CYP2B6: the enzyme behind efavirenz side effects
CYP2B6 is a smaller enzyme than its cytochrome P450 relatives in almost every sense, but for one HIV drug it is the difference between a manageable regimen and one that people stop taking because of the dreams.
Also known as: cytochrome P450 2B6.
What CYP2B6 does
The enzyme clears a short list of drugs from the body, and the reduced-function variants that matter here mostly lower how much working enzyme gets made. When clearance drops, blood levels from a standard dose rise, and for drugs that act on the central nervous system that shows up as side effects rather than as a laboratory number. The frequency of the main reduced-function allele varies substantially between populations, which is why the same fixed dose has behaved differently in trials run in different places.
The metabolizer groups
CYP2B6 is graded by diplotype into poor, intermediate, normal, rapid and ultrarapid metabolizer, plus Indeterminate where the pair cannot be graded. Most of the clinical weight sits at the slow end. The rapid and ultrarapid categories exist and are reported, but the published dosing guidance below only changes at the intermediate and poor end, which is a useful reminder that a category with a dramatic name does not always come with an action attached.
Medications where it matters
efavirenz (Sustiva, part of Atripla)
Efavirenz is taken once a day, usually at bedtime, and its characteristic side effects are nervous-system ones: vivid dreams, dizziness, trouble concentrating. Those track drug concentration, and concentration tracks this gene. The guidance below is about a lower starting dose for slow metabolizers rather than about avoiding the drug, which is a meaningful difference in settings where efavirenz-based regimens are the mainstay.
| If your result is | Published CPIC guidance |
|---|---|
| Intermediate Metabolizer | Per CPIC (A): "Consider initiating efavirenz with decreased dose of 400 mg/day" |
| Normal Metabolizer | Per CPIC (A): "Initiate efavirenz with standard dosing (600 mg/day)" |
| Poor Metabolizer | Per CPIC (A): "Consider initiating efavirenz with decreased dose of 400 or 200 mg/day" |
| Rapid Metabolizer | Per CPIC (A): "Initiate efavirenz with standard dosing (600 mg/day)" |
| Ultrarapid Metabolizer | Per CPIC (A): "Initiate efavirenz with standard dosing (600 mg/day)" |
sertraline (Zoloft)
Sertraline is cleared by several enzymes, so CYP2B6 is one contributor rather than the whole story, and CPIC grades this recommendation at a lower confidence level than the efavirenz one. What it changes is titration speed and maintenance dose at the slow end, not whether the drug is suitable.
| If your result is | Published CPIC guidance |
|---|---|
| Indeterminate | Per CPIC (B): "Initiate therapy with recommended starting dose." |
| Intermediate Metabolizer | Per CPIC (B): "Initiate therapy with recommended starting dose. Consider a slower titration schedule and lower maintenance dose." |
| Normal Metabolizer | Per CPIC (B): "Initiate therapy with recommended starting dose." |
| Poor Metabolizer | Per CPIC (B): "Consider a lower starting dose, slower titration schedule and 25% reduction of standard maintenance dose as compared to CYP2B6 normal metabolizers or select a clinically appropriate alternative antidepressant not predominantly metabolized by CYP2B6." |
| Rapid Metabolizer | Per CPIC (B): "Initiate therapy with recommended starting dose." |
| Ultrarapid Metabolizer | Per CPIC (B): "Initiate therapy with recommended starting dose." |
Is this in your raw DNA file?
CYP2B6 is not part of the standard chip or variant-file report. It is a Deep Read gene, read from aligned reads in a BAM or CRAM file. The star alleles here are built from combinations of variants across the gene rather than from single marker positions, so resolving a diplotype needs sequence. Where the reads do not settle it, the result is Indeterminate rather than an assumed normal.
A genotyping chip reads only a few hundred thousand pre-selected positions, about 0.02% of your genome, chosen for common variation. It does not sequence the rest, so it cannot find the rare or novel pathogenic variants the clinical and carrier modules look for: a “no finding” from chip data means “this chip never looked”, far more so than with whole-genome sequencing. Chip data suits common-variant traits, pharmacogenomic tag SNPs, and haplogroup ancestry, not clinical or carrier screening.
Common questions
Is CYP2B6 in my 23andMe or AncestryDNA raw data?
Some of the individual variants may be, but the star alleles that CPIC grades are combinations, and a chip cannot resolve which variants sit on which chromosome. Aimosti reads this gene from aligned reads in the Deep Read panel.
I am on efavirenz and the dreams are bad. Should I change my dose?
Not on your own, and not on the basis of this page. Antiretroviral dosing is held tightly for good reasons, and reducing a dose without the prescriber involved risks the thing the treatment exists to prevent. This result is something to bring to that conversation.
Why does the sertraline guidance look weaker than the efavirenz one?
Because it is, and CPIC labels it that way. Efavirenz guidance carries CPIC's strongest evidence classification; the sertraline recommendation carries a lower one. The tables above reproduce what CPIC published for each, including that difference.