MTHFR Variants: What the Evidence Actually Supports

Key Takeaways

Somewhere in the last decade, a single-letter variant in one folate enzyme became the most misunderstood result in consumer genetics. People are told their "MTHFR mutation" means they cannot use folic acid, cannot detoxify, cannot methylate, and that a long list of symptoms follows. Supplement ranges exist for it. Some clinics test for it routinely.

The research tells a smaller and more useful story. The variant is common. It reduces an enzyme's activity. Whether that matters depends almost entirely on how much folate — and, it turns out, riboflavin — you take in. This review goes through what the variant does, what the disease association data show, why the main genetics body advises against testing for it, and the two interventions that have actual trial evidence behind them.

If you want the practical version, the Methylation Support Protocol turns this into a test-first plan. If you came here because of a homocysteine result, start with the homocysteine guide.

What the Variant Actually Does

MTHFR sits at a junction in one-carbon metabolism. It converts one form of folate into 5-methyltetrahydrofolate, which hands a methyl group to homocysteine and turns it back into methionine. Vitamin B12 is the cofactor for that hand-off; riboflavin, as FAD, is the cofactor for MTHFR itself. If any of the three runs short, homocysteine accumulates.

The C677T variant was described in 1995. It was found on about 38% of the chromosomes tested, and it produces a thermolabile enzyme — one that loses activity under mild heat in the lab — with reduced activity in both carriers and homozygotes. TT homozygotes had higher plasma homocysteine. A second common variant, A1298C, followed in 1998: about 10% of Canadians tested were homozygous for it, with roughly 60% of normal enzyme activity, and people carrying one copy of each variant had 50–60%. A1298C on its own has a much weaker effect on homocysteine than 677TT.

Two figures do most of the work here. About one person in ten of European or North American descent is 677TT — 12% in a 6,076-person Irish cohort, around 10% in North American samples. And the original papers were explicit: TT produces mild hyperhomocysteinaemia "when their folate status is low." The variant is a vulnerability that adequate folate largely cancels.

The Disease Association Data — and Why Geography Matters

If the variant caused heart disease, its effect should look the same everywhere. It does not.

A 2012 analysis then tested whether even the overall coronary signal was real. It pooled 19 unpublished genetic datasets — 48,175 cases — and found TT vs CC OR 1.02 (0.98–1.07), and 1.01 in low-folate populations. The 86 published studies had given 1.15; the difference was statistically significant. Much of the original coronary association was publication bias.

The pattern is the same in every row. Where diets are low in folate — or where flour is not fortified — the TT genotype carries a measurable extra risk. Where folate intake is high, the association fades to nothing. The 2011 stroke meta-analysis showed the mechanism directly: the homocysteine gap between TT and CC was 3.12 µmol/L in Asia and 0.13 µmol/L in fortified regions.

This is also why the variant was once treated as a natural experiment for homocysteine itself. A 2002 Mendelian-randomisation analysis predicted that lowering homocysteine with folic acid would cut ischaemic heart disease by 16%. The trials that followed did not deliver that. The same 2011 meta-analysis pooled 13 randomised homocysteine-lowering trials and found a stroke relative risk of 0.94 (0.85–1.04) — not significant. The full trial story is in the homocysteine guide.

Why the Genetics Body Says Don't Test

In 2013 the American College of Medical Genetics and Genomics published a practice guideline with a title that is its own summary: lack of evidence for MTHFR polymorphism testing. It concluded that meta-analyses had disproven an association between hyperhomocysteinaemia and coronary heart disease, and between MTHFR status and venous thromboembolism, and that MTHFR testing "should not be ordered as a part of a routine evaluation for thrombophilia." A 2020 addendum reaffirmed the position.

The logic is straightforward once the geography is clear:

• A positive result is common. About one in ten people of European descent is TT, and more than half carry at least one T.
• It changes nothing that a cheaper test would not. If the variant matters, it matters through homocysteine — and homocysteine can be measured directly, in a blood test that also captures B12, renal function and everything else that raises it.
• It invites harm. People with a TT result are routinely told to avoid folic acid, including before pregnancy — the one setting where folic acid has the strongest evidence in nutrition.

If you already have the result — often from a consumer DNA kit — the useful next step is a fasting homocysteine, serum B12 (ideally with methylmalonic acid) and folate. That tells you whether the variant is doing anything in your body at your current intake.

Folic Acid vs Methylfolate: What the Trials Show

The central claim of the MTHFR supplement market is that TT carriers cannot process folic acid and must take L-methylfolate instead. Four randomised trials test that claim directly.

Read together, they say this: folic acid works in TT carriers — in the trial that stratified by genotype, TT had the largest fall. Methylfolate is equal or modestly better on blood markers, and it reaches higher plasma levels faster in a pharmacokinetic study that included both TT and CC women. In that study, plasma folate was essentially all 5-MTHF regardless of which form was swallowed: the body converts folic acid fine, even with a slower enzyme.

