CRP is one of the best predictors of cardiovascular events in the blood, and the genetics is the reason we know it is not a cause. Variants that raise CRP for life do not raise heart disease — which is exactly the test a marker has to fail to be demoted from a cause.
Solid lines are connections this site curates. Dashed lines mean the two ends share a research paper — worth knowing, and not a claim that one explains the other.
C-reactive protein is made by the liver when the body is inflamed. Measured in blood, it predicts cardiovascular events well enough that it is used clinically to refine risk. For years that raised an obvious question: if lowering CRP lowered heart attacks, it would be a drug target worth billions.
You cannot randomise people to a lifetime of high CRP. But genetics did it for us. Some people carry variants that raise their CRP slightly from birth, and they carry them at random with respect to everything else — diet, income, smoking.
So the question becomes answerable. Do the people who inherited higher CRP have more heart disease?
They do not. Repeated analyses using CRP-raising variants as a natural experiment find no matching increase in cardiovascular events. CRP goes up with inflammation, inflammation goes with heart disease, and CRP itself is along for the ride.
This is one of the cleanest examples in human genetics of a marker being separated from a cause, and the method is now standard: if a variant that moves a marker for a lifetime does not move the outcome, the marker is unlikely to be causing it.
The same reasoning has since demoted several other promising targets and confirmed others — LDL cholesterol passes this test, which is part of why lowering it works.
All three come from a meta-analysis of more than 80,000 people — 66,185 in the discovery set across 15 population studies, with replication in 16,540 more. It found 18 genome-wide significant loci, with replication evidence for 8.
The strongest signals for CRP are in CRP itself and in IL6R, the receptor for the cytokine that induces it — neither is in our catalogue yet.
A high CRP measured in blood is worth attention: it means something is inflamed, and the something matters more than the number. Common causes are infection, injury, obesity, smoking and autoimmune disease.
What a high CRP does not tell you is why. And what these variants do not tell you is your CRP — they shift a population average, and a blood test measures the thing directly for a few pounds.
Positions joined since this page was written
What this is The text above discusses the variants this page was written around. Since then the catalogue has joined 587 more positions to it, by shared trait or shared paper. They are listed here by the paper each came from; the text does not describe them, and each variant page carries that study's own record.
Koskeridis F et al. 2022, Nature communications rs11065026 (near RAB35-AS1), rs74068089 (SOX5), rs12760385 (near DUSP23), rs145663198 (LINC02819), rs6989383 (NSMCE2), rs11556510 (TOMM40), rs2479464 (DOCK7), rs9675376 (near MC4R), rs1037144 (near OR10J6P), rs17807945 (near LINC00208), rs11643744 (FTO), rs6582586 (near RPL13AP21) and 160 more — PMID:36376304
Han X et al. 2020, European journal of epidemiology rs34415150 (near HLA-DRB1), rs288183 (FBXL17), rs7008413 (PPP1R3B-DT), rs61542988 (near SNHG26), rs144398877 (near OR10J9P), rs72743115 (near WDR72), rs12448756 (near SOD1P2), rs663015 (ARAP1), rs117368206 (near RPL11P5), rs846879 (CEACAM22P), rs7958316 (P2RX7), rs787488 (near LINC01702) and 125 more — PMID:31900758
Sinnott-Armstrong N et al. 2021, Nature genetics rs77832441 (CRP), rs10494326 (near CRPP1), rs1169282 (HNF1A), rs28360457 (P2RX7), rs1169721 (near P2RX4), rs1386821 (IL6R), rs17459069 (near CRPP1), rs12819210 (OASL), rs17434647 (P2RX7), rs7302482 (near CABP1-DT), rs115585839 (near DUSP23), rs11611673 (C12orf43) and 94 more — PMID:33462484
Loya H et al. 2025, Nature genetics rs180689986 (near CRP), rs28360463 (near P2RX4), rs140853922 (IFI16), rs548069293 (TOMM40), rs846881 (CEACAM22P), rs139996812 (near DUSP23), rs138131618 (near CRPP1), rs191485686 (near CRPP1), rs56261258 (CEACAM16-AS1), rs148811990 (near CRPP1), rs145670116 (near CRPP1), rs11612787 (C12orf43) and 54 more — PMID:39789286
Said S et al. 2022, Nature communications rs12929503 (near NFAT5), rs10106298 (near ADI1P2), rs34761529 (near PPIAP34), rs12516176 (TNIP1), rs6443429 (TBL1XR1), rs3808348 (CYP2W1), rs10760691 (NAMA), rs1801272 (CYP2A6), rs197419 (DDX20), rs4656241 (near OR10J6P), rs75229905 (BAZ1B), rs11047224 (SOX5) and 31 more — PMID:35459240
