Two people walk into a pharmacy with the same prescription — say, codeine for post-surgical pain. One feels relief. The other feels nothing for hours, then suddenly experiences severe opioid effects. A third person, thanks to a CYP2D6 ultra-rapid metabolizer phenotype, is at risk of a fatal overdose at the standard dose.
This isn't bad pharmacy. It's pharmacogenomics — the difference your genes make in how you respond to medications.
The basics
Pharmacogenomics (PGx) is the study of how genetic variation affects drug response. Two main mechanisms:
- Pharmacokinetics — how fast your body metabolizes the drug. The CYP450 enzyme family (CYP2D6, CYP2C19, CYP3A4, etc.) metabolizes about 75% of clinically used drugs. Variants in these genes change enzyme activity from "poor metabolizer" (drug builds up) to "ultra-rapid metabolizer" (drug clears too fast).
- Pharmacodynamics — how the drug interacts with its target receptor or downstream pathway. Variants in receptors (OPRM1 for opioids, MTHFR for methotrexate) can shift response.
A third mechanism gets less attention but matters most for safety: immune-mediated hypersensitivity. Variants in HLA genes (HLA-B15:02, HLA-B57:01) predict severe drug reactions including Stevens-Johnson Syndrome and DRESS.
Why this isn't fringe science anymore
The FDA includes pharmacogenomic information on over 280 drug labels in 2026. The Clinical Pharmacogenetics Implementation Consortium (CPIC) publishes evidence-graded guidelines for over 100 drug-gene pairs.
For some drugs, genetic testing is now standard of care:
- HLA-B*57:01 before abacavir (HIV) — failure to test is below standard of care
- HLA-B*15:02 before carbamazepine in patients of Han Chinese, Thai, or Filipino ancestry
- DPYD before fluoropyrimidine chemotherapy (5-FU, capecitabine)
- TPMT/NUDT15 before thiopurine drugs (azathioprine, mercaptopurine)
- CYP2C19 for clopidogrel after stenting
- CYP2D6 for codeine in pediatric patients (after 2013 FDA boxed warning)
What a pharmacogenomic test actually tells you
For each drug-gene pair, a PGx test gives you:
- Phenotype — your metabolic status (poor / intermediate / normal / rapid / ultra-rapid)
- Recommended action — standard dose, reduced dose, avoid, alternative drug
- Evidence level — CPIC Level A (strong) through D (limited)
It does not tell you to change your medication. That decision belongs to your prescriber. But it gives them an evidence-based starting point for dose selection.
Common pharmacogenomic findings
Some examples of how often these variants show up:
- ~30% of people are CYP2D6 intermediate or poor metabolizers
- ~30% are CYP2C19 intermediate or poor metabolizers
- ~3-5% carry reduced-function DPYD variants (life-threatening 5-FU toxicity)
- ~7% are TPMT/NUDT15 intermediate metabolizers
If you've ever wondered why a medication "didn't work for you" or why side effects were worse than your doctor expected, PGx is worth checking.
How Atlagene's PGx report works
Atlagene's pharmacogenomics analysis covers 200+ drug-gene interactions following CPIC Level A/B guidelines. We include:
- Major CYP enzymes (CYP2D6, CYP2C19, CYP2C9, CYP3A5, CYP2B6)
- HLA hypersensitivity alleles (HLA-B15:02, HLA-B57:01, HLA-B*58:01)
- Chemotherapy safety variants (DPYD, TPMT, NUDT15, UGT1A1)
- Anticoagulant pharmacogenomics (CYP2C9 + VKORC1 for warfarin)
- Statin transport (SLCO1B1, ABCG2)
Population-equity coverage is built in: African-ancestry CYP2C9 *5/*6/8/11, CYP2D617, CYP2B618, and additional CYP2C19 alleles common in non-European populations.
How to use a PGx report
- Save it. Bring it to medical appointments.
- When your doctor prescribes something new, mention "I have pharmacogenomic data — should we check before I fill this?"
- If you have an FDA boxed-warning variant (HLA-B15:02, HLA-B57:01, DPYD), put it on file with your primary care and any specialists.
- Don't change medications based on the report alone. CPIC guidelines exist; your doctor follows them.
What this isn't
Pharmacogenomics doesn't predict whether a drug will work for your specific condition. It predicts dose-response and safety based on metabolism and immune-system markers. Plenty of medications work or fail for non-genetic reasons — drug interactions, adherence, comorbidities, or simply individual variation.
PGx narrows the search space. It doesn't replace clinical judgment.