In the 1950s, American pharmacologist Elliot Proctor Josiah noticed something strange: some patients taking the antimalarial drug primaquine developed severe anaemia, while others tolerated the same medicine without any ill effects. Genetics at the time could not explain why. But that observation became one of the first bricks in the foundation of the science now called pharmacogenomics.
Pharmacogenomics studies how genetic differences between people affect their response to drugs. This is not exotic territory. It explains why one antidepressant works for one person and causes severe side effects in another. Why some people need double the warfarin dose, while others need a quarter of the standard amount. Why caffeine energises some and triggers anxiety in others.
When you take a tablet, it goes through several stages: absorption in the gut, distribution to tissues, metabolism (breakdown) in the liver, and elimination from the body. Each stage involves proteins — transporters and enzymes — encoded by genes. If a gene differs slightly from the 'standard' version, the protein works differently: faster, slower, or not at all.
The cytochrome P450 enzyme system is the main metabolic hub for most drugs. Approximately 70–80% of all clinically used medicines are metabolised with the involvement of enzymes from this family. CYP2D6, CYP2C19, CYP2C9, CYP3A4 — these are not abstract labels but specific enzymes whose variants determine what happens to a drug in your body.
Poor metabolisers: the enzyme works weakly or not at all. The drug accumulates in the blood, concentration builds, and a standard dose becomes toxic. For CYP2D6, approximately 5–10% of Europeans fall into this category.
Intermediate metabolisers: the enzyme works at reduced capacity. The drug is processed more slowly than normal, but toxicity is usually absent — a dose adjustment is simply needed.
Normal metabolisers (also called extensive metabolisers): standard enzyme activity. This is the group for which drug doses are calibrated in clinical trials. Most people are here.
Ultrarapid metabolisers: the enzyme works with excessive activity. The drug is broken down so quickly that a standard dose has no chance to work. For CYP2D6, approximately 1–2% of Europeans fall here, but among North African and Middle Eastern populations, the figure reaches 20–30%.
Codeine and CYP2D6. Codeine is a prodrug: inactive on its own, it only works after being converted to morphine by the CYP2D6 enzyme. In poor metabolisers, this conversion barely happens — codeine fails as a painkiller. In ultrarapid metabolisers, the reverse: conversion is so rapid that dangerously high morphine concentrations are produced. In 2013, the FDA warned of fatal cases in children with the ultrarapid metaboliser genotype whose mothers were taking codeine while breastfeeding.
Warfarin and CYP2C9/VKORC1. Warfarin is an anticoagulant and one of the most widely prescribed drugs in the world. Its therapeutic window is extremely narrow: too little — thrombosis; too much — haemorrhage. Traditionally, dose calibration took weeks of trial and error. Genetic testing for CYP2C9 and VKORC1 allows the right dose to be estimated from the outset. Approximately 30% of adverse reactions to warfarin are estimated to have a genetic basis.
Clopidogrel and CYP2C19. Clopidogrel is prescribed after myocardial infarction and stenting to prevent thrombosis. It is also a prodrug, activated by CYP2C19. Approximately 25–30% of Europeans carry a reduced-function CYP2C19 variant — in these patients, the drug works substantially less effectively. The FDA requires this information to be stated on the label. An alternative antiplatelet agent is recommended for such patients.
Antidepressants and CYP2D6/CYP2C19. Most SSRI antidepressants (fluoxetine, sertraline, citalopram, escitalopram) are metabolised by these enzymes. In poor metabolisers, blood concentrations are substantially higher than normal — hence the frequent complaints about side effects at 'standard' doses. Selecting the right antidepressant traditionally takes months. Pharmacogenomic testing can substantially shorten this period.
Not all drug reactions are related to metabolism. Some involve the immune system. HLA (Human Leukocyte Antigen) genes encode proteins that present foreign molecules to immune cells. If a drug 'hides' inside such a protein and an immune cell recognises it as a threat — a severe hypersensitivity reaction occurs.
Abacavir (an HIV drug) and HLA-B*57:01. Carriers of this gene variant who take abacavir develop a severe, potentially fatal hypersensitivity reaction. Testing for HLA-B*57:01 before prescribing abacavir is now a global standard of care. This is one of the most compelling examples of pharmacogenomics saving lives right now.
Carbamazepine (an anti-epileptic) and HLA-B*15:02. Carriers of this variant — predominantly in South-East Asian and Chinese populations — face a high risk of Stevens-Johnson syndrome, a severe skin condition with a mortality rate of up to 30%. Since 2007, the FDA has recommended testing before prescribing carbamazepine to patients of Asian descent.
According to the FDA, more than 250 drugs carry pharmacogenomic recommendations in their official labelling — from anticoagulants to cancer drugs, from antidepressants to anaesthetics. The Clinical Pharmacogenomics Implementation Consortium (CPIC) publishes regularly updated guidelines on genotype-guided dosing.
The reality: in many developed countries, pharmacogenomic testing before prescribing certain drugs is already a standard of care. In others, it remains the domain of specialist centres. The technology is available. The question is the pace of integration into routine practice.
A pharmacogenomic test is a DNA analysis that determines your genotype for the key genes that influence drug metabolism. The result is not a disease diagnosis — it is a profile: how quickly your body processes various classes of drugs, and whether you carry variants associated with elevated risk of drug reactions.
This profile does not become obsolete — your genotype does not change. Tested once, you have information valid for life. When a new drug is prescribed — particularly from classes with known pharmacogenomic interactions — this information allows the prescriber to choose the right drug and dose from the start, rather than by trial and error.
Drugs are developed for the 'average' patient. But average patients do not exist. There are people with specific genotypes that determine exactly how their body will respond to a specific drug. Pharmacogenomics is the tool that moves medicine from 'let's try and see' to 'we know in advance'. This is not the future. It is already working for hundreds of drugs right now.