CYP2D6 Activity Score Interpreter
CYP2D6 Activity Score Interpreter
Add the two allele activity values and read the sum against the current CPIC boundaries — which moved in the 2019–2020 CPIC/DPWG consensus, taking an activity score of 1 out of normal and into intermediate. The score belongs to the gene, not to the drug, and the page asks which drug you mean because the consequence inverts.
CYP2D6 activity score
Two allele values → metaboliser phenotypeA CYP2D6 *1/*4 genotype — one normal-function allele and one no-function allele — in a patient being considered for codeine
The score, and the boundaries it is read against
a duplicated allele contributes its value × its copy number
- 0
- no function — *3, *4, *5, *6, *36, *40. *4 is the commonest no-function allele in European populations; *5 is a whole-gene deletion
- 0.25
- severely decreased function, and *10 is the only allele in this class. The value was 0.5 before the 2019–2020 consensus. *10 is the commonest CYP2D6 allele in East Asian populations, so this single change moved a large number of people between phenotypes
- 0.5
- decreased function — *9, *17, *29, *41. *17 and *29 are common in African populations, *41 in European and South Asian
- 1
- normal function — *1, *2, *35
- the boundaries
- Activity score 0 is poor, 0.25 to 1 is intermediate, 1.25 to 2.25 is normal, and above 2.25 is ultrarapid. Written as CPIC writes it: poor is 0, intermediate is greater than 0 and less than 1.25, normal is 1.25 to 2.25 inclusive, ultrarapid is greater than 2.25
- what the score cannot see
- phenoconversion. A strong CYP2D6 inhibitor — bupropion, fluoxetine, paroxetine, quinidine, terbinafine — converts a genotypic normal metaboliser into a phenotypic poor one, and no genotype-based score accounts for it. Check the drug list before you trust the phenotype
Worked example
A CYP2D6 *1/*4 genotype — one normal-function allele and one no-function allele — in a patient being considered for codeine
*1 is a normal-function allele: activity value 1
*4 is a no-function allele: activity value 0
1 + 0 = activity score 1.00
Against the current CPIC boundaries, 1 falls in the intermediate metaboliser band (greater than 0, less than 1.25)
Under the CPIC scheme in use before the 2019–2020 consensus, the normal band began at 1, so the same genotype was reported as a normal metaboliser. Nothing about the patient changed
Codeine is a prodrug, so intermediate metabolism means reduced conversion to morphine: CPIC advises the label dose but says to consider an alternative analgesic if the response is inadequate
Change the drug to ondansetron and the reading changes completely. Ondansetron is inactivated by CYP2D6, so reduced activity means slightly higher exposure, not lower — and CPIC makes no recommendation for intermediate metabolisers on that drug because the evidence is insufficient
Activity score to phenotype — before and after the consensus
| Activity score | CPIC before the 2019–2020 consensus | CPIC now | Example genotype |
|---|---|---|---|
| 0 | Poor | Poor | *4/*4, *4/*5 |
| 0.25 | not defined (*10 scored 0.5) | Intermediate | *10/*4 |
| 0.5 | Intermediate | Intermediate | *41/*4 |
| 0.5 (*10/*10) | Normal (as 1.0 under the old *10 value) | Intermediate | *10/*10 |
| 1 | Normal | Intermediate | *1/*4, *41/*41 |
| 1.25 | not reachable under the old values | Normal | *1/*10 |
| 1.5 | Normal | Normal | *1/*41 |
| 2 | Normal | Normal | *1/*1, *1/*2 |
| 2.25 | not reachable under the old values | Normal | *2×2/*10 |
| 3 | Ultrarapid | Ultrarapid | *1×2/*1 |
The same score, two opposite consequences
| Phenotype | Codeine or tramadol (prodrug: CYP2D6 makes the active drug) | Ondansetron (active drug: CYP2D6 destroys it) |
|---|---|---|
| Ultrarapid | Avoid. Too much morphine is formed — CPIC: avoid “because of potential for serious toxicity” | Avoid. The drug is destroyed too fast — CPIC: select an alternative not predominantly metabolised by CYP2D6, such as granisetron, because of increased risk of vomiting |
| Normal | Label dose | Standard starting dose |
| Intermediate | Label dose; consider an alternative if analgesia is inadequate | No recommendation — insufficient evidence |
| Poor | Avoid. Too little morphine is formed — CPIC: avoid “because of possibility of diminished analgesia” | No recommendation — insufficient evidence |
Why the boundaries moved, and why the score is not a recommendation
The CYP2D6 activity score exists because star-allele nomenclature does not add up. A laboratory reports a diplotype — *1/*4, *10/*41, *2×2/*5 — and the clinician needs a phenotype. The activity score bridges them by giving each allele a number for how much functioning enzyme it produces, summing the two, and reading the sum against boundaries. It is the instrument that made CYP2D6 pharmacogenomics usable, and it is the model every other pharmacogene’s phenotype assignment has since copied.
