Atomoxetine
Atomoxetine (Strattera) is a non-stimulant ADHD medicine for adults and children 6 and older. CYP2D6 does most of the work of clearing it, so poor metabolizers have about 10 times the exposure of other people and a half-life of about 22 hours instead of about 5. The FDA label recommends a slower, 4-week titration step for poor metabolizers, and CPIC gives dosing advice by activity score and age, including a blood-level check when the response is poor.
| Brand names | Strattera; generic atomoxetine capsules |
|---|---|
| Used for | Attention-deficit/hyperactivity disorder (ADHD) in adults and children 6 and older |
| Genes | CYP2D6 |
| Guidance | CPIC guideline (2019) [1]; FDA label (sections 2.5, 8.7 and 12.5) [2] |
| Gene2Rx coverage | Consumer DNA tests and whole-genome sequencing. |
How atomoxetine works
Atomoxetine is thought to work by selectively blocking the norepinephrine transporter on nerve endings [2]. Unlike stimulants it acts slowly: CPIC notes that full benefit typically takes 2 to 4 weeks [1]. The label's usual dosing starts at 0.5 mg/kg/day for children under 70 kg (target 1.2 mg/kg/day) and 40 mg/day for heavier children and adults (target 80 mg/day, maximum 100 mg/day), raised after at least 3 days [2].
Atomoxetine is the active drug; it is not a prodrug. CYP2D6 converts it to 4-hydroxyatomoxetine, which is as potent but is quickly joined to glucuronic acid and excreted, so very little of it circulates. CYP2C19 makes a second metabolite with about 20-fold less activity [2]. When CYP2D6 is missing, the drug stays in the body far longer:
| Measure | Other CYP2D6 types | Poor metabolizers |
|---|---|---|
| Half-life | 5.2 hours | 21.6 hours |
| Oral clearance | 0.35 L/hr/kg | 0.03 L/hr/kg |
| Absolute bioavailability | 63% | 94% |
| Total exposure (AUC) | Reference | About 10 times higher |
| Peak level at steady state | Reference | About 5 times higher |
These figures come from the label's pharmacokinetic studies [2].
How CYP2D6 affects atomoxetine
CPIC turns a person's two CYP2D6 alleles into an activity score. Under its current scale, 0 is a poor metabolizer, 0.25 to 1.0 an intermediate metabolizer, 1.25 to 2.25 a normal metabolizer, and above 2.25 an ultrarapid metabolizer, a group that requires duplicated genes [3]. The 2019 guideline paper used an older scale, which is why it describes a score of 1.0 as "normal or intermediate" [1]. The codeine page explains the scoring in more detail.
Exposure rises steadily as CYP2D6 activity falls. CPIC describes a 30-fold range in exposure among 23 children given the same starting dose, and 5-fold higher peak levels in people with two copies of the reduced-function *10 allele than in people with at least one normal allele [1]. Response follows exposure. In the label's trials, poor metabolizers had larger heart rate increases (9.4 against 5 beats per minute in children; 11 against 7.5 in adults), and in children 11% stopped because of side effects, against 6% of other metabolizer types [2]. Some side effects were clearly more common in poor metabolizers:
| Side effect | Poor metabolizers | Other CYP2D6 types |
|---|---|---|
| Insomnia (children) | 11% | 6% |
| Tremor (children) | 5% | 1% |
| Dry mouth (adults) | 35% | 17% |
| Erectile dysfunction (adults) | 21% | 9% |
| Urinary retention (adults) | 6% | 1% |
The same higher exposure has an upside. CPIC reports that poor metabolizers who tolerate atomoxetine show greater symptom improvement, and that non-poor metabolizers more often stop it because it does not work. That evidence compares poor metabolizers with everyone else; CPIC found no evidence yet on response in the other groups separately [1]. The label puts poor metabolizers at about 7% of White, 2% of Asian and 2% of Black or African American people [2].
