ABCG2: the second transporter in the rosuvastatin story
ABCG2 is a small page for a narrow question: it changes how much rosuvastatin ends up in your blood. That is the whole of its guideline scope, and pretending otherwise would be padding.
Also known as: BCRP, breast cancer resistance protein, ATP binding cassette subfamily G member 2.
- Gene
- ABCG2
- Medications on this page
- rosuvastatin
What ABCG2 does
ABCG2 makes a pump, usually called BCRP, that sits in cell membranes in the gut and the liver and pushes certain drugs back out. For rosuvastatin the practical effect is on absorption and clearance: a weaker pump means more of the drug gets in and less of it leaves, so the same tablet produces higher blood levels. The gene has a second life outside pharmacology as a determinant of uric acid handling, which is why you may see it mentioned in the context of gout, but that is not what CPIC grades it for and not what Aimosti reports here.
The transporter function groups
The grading is short because the evidence is. CPIC recognises normal function, decreased function and poor function for ABCG2, driven mainly by one common variant, rs2231142, also written Q141K. One copy gives decreased function, two give poor function. There is no metabolizer scale here, because ABCG2 is a pump rather than an enzyme.
Medications where it matters
rosuvastatin (Crestor)
Rosuvastatin is the only drug in CPIC's ABCG2 guidance, and the recommendation is written for the pair of transporters rather than for this gene alone. The rows below assume a typical SLCO1B1 result. If your SLCO1B1 result is not typical, CPIC's advice for the pair changes. From a chip or a plain VCF the report's ABCG2 card quotes the row for the SLCO1B1 result your file gives; the gVCF and Deep Read cards quote the row for a typical SLCO1B1 and say so. The rosuvastatin page sets out both tables.
| If your result is | Published CPIC guidance |
|---|---|
| Decreased Function | Per CPIC (A): "Prescribe desired starting dose and adjust doses of rosuvastatin based on disease-specific and specific population guidelines." |
| Normal Function | Per CPIC (A): "Prescribe desired starting dose and adjust doses of rosuvastatin based on disease-specific and specific population guidelines." |
| Poor Function | Per CPIC (A): "Prescribe ≤20mg as a starting dose and adjust doses of rosuvastatin based on disease-specific and specific population guidelines. If dose >20mg needed for desired efficacy, consider an alternative statin or combination therapy (i.e., rosuvastatin plus non-statin guideline directed medical therapy) (PMID: 30423391)." |
Is this in your raw DNA file?
ABCG2 is part of the standard report from a chip export or a variant file as well as from aligned reads, because CPIC grades it on one position, rs2231142, and most consumer arrays type it. Some exports leave it out: we have seen MyHeritage, FamilyTreeDNA and Living DNA files with no row for it, and on those the card says the position was not on your array instead of calling the gene normal. A gVCF counts the position only where the file actually called it. A plain VCF cannot prove a missing position was covered, so a reference result from one carries that caveat. CPIC reads the two transporters together for rosuvastatin, so from a chip or a plain VCF the ABCG2 card quotes CPIC's row for the SLCO1B1 result read from the same file.
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.
Free · in your browser
Find out what your own file can read.
The free check opens your DNA file on your device and shows which format you have and what a report could read from it. The file is never uploaded.
Common questions
Is ABCG2 in my 23andMe or AncestryDNA raw data?
Usually. Every 23andMe (v3, v4 and v5) and AncestryDNA export we have tested carries rs2231142, and Aimosti reads ABCG2 from it in the standard chip report, with the rosuvastatin line matched to the SLCO1B1 result from the same file. If your export has no row for that position, the result is reported as unresolved rather than assumed to be normal function.
I have seen ABCG2 linked to gout. Is that in the report?
No. The uric-acid association is real and well studied, but it is not pharmacogenomic guidance and CPIC does not grade the gene for it. This page and the report stay inside what CPIC publishes, which for ABCG2 means rosuvastatin.
Why is there only one drug here when the statin guideline covers seven?
Because ABCG2 only carries weight for rosuvastatin. The other six statins in the same CPIC guideline are graded on SLCO1B1, sometimes with CYP2C9, and they appear on those pages instead.