and what you are and are not entitled to say. The material is the phase I trial of a TNIK inhibitor published in Nature Biotechnology in 2024. Python runs right inside this page.
How you'll learn
A single phase I holds a dozen decisions: five dose levels, 3:1 allocation, seven days of repeat dosing, one high-fat meal. We follow why each was set that way.
Not a made-up example. NCT05154240 — the design and results of a trial 78 real people took part in.
Randomisation and blinding are not manners, they are machinery. You turn the handle yourself and watch what goes wrong without them.
Phase I does not prove efficacy. The last lesson is about how far safety data from 78 people entitles you to go.