The boundaries are not where they were. A CPIC and Dutch Pharmacogenetics Working Group consensus, worked through in 2019 and published in 2020, standardised the translation and made two changes at once. It moved the line between intermediate and normal up, so that an activity score of exactly 1 is now an intermediate metaboliser where it used to be a normal one — and an activity score of 1 is the score of *1/*4, one of the commonest genotypes in European populations. And it gave *10 its own class, “severely decreased function”, with an activity value of 0.25 instead of the 0.5 that every reduced-function allele previously carried. *10 is the commonest CYP2D6 allele in East Asian populations, so that second change moved a large number of people too. The two changes interact: *10/*10 scored 1.0 and read as normal before, and scores 0.5 and reads as intermediate now. Nothing about anybody’s enzyme changed. If you are comparing a new report against an old one, compare the genotypes, not the phenotypes.
The second thing this page exists to say is that the activity score is a property of the gene and not of any drug. This is stated everywhere and misread constantly, and the misreading has a direction: people assume that more enzyme is better and less is worse. For a prodrug that is exactly backwards at one end. Codeine does almost nothing until CYP2D6 turns it into morphine, so the ultrarapid metaboliser is the dangerous one — this is the genotype behind the reported deaths of children given codeine after tonsillectomy, and the breastfed infant whose mother was an ultrarapid metaboliser taking codeine — and the poor metaboliser simply gets no analgesia. For a drug that CYP2D6 destroys, the arrows reverse: the ultrarapid metaboliser loses the drug too quickly and the risk is treatment failure. CPIC’s 2026 recommendation for ondansetron in ultrarapid metabolisers is to use a different antiemetic altogether, for precisely that reason. Same score, opposite hazard, which is why this page asks which kind of drug you mean before it says anything.
Three limitations are worth carrying away. The score cannot be right if the genotype is incomplete: a platform that does not test copy number cannot call an ultrarapid metaboliser at all, and will report a duplication carrier as normal. The score says nothing about phenoconversion — a strong CYP2D6 inhibitor such as bupropion, fluoxetine, paroxetine or terbinafine turns a genotypic normal metaboliser into a phenotypic poor one, and no genotype-based number will show it. And activity values are assigned by allele, so a novel or unassigned star allele has no value and leaves the score undefined rather than zero. Related pages: the ACMG variant classification interpreter for classifying a variant rather than an allele, and the NGS variant confirmation interpreter for whether a genotype call needs orthogonal confirmation before it is reported.
Frequently asked questions
What activity score is a CYP2D6 poor metaboliser?
Zero. CPIC defines the poor metaboliser as an activity score of 0 — two no-function alleles, such as *4/*4 or *4/*5. Anything above 0 and below 1.25 is an intermediate metaboliser, 1.25 to 2.25 inclusive is a normal metaboliser, and above 2.25 is an ultrarapid metaboliser. Because allele activity values move in steps of 0.25, the intermediate band in practice means an activity score of 0.25, 0.5, 0.75 or 1.
Did the CYP2D6 intermediate and normal metaboliser boundaries change?
Yes, in the CPIC and Dutch Pharmacogenetics Working Group consensus worked through in 2019 and published in 2020. The normal metaboliser band used to start at an activity score of 1 and now starts at 1.25, so a score of exactly 1 — the score of the common *1/*4 genotype — moved from normal to intermediate. At the same time *10 was reclassified as severely decreased function with an activity value of 0.25 instead of 0.5. A report issued before the consensus can therefore carry a different phenotype for the same genotype, and the genotype is the part to re-read.
Why is an ultrarapid metaboliser dangerous for codeine but not for every drug?
Because codeine is a prodrug. It has little analgesic effect of its own and depends on CYP2D6 to convert it to morphine, so more enzyme means more morphine, faster — which is why CPIC advises avoiding codeine and tramadol in ultrarapid metabolisers because of the potential for serious toxicity. For a drug that CYP2D6 destroys rather than creates, the same genotype causes the opposite problem: ondansetron is cleared too quickly, exposure falls, and CPIC’s 2026 recommendation is to use granisetron instead because of an increased risk of vomiting. The activity score describes the enzyme. Whether a high score is good or bad depends on which side of the reaction the useful molecule sits.