Sakaue S et al. 2021, Nature genetics rs2836881 (near LINC02940), rs3935032 (MIB2), rs185738008 (TAF4), rs12300845 (SOX5), rs1546721 (CCDC162P), rs1933737 (FRK), rs1644005 (BHMT), rs371271858 (near BAAT), rs12540285 (near TRPC6P3), rs4640929 (near RN7SL554P), rs5906271 (SLC9A7), rs11738559 (near TNIP1) and 9 more — PMID:34594039
Oh K et al. 2025, Scientific reports rs78821961 (STEAP1B), rs3093068 (near CRP), rs13306436 (IL6), rs12832102 (near ACADS), rs138233898 (LINC02819), rs384653 (NECTIN2), rs148665049 (LNCOB1), rs2239686 (ACADS), rs117103761 (SHE), rs74987345 (near MTCYBP42), rs75276031 (LEPROT), rs208296 (P2RX7) and 5 more — PMID:39747571
Ligthart S et al. 2018, American journal of human genetics rs12749227 (CRP), rs72698571 (CRP), rs2246469 (CRP), rs149520992 (CRP), rs12083537 (IL6R), rs67129560 (LEPR), rs4752829 (SPI1), rs4845623 (IL6R), rs9271608 (HLA-DQA1), rs56288844 (FCER1A), rs6987444 (LINC00536), rs2852151 (PTPN2) and 2 more — PMID:30388399
Jiang MZ et al. 2024, Human molecular genetics rs6588153 (LEPR), rs370370301 (near CRP), rs553202904 (near CRP), rs181704186 (near DUSP23) — PMID:38747556
Jacobs BM et al. 2024, Nature communications rs77289344 (near CRP), rs138594975 (near OR10J6P) — PMID:39414775
Sherman BT et al. 2021, AIDS (London, England) rs6667499 (CRP) — PMID:33095540
Kocarnik JM et al. 2018, Human molecular genetics rs16842568 (CRP) — PMID:29878111
Cheng S et al. 2023, Cell discovery rs12081480 (near CRP) — PMID:37479695
Sun BB et al. 2018, Nature rs2808628 (near CRPP1) — PMID:29875488
Jeon S et al. 2024, GigaScience rs11588887 (near CRP) — PMID:38626723
The source. Dehghan et al. (Circulation 2011) meta-analysed genome-wide association data for CRP in 66,185 participants from 15 population-based studies, with replication sought in 16,540 individuals from 10 independent studies — more than 80,000 subjects in total. Eighteen genome-wide significant loci were identified, with replication evidence provided for 8 and confirmation of 7 previously reported loci.
The variants recorded here. rs1183910 (HNF1A), rs6734238 (near the IL1F10/interleukin-1 cluster) and rs13233571 (BCL7B). None is among the strongest CRP signals, which lie in CRP itself and IL6R; those are not currently in this catalogue.
Marker versus cause. Circulating CRP is robustly associated with incident cardiovascular disease in observational cohorts. Mendelian randomisation analyses using CRP-raising alleles as instruments have not found a corresponding increase in coronary risk, which is the basis for the current view that CRP is a marker of inflammatory burden rather than a causal mediator. The contrast with LDL cholesterol, where the same design supports causality, is the reason the method is now routine.
Clinical boundary. High-sensitivity CRP is used in some cardiovascular risk frameworks as an adjunct, measured directly from blood. No guideline uses genotype at these or any other positions to estimate CRP or to modify risk assessment, and nothing on this page is a reason to request a CRP test.
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about C-Reactive Protein Levels comes down to these specific, well-studied positions — not a diagnosis. 636 positions are linked to this page; the ones this page's own text discusses are shown first.
CRP · rs12755606
See detailed info →CRP · rs6667499
See detailed info →The studies behind these variants recruited participants from different ancestries — a result found in one population doesn't always transfer to another. Based on 145 of 636 linked studies with a resolved discovery ancestry.
Free to quote and reuse under CC BY 4.0. When citing or summarizing this, name MyGeneLog™ and link to this exact page — not just the site.
C-Reactive Protein Levels. MyGeneLog™. https://www.mygenelog.com/conditions/c-reactive-protein-levels
Because prediction is not causation, and genetics can tell them apart. People who inherit variants that raise their CRP for life do not have correspondingly more heart disease. CRP rises with inflammation; inflammation goes with heart disease; CRP is along for the ride.
No. A blood test measures CRP directly and cheaply, and a measurement beats a prediction every time. These positions shift a population average and say nothing about your level.
That something is inflamed — and what that something is matters far more than the number. Infection, injury, obesity, smoking and autoimmune disease are the common reasons. It is a question for a doctor, not for a genome.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog™. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.