THE EVIDENCE BASE

MTHFR A1298C references: the verified source library

Every factual claim on this site traces to one of these records. Each entry shows its evidence class, what it supports and what it does not show; citation markers like [2] throughout the site link here. Null and unfavourable results are listed at equal prominence — they are the majority of the A1298C literature.

All citations fetch-verified against PubMed, ClinicalTrials.gov and Ensembl REST on 2026-08-09.

Discovery and variant biochemistry

[1]Case-control / biochemical1988

Kang SS, et al. Metabolism, 37:611-613 (1988)

Kang SS, et al. Metabolism. 1988;37:611-613. PMID: 3386531.

Original source ↗ · doi:10.1016/0026-0495(88)90076-5 · PMID 3386531

What it supports: The thermolabile-MTHFR discovery: heat-treating lymphocyte extracts at 46 °C for five minutes left controls with 37.6% ± 5.6% of MTHFR activity, but 6 of 21 young coronary-artery-disease patients retained only 13.6% ± 5.1% — a heat-sensitive enzyme, running in families and, apparently, in arteries.
What it does not show: A biochemical phenotype in lymphocyte extracts from a small case series; the underlying genotype was still seven years away.

[2]Family / genetic1991

Kang SS, et al. Am J Hum Genet, 48:536-545 (1991)

Kang SS, et al. Am J Hum Genet. 1991;48:536-545. PMID: 1998339.

Original source ↗ · PMID 1998339

What it supports: Extended the 1988 observation into inheritance: thermolabile MTHFR runs in families — the trait was heritable before anyone knew its molecular basis.
What it does not show: Pre-molecular genetics: the variant behind the trait had not yet been identified.

[3]Case-control1993

Kang SS, et al. Circulation, 88:1463-1469 (1993)

Kang SS, et al. Circulation. 1993;88:1463-1469. PMID: 8403293.

Original source ↗ · doi:10.1161/01.cir.88.4.1463 · PMID 8403293

What it supports: Further case-control extension of the thermolabile-MTHFR coronary story — the body of association evidence that made solving the variant urgent.
What it does not show: Association-level evidence in selected clinical samples; mechanism still unknown at the time.

[4]Case-control / biochemical1995

Frosst P, et al. Nat Genet, 10:111-113 (1995)

Frosst P, et al. Nat Genet. 1995;10:111-113. PMID: 7647779.

Original source ↗ · doi:10.1038/ng0595-111 · PMID 7647779

What it supports: Pinned the thermolabile ghost to a single base: 677C→T (Ala222Val), present on about 38% of unselected chromosomes — one of the most common “disease” variants ever described. Homozygotes show elevated homocysteine; it is the strongest common genetic cause of mild hyperhomocysteinemia. The same paper defines the MTHFR product as the predominant circulatory form of folate and the carbon donor for homocysteine remethylation.
What it does not show: About C677T, not A1298C — the second common variant was discovered three years later.

[5]Case-control / biochemical1998

A second common mutation in the MTHFR gene: an additional risk factor for neural-tube defects?

van der Put NMJ, et al. Am J Hum Genet. 1998;62:1044-1051. PMID: 9545395.

Original source ↗ · doi:10.1086/301825 · PMID 9545395

What it supports: The A1298C discovery paper: 1298C allele frequency 0.33 in a Dutch sample; lymphocyte MTHFR activity decreased, most pronounced in homozygotes (ANOVA P<.0001). Neither 1298CC nor 1298AC was associated with higher homocysteine or lower folate. Compound heterozygotes (677CT/1298AC) showed the C677T-homozygote-like phenotype — reduced specific activity (P<.0001), higher homocysteine and decreased plasma folate (P<.03) — with an NTD odds ratio of 2.04 whose 95% CI of 0.9–4.7 crosses 1.
What it does not show: A1298C alone did not raise NTD risk, and the compound-heterozygosity signal was not statistically significant; the authors framed it as an additional genetic risk factor while the interval admitted no effect.

[6]Biochemical / cohort1998

Weisberg I, et al. Mol Genet Metab, 64:169-172 (1998)

Weisberg I, et al. Mol Genet Metab. 1998;64:169-172. PMID: 9719624.

Original source ↗ · doi:10.1006/mgme.1998.2714 · PMID 9719624

What it supports: The independent discovery of A1298C: 1298CC homozygotes carried about 60% of control enzyme activity and about 10% of Canadian individuals were 1298CC; about 15% were compound heterozygotes (677T + 1298C) with 50–60% of control activity — lower than single C677T heterozygotes. No individual homozygous for both variants was observed.
What it does not show: A Canadian sample with biochemical phenotyping; genotype frequencies vary by ancestry.

[7]In-vitro + cohort2001

Weisberg IS, et al. Atherosclerosis, 156:409-415 (2001)

Weisberg IS, et al. Atherosclerosis. 2001;156:409-415. PMID: 11395038.

Original source ↗ · doi:10.1016/s0021-9150(00)00671-7 · PMID 11395038

What it supports: Recombinant expression separated the two variants: the 1298 mutant retained 68% ± 5.0 of wild-type activity (against 45% ± 10.8 for the 677 mutant) and was not thermolabile. In the Family Heart Study, the 1298 mutation alone did not raise homocysteine; only double heterozygotes showed significantly higher fasting homocysteine (P<0.05).
What it does not show: Heterologous expression measures the protein outside its cellular context; the in-vivo mechanism of the mild cellular effect is less clear than for C677T.