What the guidelines recommend
The Clinical Pharmacogenetics Implementation Consortium (CPIC) published its atomoxetine guideline in 2019, with separate recommendations for children and adults [1]. CPIC rates the CYP2D6 and atomoxetine pair level A, its highest level, meaning genetic information should be used to change prescribing [4]. Its advice centers on an "exposure check": if the response is poor, a blood level taken at a set time after a dose shows whether the person is reaching the 200 to 1000 ng/mL range generally considered therapeutic [1]. The table follows CPIC's current tables, which use the updated activity-score scale [3].
| Result | What it means | Guideline recommendation |
|---|---|---|
| Ultrarapid Metabolizer | Very limited data; unlikely to reach adequate levels at standard doses. | Same as normal metabolizers (moderate). |
| Normal Metabolizer | Less likely to respond than poor metabolizers, and more likely to stop for lack of effect. | Children: start 0.5 mg/kg/day, raise to 1.2 mg/kg/day after 3 days. Adults: start 40 mg/day, raise to 80 mg/day after 3 days, and consider 100 mg/day if after 2 weeks there is no response and no side effects. If there is still no response after 2 weeks, consider a peak level 1 to 2 hours after a dose; if it is below 200 ng/mL, consider a proportional increase toward 400 ng/mL. Adults may need more than 100 mg/day (moderate). |
| Intermediate Metabolizer, activity score 1.0 | Possibly higher levels than normal metabolizers, of questionable clinical significance. | Same as normal metabolizers (moderate). |
| Intermediate Metabolizer, activity score 0.25 to 0.75 | Slower metabolism and higher levels than normal metabolizers. | Children: start 0.5 mg/kg/day. Adults: start 40 mg/day and raise to 80 mg/day only if after 2 weeks there is no response and no side effects. If the response is inadequate, consider a level 2 to 4 hours after a dose and a proportional increase toward 400 ng/mL if it is below 200 ng/mL. Reduce the dose if side effects are unacceptable (moderate). |
| Poor Metabolizer | Much higher levels: more side effects, but greater improvement in people who tolerate it and lower final doses. | As for activity scores 0.25 to 0.75, except that in children the level is checked 4 hours after a dose (children: strong; adults: moderate). FDA label: allow 4 weeks between dose increases. |
| Indeterminate or not available | An allele was found whose function is uncertain or unknown, so no phenotype can be given. | CPIC makes no recommendation. The report defers to the usual label dosing. |
Gene2Rx reports show CPIC's adult recommendations, chosen by activity score, for every age; CPIC's weight-based starting doses for children are in the table above. For an activity score of 1.0 the report shows the slower schedule that CPIC uses for scores of 0.25 to 0.75, while CPIC's current tables give the normal-metabolizer schedule [3]. CPIC also notes that doses above 120 mg have not been studied extensively [1].
What the FDA label says
The current FDA label tells prescribers to consider genetic testing to determine CYP2D6 metabolizer status. For poor metabolizers, and for anyone taking a strong CYP2D6 inhibitor, it recommends the same starting, target and maximum doses as everyone else but a 4-week titration interval, used if symptoms have not improved and the starting dose is well tolerated. Other metabolizer types follow the usual minimum of 3 days [2]. The FDA's Table of Pharmacogenetic Associations lists the poor metabolizer finding in its first section, where the data support management recommendations [5].
Want to know what your genetics say about how you'll respond to Atomoxetine?
A Gene2Rx report reads your own DNA to show how it may affect your response to Atomoxetine and your other medications.
Using your result
Your report lists your CYP2D6 diplotype and activity score. A poor metabolizer result is the one the label acts on directly, so it is worth raising before atomoxetine starts or before a dose increase. For other results, CPIC's exposure check is the practical tool if the medicine does not seem to work.
Other drugs can make a person behave like a poor metabolizer. CPIC names bupropion, fluoxetine and paroxetine as strong CYP2D6 inhibitors that raise atomoxetine levels, and notes that fluoxetine's effect can last 2 to 3 months after it is stopped [1]. The label applies its 4-week titration interval to anyone taking a strong inhibitor [2].
The label also cuts the starting and target doses to 50% in moderate liver impairment and 25% in severe impairment, and advises checking heart rate and blood pressure at baseline, after dose increases and during treatment. It carries a boxed warning about suicidal thoughts in children and teenagers [2].
Do not start, stop or change atomoxetine or another medicine because of a genetic result without talking to your prescriber.
What this result cannot tell you
Consumer arrays (23andMe, AncestryDNA, MyHeritage) are not clinical-grade tests, and their raw data is labeled not for medical use. They read fixed positions, and Gene2Rx uses imputation to infer some variants that were not measured directly. A normal result means none of the assessed variants was found among the positions your data covers; it cannot rule out a variant at a position the data does not cover. CPIC warns that rare variants missed by a test are usually reported as the default *1 allele [1].