What methylfolate does avoid is unmetabolised folic acid in blood. That appears once a single dose exceeds roughly 200–270 µg, and a 2015 national survey detected it in over 95% of Americans, supplement users or not. Whether it does any harm is unknown: an NIH workshop in 2020 concluded the evidence was inconclusive and did not justify changing public-health recommendations.

The one setting where the form is not a matter of preference is pregnancy. The 1991 Medical Research Council trial in 1,817 high-risk women cut neural tube defects by 72% (RR 0.28) with folic acid. In 2023 the US Preventive Services Task Force reaffirmed a grade-A recommendation for 0.4–0.8 mg of folic acid daily for anyone who could become pregnant. No trial has tested methylfolate against neural tube defects. A TT result is not a reason to switch away from the form that has the outcome data — and a methylfolate product taken alongside a standard prenatal is not harmful either.

Methylfolate is also sold at 7.5–15 mg as an add-on for depression. That is a pharmacological dose, the evidence is one null and one positive small trial, and neither selected patients by genotype. It is outside the scope of a methylation protocol.

Riboflavin: The Genotype-Specific Intervention

The most interesting MTHFR finding has nothing to do with folate. The thermolabile TT enzyme loses its FAD cofactor more easily — so riboflavin, the precursor of FAD, should stabilise it. A series of trials from one research group in Northern Ireland tested that.

• Homocysteine (2006): riboflavin 1.6 mg/day for 12 weeks lowered homocysteine by 22% in TT carriers (16.1 to 12.5 µmol/L), and by 40% in those with low riboflavin at baseline. No effect in CC or CT.
• Blood pressure in cardiovascular patients (2010): in 181 patients, the same dose over 16 weeks took TT blood pressure from 144/87 to 131/80 mmHg. No effect in other genotypes.
• Treated hypertension (2013): in 91 TT patients already on medication, systolic blood pressure fell a further 5.6 mmHg versus placebo (P = 0.033); diastolic did not change significantly.
• Population data (2020): in 6,076 Irish adults, TT raised hypertension risk (OR 1.42), and TT combined with low riboflavin status tripled it (OR 3.00) versus CC with normal riboflavin.

This is genuinely genotype-specific, cheap and physiological — 1.6 mg is roughly the adult daily requirement. It is also limited: sample sizes of 31–91 TT participants, one research group, authors with patent interests in riboflavin for hypertension, and no independent replication. The fair summary is a grade C with a strong mechanism: a reasonable thing for a TT carrier with high blood pressure or high homocysteine to try, alongside — never instead of — their antihypertensive. Profile and dosing in Riboflavin.

The Folate–B12 Trap

There is one real reason for caution with high folate, and it is not MTHFR. It is vitamin B12.

Folate can correct the anaemia of B12 deficiency while the neurological damage continues. In US national survey data, older adults with low B12 and high serum folate had three times the odds of anaemia and 2.6 times the odds of cognitive impairment; with normal B12, high folate was protective. A companion analysis found that in B12 deficiency, homocysteine and methylmalonic acid actually rose as folate rose. These are cross-sectional associations and a 2020 re-analysis found them inconsistent across cognitive tests — but the practical rule is clear: check B12 before taking folate in any form, and take them together. See Vitamin B12.

What to Do With an MTHFR Result

1. Do not treat it as a diagnosis. It is a common variant whose effect is largely neutralised by adequate folate.
2. Measure homocysteine, B12 (with MMA if borderline) and folate. These tell you whether anything needs fixing.
3. If homocysteine is normal, nothing needs to change.
4. If it is raised: B12 first if low, then folate — folic acid or methylfolate both work — then riboflavin 1.6 mg if you are TT. Retest at 8–12 weeks.
5. If you are planning pregnancy: take 0.4–0.8 mg of folic acid as recommended. The variant does not change that.
6. If you are TT with high blood pressure: riboflavin is a reasonable low-risk add-on to discuss with whoever manages your blood pressure.

Frequently Asked Questions

Should I avoid folic acid if I have MTHFR? No. Folic acid lowered homocysteine in TT carriers in randomised trials — by more than in other genotypes. Methylfolate is an acceptable alternative, not a requirement.

Is methylfolate better? Slightly, on some blood markers, and it avoids unmetabolised folic acid. It has not been shown to be better on any clinical outcome, and it has never been tested for preventing neural tube defects.

What about A1298C? Homozygous A1298C reduces enzyme activity less than 677TT and has a much weaker effect on homocysteine. Compound heterozygotes (one copy of each) sit in between. The same rule applies: measure homocysteine.

Does MTHFR cause my fatigue, anxiety or "poor detox"? There is no trial evidence linking the variant to these symptoms. If homocysteine is normal, the variant is not doing anything measurable.

Why did my doctor refuse to test? Because the guideline advises against it. A homocysteine test answers the same question more directly.

Related Research

• The Methylation Support Protocol
• Homocysteine: What Your Level Means and How to Lower It
• The Methylation Support Protocol: Test, Then Target
• Methylfolate
• Riboflavin
• Vitamin B12
• The Biohacker's Blood Panel: 40 Biomarkers