How is the CYP2D6 activity score calculated?
Give each of the two alleles its published activity value — 0 for a no-function allele, 0.25 for *10, 0.5 for a decreased-function allele such as *9, *17, *29 or *41, and 1 for a normal-function allele such as *1, *2 or *35 — and add them. A duplicated or multiplied allele contributes its value times its copy number, so *1×2 counts 2 and *41×2 counts 1. The sum is the activity score.
Can a genotype test miss an ultrarapid metaboliser?
Easily, and this is the commonest false-normal on a CYP2D6 report. An activity score above 2.25 essentially requires a duplication or multiplication of a functional allele, and copy number is not measured by every genotyping platform. A platform that calls only single-nucleotide variants will report a *1×2/*1 individual as *1/*1 and assign a normal phenotype. If an ultrarapid result matters clinically — codeine in a child, tramadol, tamoxifen — check that the assay reports copy number, and check that it can say which allele was duplicated, because a duplicated no-function allele adds nothing.
Does the activity score account for drug interactions?
No, and this is a real gap rather than a technicality. Strong CYP2D6 inhibitors — bupropion, fluoxetine, paroxetine, quinidine, terbinafine — reduce enzyme activity so far that a genotypic normal metaboliser behaves as a phenotypic poor metaboliser, a phenomenon called phenoconversion. The activity score is computed from DNA and cannot see it. Any patient on one of those drugs should have their medication list read alongside the genotype, and in that situation the genotype is the weaker of the two pieces of information.
Related calculators
References
- Caudle KE, et al. Standardizing CYP2D6 Genotype to Phenotype Translation: Consensus Recommendations from the Clinical Pharmacogenetics Implementation Consortium and Dutch Pharmacogenetics Working Group. Clin Transl Sci. 2020. doi:10.1111/cts.12692. The consensus that moved the intermediate/normal boundary and revalued *10.
- Moore CA, et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6 Genotype and Use of 5-HT3 Receptor Antagonists: 2026 Update. Clin Pharmacol Ther. 2026. doi:10.1002/cpt.70291. Table 1 prints the current activity score bands as “0”, “0<x<1.25”, “1.25≤x≤2.25” and “>2.25”; and for ultrarapid metabolisers, “Select alternative drug not predominantly metabolized by CYP2D6 (i.e., granisetron)” because of “increased metabolism to less active compounds” and “increased risk of vomiting”.
- Crews KR, et al. Clinical Pharmacogenetics Implementation Consortium Guideline for CYP2D6, OPRM1, and COMT Genotype and Select Opioid Therapy. Clin Pharmacol Ther. 2021. The same four activity score bands, and for codeine and tramadol: ultrarapid, “Avoid codeine use because of potential for serious toxicity”; poor, “Avoid codeine use because of possibility of diminished analgesia”.
- Gaedigk A, et al. The CYP2D6 activity score: translating genotype information into a qualitative measure of phenotype. Clin Pharmacol Ther. 2008. The original activity score.
- CYP2D6 Overview: Allele and Phenotype Frequencies. In: Medical Genetics Summaries. Bethesda (MD): National Center for Biotechnology Information; NBK574601. “Ultrarapid metabolizers (UM) have a diplotype with an activity score greater than 2.25”; “Normal metabolizers (NM) have an activity score of 1.25 to 2.25”; “Intermediate metabolizers (IM) have an activity score of 0.25 to 1”; “Poor metabolizers (PM) have an activity score of 0”; and, on the change, “in the first an AS of one results in an NM phenotype, and in the second an AS of one results in an IM phenotype”.
- Goetz MP, et al. Clinical Pharmacogenetics Implementation Consortium (CPIC) Guideline for CYP2D6 and Tamoxifen Therapy. Clin Pharmacol Ther. 2018. Tamoxifen as the third prodrug: alternative hormonal therapy recommended for poor and intermediate metabolisers.
- US Food and Drug Administration. Drug Development and Drug Interactions: Table of Substrates, Inhibitors and Inducers. Bupropion, fluoxetine, paroxetine, quinidine and terbinafine are listed as strong CYP2D6 inhibitors — the drugs that phenoconvert a genotypic normal metaboliser into a phenotypic poor one, which the activity score cannot see.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.