Expressive
← Blog · · 6 min read
Gene deep-diveFTOEducation

FTO gene variants and weight, what the GWAS research actually says

The FTO gene was the first common genetic variant linked at genome-wide significance to obesity. Since the original Frayling et al. paper in 2007, it's been replicated across hundreds of cohorts. It's real. It's also massively over-interpreted in consumer wellness content. Understanding why both of those things are true tells you a lot about how to read GWAS findings in general.

This post walks through what the actual evidence says about FTO, what carrying a risk allele does and doesn't mean, and what the research supports doing about it (very little, specifically targeted at FTO, but useful general principles).

The variant: rs9939609

The most-studied FTO SNP is rs9939609. It's in the first intron of FTO, which sits on chromosome 16. The two alleles are A and T. Frequencies vary by population, about 42% A in European-ancestry populations, lower in East Asian populations, similar in some African populations. If you want to see the full catalog of FTO entries with clinical interpretation, ClinVar's FTO page and the NHGRI-EBI GWAS Catalog FTO record are the canonical references.

Risk-allele homozygotes (AA) weigh, on average, about 3 kg more than non-carriers (TT). Heterozygotes (AT) sit in the middle. That's the headline finding, replicated across studies with hundreds of thousands of participants. (If you want to skim the primary literature directly, the PubMed query for FTO + rs9939609 returns the main cohort and meta-analysis papers.)

This is the kind of finding the "FTO gene obesity" search history is built on, and it's also where most of the over-interpretation starts. The honest read on the rs9939609 FTO variant requires sitting with both facts at once: real association, small individual effect.

A 3 kg difference is meaningful at the population level. It's small at the individual level. It explains roughly 1% of variance in BMI. Other genetic variants contribute similar small amounts; in aggregate, common-variant genetics explain perhaps 20% of BMI variance in well-powered studies. The rest is environment, behavior, and (small but real) gene-by-environment interactions.

What the mechanism is

The original FTO finding was framed as "FTO is the obesity gene." That framing turned out to be partly wrong. The interesting wrinkle: the rs9939609 SNP is in an FTO intron, but the actual functional consequence appears to involve a different gene, IRX3, located further away on the chromosome.

Smemo et al. (2014) showed that the rs9939609 region contains a regulatory element that controls IRX3 expression in early adipocyte development, not FTO. Later work (Claussnitzer et al. 2015) developed the mechanistic story further: in carriers of the risk allele, the regulatory element more strongly activates IRX3 and IRX5, which biases preadipocyte development toward energy-storing (white) fat rather than energy-burning (beige) fat.

The mechanism is biologically real. It's also tiny in effect. We're talking about a modest shift in the fat-cell developmental program over many years.

What the GWAS evidence supports

A few claims about FTO are well-supported by the literature:

Modest BMI effect. As above. ~3 kg average difference between AA and TT genotypes, replicated across many large cohorts.

Effect varies with physical activity. Multiple meta-analyses (Kilpeläinen et al. 2011, Ahmad et al. 2013) show that the BMI effect of FTO risk alleles is reduced, sometimes substantially, in physically active individuals. The interaction isn't quite "exercise cancels FTO" but is closer to "exercise reduces the effect by half or more in active populations." This is one of the better-replicated gene-by-environment interactions in obesity genetics.

Modest effect on appetite signaling. Several studies show FTO risk-allele carriers report slightly different satiety responses and have small differences in eating behavior measures. The effect sizes are small and inconsistent across studies.

Modest effect on overall mortality. Some large cohort studies show small mortality effects in AA carriers, mediated partly through cardiometabolic disease pathways. The effect is small and not consistently found across populations.

What the evidence does NOT support

A much longer list of claims has been attached to FTO without solid support:

"FTO makes you fat." No. The variant modestly tilts your probability. Most AA carriers are not obese. Many TT carriers are. The variant is one of many contributors and is dominated in effect by behavior and environment.

"FTO-positive people need a specific diet." No solid evidence supports differential dietary response to common macronutrient changes by FTO genotype. Some studies suggest small effects (e.g., FTO carriers may benefit slightly more from higher-protein diets in some analyses) but the effects are inconsistent and not large enough to drive personalized recommendations.

"FTO-positive people can't lose weight." Strongly false. RCTs show that AA carriers lose weight in response to caloric restriction and exercise at rates similar to non-carriers. The starting weight is shifted; the responsiveness to intervention is not.

"FTO is why I'm overweight." For any one person, almost certainly not. The variant's contribution is small, and the dominant factors for individual body weight are dietary intake, physical activity, sleep, and broader metabolic context.

"Anti-FTO supplements." No such thing has solid evidence. The supplement industry has marketed various compounds as "addressing the FTO pathway" with no RCT support.

What the evidence DOES support, at a general level

This is where carrying FTO risk variants intersects with the broader literature on weight management and metabolic health:

The summary: there's no FTO-specific intervention. There's also no reason for an FTO risk-allele carrier to do less of the general evidence-supported weight management interventions; if anything, the small interaction effects make the standard advice slightly more important.

How to think about FTO if you carry the risk allele

A few practical takeaways:

  1. It's not destiny. The variant shifts your starting point modestly. It doesn't determine your outcome.
  2. The general interventions work. Physical activity, sleep, eating patterns, all evidence-anchored, all useful, all not FTO-specific.
  3. Skip the FTO-marketed products. Supplements or diet plans claiming to "address" FTO are exploiting a real association with a fabricated intervention.
  4. Pay attention to the broader metabolic picture. Lipid panel, A1C, blood pressure, body composition. These are the actually-actionable signals.

How Expressive handles FTO

If you upload your raw genetic file to Expressive, the FTO rs9939609 status surfaces with its actual evidence quality: a modest but well-replicated BMI effect with a confirmed exercise-interaction effect. We don't prescribe, we describe. You see the magnitude, you see the studies, you see the gap between what the literature supports and what the supplement industry claims.

This is what evidence-based genomics looks like in practice. We don't recommend FTO-specific interventions because the literature doesn't support them. We note the variant, link to the underlying studies, and flag the gene-by-environment interaction so you can see why the strongest evidence-based response is "stay active" rather than "buy a supplement." No wellness washing. Your genome stays yours: privacy-first DNA analysis with the raw file and the interpretation under your control, not sold to a downstream broker.

You can also look at the rs9939609 variant page directly. The honest read on FTO is one of the better case studies in genetics and longevity content: small effect, real mechanism, and a wide gap between the literature and the marketing. Know your magnitude. That's the whole job.

Want updates when we ship new variant pages or a research deep-dive? Read the latest issue or get notified about early access.