Why one trial population still decides the dose for eight billion different livers
For most small-molecule drugs, the liver decides what actually happens once a pill is swallowed. Somewhere between 70 and 80% of small molecules are cleared hepatically — broken down by Phase I enzymes (chiefly the cytochrome P450 family) and Phase II conjugation before whatever’s left reaches systemic circulation. This single biological fact quietly underwrites almost every dose printed on every label in the world.
The trouble is that the liver doing this work is not the same liver everywhere. CYP2D6, CYP2C9, CYP2C19, and CYP3A4/5 — the handful of enzymes responsible for clearing most drugs — are shaped by genetics that vary sharply by ancestry. CYP2C19 poor-metabolizer prevalence, for instance, runs at roughly 15% in East Asian populations versus 2–4% in European, African, or Latin American groups. Layer onto that the effects of age, sex and hormonal status, body weight and composition, diet, chronic infection (malaria, TB, HIV, and viral hepatitis all suppress or induce CYP activity), co-medication with herbal remedies, and lifestyle exposures like smoking and alcohol — and the “standard dose” starts to look less like a scientific constant and more like a regional accident of where the original trial happened to be run.
A Pattern Documented for Decades
This is not a theoretical concern. It has been documented, repeatedly, across decades and continents, in study after study of how differently the same drug behaves in different populations:
| Country / Region | Year | N (Patients) | Study | Key Finding |
|---|---|---|---|---|
| Pakistan (Rawalpindi/Islamabad) | 1989 | 157 | Isoniazid acetylator phenotyping | 31.8% fast acetylators |
| Iran | 2009 | 200 | Isoniazid acetylation, TB vs. healthy | 62% slow acetylators in TB patients |
| Iran & Afghanistan | 1977 | 371 | Isoniazid N-acetylation phenotyping | ~59% slow acetylators, bimodal pattern |
| Tunisia | 2005 | 620 | Isoniazid acetylator phenotype, TB | Dose cut from 5→3 mg/kg recommended |
| Mexico (Monterrey) | 2007 | 111 | Isoniazid acetylation, pulmonary TB | 60% rapid / 40% slow acetylators |
| India | 2023 | 194 | Isoniazid N-acetylation phenotyping | 70% slow acetylators — among the highest globally |
| Sweden | — | 310 | Isoniazid acetylator phenotype by age/sex | Metabolism slows further with age in men |
| 9 countries (US, Sweden, Japan, Taiwan, Singapore, etc.) | 2009 | 5,700 | International Warfarin Pharmacogenetics Consortium | Single global dose deemed unsafe |
| Japan / China / Caucasian | Ongoing | 54–135/cohort | Ethnobridging PK studies | Exposure differs by ethnicity at identical mg dose |
| Japan (PMDA national review) | 2001–2007 | 137 approved drugs | Dose-ratio analysis vs. US/EU | 31% of approvals used a lower dose in Japan |
Table 1. Ten documented dosing and metabolism studies spanning six continents and five decades, compiled from published pharmacogenomic and regulatory literature.
Ten studies, six continents, five decades — and the same conclusion each time: metabolism is not a constant that travels safely with the drug across a border.
The Patient Has Also Changed
What makes this more urgent now than when most of these studies were run is that the patient has changed too. Global age-standardised obesity prevalence climbed from under 1% in 1975 to nearly 8% in boys by 2016, with the steepest rise concentrated in East and South Asia, Central Latin America, and the Pacific — regions whose diets have shifted just as fast toward processed, energy-dense food. Alcohol tells a similar story in reverse-mirror: Europe’s per-capita consumption fell 23% between 1980 and 2020, while China’s more than doubled (4.1 to 7.2 litres) and India’s more than doubled again (2.4 to 5.7 litres) between 2000 and 2016. The population a drug was dosed against in 1985 is, in a meaningful biological sense, no longer the population taking it in 2026.
More than 190 national regulatory authorities continue to accept a single global dose, built on one narrow trial population, for continents that dose has never been bridged to.
Millions of physicians prescribe that same label dose every day — not out of indifference, but because no one has handed them a better number to write instead.
A Correction Already Within Reach
The science to close this gap already exists. Bridging studies, PK/PD cohorts, and ethnic-sensitivity frameworks such as ICH E5 have been standard regulatory science since the late 1990s. What has been missing is the will to apply them proactively, as a normal part of launch planning, rather than reactively, after an adverse-event signal forces the question.
For pharmaceutical companies with the pipelines and the global reach to do this at scale, continent-based dose-finding is not a regulatory burden — it is a chance to get the dose right the first time, reduce adverse drug reactions, lower the effective cost of therapy in price-sensitive markets, and lead an industry correction that is, at its core, about treating people as they actually are rather than as a decades-old trial assumed them to be.
It’s time the dose remembered the world.
The author is Chief Operating Officer, Meraki Life Sciences (SMC Pvt Ltd). He can be reached at sohailmanzoor@merakils.org