Genotype explains only part of the response. In an analysis of 618 patients from six trials, CPIC notes, 47% responded well, 13% minimally and 40% not at all, a split that poor metabolizer status alone cannot account for [1]. Variants in SLC6A2, the gene for the norepinephrine transporter, have been studied but are not used in guidance. CPIC notes that a blood level also captures non-genetic influences on CYP2D6 activity, such as interacting medicines [1].
Common questions
I am a poor metabolizer. Can I still take atomoxetine?
Yes. Neither CPIC nor the label advises avoiding it; both call for the usual starting dose with slower increases and closer attention to side effects (see What the guidelines recommend).
Why might a normal metabolizer need a blood test?
Normal metabolizers clear atomoxetine quickly, and some do not reach the target levels at standard doses. CPIC suggests checking a level if there is no response after 2 weeks.
Sources and further reading
- Brown JT et al. Clinical Pharmacogenetics Implementation Consortium guideline for cytochrome P450 (CYP)2D6 genotype and atomoxetine therapy. Clin Pharmacol Ther (2019). Evidence review, exposure checks and recommendations for children and adults.
- Strattera (atomoxetine) capsules: prescribing information (Eli Lilly). DailyMed. Dosing, CYP2D6 poor metabolizer data, side effects, warnings.
- CPIC. Current atomoxetine recommendations by CYP2D6 activity score (CPIC database record). The recommendation rows in the table above.
- CPIC. Gene-drug pairs and CPIC levels (spreadsheet). Level A for CYP2D6 and atomoxetine.
- FDA. Table of Pharmacogenetic Associations. Atomoxetine entry for poor metabolizers.
Considering GeneSight for this medication?
Gene2Rx uses the same CPIC and FDA guidelines as GeneSight, covers more medications, costs $35 with the Psychiatric Report or $50 with the Full Report instead of several hundred, and works with the 23andMe or AncestryDNA data you may already have. See the full side-by-side comparison before ordering. Already have a GeneSight report? Use your GeneSight results.
Read the comparisonGenes that affect Atomoxetine
See the full list of drugs affected by each gene:
Brand names containing Atomoxetine
See how your genetics affect each product Atomoxetine shows up in:
Related guides
Learn more about how genetics may affect your response to Atomoxetine and related medications:
- 23andMe Pharmacogenetics: How to Get a Drug Response Report From Your Existing Data
- Adderall Not Working? Common Reasons, and What CYP2D6 Can and Cannot Explain
- ADHD Medication Not Working? What Your DNA Might Have to Do With It
- AncestryDNA for Drug Testing: Get Pharmacogenetics From Your Ancestry Data
- Atomoxetine Not Working? Why Strattera Might Not Be Helping Your ADHD
- CYP2D6 Structural Variants: Deletions, Duplications and Gene Fusions Explained
- Genera Pharmacogenetics: Turn Your Genera Raw Data Into a Drug Response Report
- Looking for a GeneSight Alternative? Here's the Short Answer
- How Much Does GeneSight Cost? Full Breakdown Plus Alternatives
- GeneSight vs Genomind vs Gene2Rx: Three-Way Pharmacogenetic Test Comparison
- Can I Get GeneSight Without a Doctor? Direct-to-Consumer Alternatives
- Looking for a Genomind Alternative? Here's What to Know
- MyHeritage Pharmacogenetics: Use Your MyHeritage DNA for a Drug Response Report
- Nebula Genomics Pharmacogenetics: How to Get a Drug Response Report From Your WGS Data
- How Much Does a Pharmacogenetic Test Cost? 2026 Prices and Cheaper Alternatives
- Pharmacogenetic Testing Without a Doctor's Order: Your Options
- SelfDecode Alternative: Comparing Genetic Health Report Options
- Tamoxifen Not Working? Why Genetics Matters for Breast Cancer Treatment
- TellMeGen Pharmacogenetics: Turn Your TellMeGen Raw Data Into a Drug Response Report
- How to Read Your GeneSight Report: Green, Yellow, and Red Explained
- Venlafaxine Side Effects: When Effexor's Side Effects Are Too Much
Interested in learning more about how your genetics may affect your response to medication? Get started with Gene2Rx today.
I'm Interested