[8]In-vitro2001

Yamada K, et al. PNAS, 98:14853-14858 (2001)

Yamada K, et al. Proc Natl Acad Sci USA. 2001;98:14853-14858. PMID: 11742092.

Original source ↗ · doi:10.1073/pnas.261469998 · PMID 11742092

What it supports: The famous caveat: highly purified recombinant human Glu429Ala MTHFR was “indistinguishable from the wild-type enzyme” in biochemical properties — the cellular activity reduction measured in lymphocyte extracts does not survive purification. The same work established C677T’s mechanism: accelerated FAD cofactor loss, slowed by folate and SAM binding.
What it does not show: Purified-protein biochemistry removes the cellular environment; a normal purified enzyme and a real cellular phenotype can both be true.

[9]In-vitro / structure1999

Guenther BD, et al. Nat Struct Biol, 6:359-365 (1999)

Guenther BD, et al. Nat Struct Biol. 1999;6:359-365. PMID: 10201405.

Original source ↗ · doi:10.1038/7594 · PMID 10201405

What it supports: The domain architecture: MTHFR carries a C-terminal regulatory domain that binds SAM, the enzyme’s allosteric inhibitor — the domain in which A1298C (Glu429Ala) sits, distinct from the N-terminal catalytic domain that carries C677T (Ala222Val).
What it does not show: Structural and in-vitro work; it establishes architecture, not clinical effect.

[10]In-vitro / structure2006

Pejchal R, et al. Biochemistry, 45:4808-4818 (2006)

Pejchal R, et al. Biochemistry. 2006;45:4808-4818. PMID: 16605249.

Original source ↗ · doi:10.1021/bi052294c · PMID 16605249

What it supports: The C677T mechanism in structural detail: the Ala222Val substitution accelerates FAD cofactor loss and monomer dissociation, producing the thermolabile enzyme — why C677T is the stronger of the two common variants.
What it does not show: The FAD-loss mechanism is established for C677T, not for A1298C; the 1298 variant shows no thermolability and no FAD-loss behaviour.

[11]Cohort (n=10,601)2007

Fredriksen A, et al. Hum Mutat, 28:856-865 (2007)

Fredriksen A, et al. Hum Mutat. 2007;28:856-865. PMID: 17436311.

Original source ↗ · doi:10.1002/humu.20522 · PMID 17436311

What it supports: The nomenclature and the nuance: anchors the modern name c.1286A>C (known as 1298A>C; p.Glu429Ala), and in the largest dataset (10,601 Norwegians) c.1286A>C did show statistically significant (P<0.01) associations with homocysteine, folate and betaine — far weaker than c.665C>T’s strong associations.
What it does not show: A Norwegian cohort; the A1298C-alone signals are weak and sit in the shadow of C677T and of vitamin status.

[12]Case report2001

Tonetti C, et al. J Inherit Metab Dis (2001)

Tonetti C, et al. J Inherit Metab Dis. 2001. PMID: 11916316.

Original source ↗ · doi:10.1023/a:1013988123902 · PMID 11916316

What it supports: The exon-mapping anchor: published MTHFR case literature places “the polymorphism 1298A>C on exon 7… 677C>T on exon 4”, consistent with Ensembl VEP transcript mapping for rs1801131.
What it does not show: Case literature cited for exon mapping; volume and pages were not captured in the 2026-08-09 verification pass.

[13]Case report2011

Prasad AN, et al. Brain Dev (2011)

Prasad AN, et al. Brain Dev. 2011. PMID: 21778025.

Original source ↗ · doi:10.1016/j.braindev.2011.05.014 · PMID 21778025

What it supports: Context for the far end of the same gene: severe MTHFR deficiency (a rare recessive disorder) and epilepsy — a reminder that the common polymorphisms discussed on this site are not the severe-deficiency state.
What it does not show: A case report about severe MTHFR deficiency, not about the common A1298C or C677T variants.

Population genetics

[14]Database (gnomAD)2020

Karczewski KJ, et al. Nature, 581:434-443 (2020)

Karczewski KJ, et al. Nature. 2020;581:434-443. PMID: 32461654.

Original source ↗ · doi:10.1038/s41586-020-2308-7 · PMID 32461654

What it supports: The gnomAD exome resource behind every allele frequency on this site: global 1298C (G-allele) frequency ≈ 0.307, non-Finnish European 0.313, South Asian 0.411, East Asian 0.205, African/African-American 0.157 (retrieved via Ensembl REST for rs1801131, 2026-08-09).
What it does not show: Exome allele frequencies; the genotype percentages quoted on this site are Hardy-Weinberg arithmetic from those frequencies, not direct genotype counts.

Homocysteine — observational evidence

[15]Meta-analysis2002

Homocysteine and risk of ischemic heart disease and stroke: a meta-analysis

Homocysteine Studies Collaboration. JAMA. 2002;288:2015-2022. PMID: 12387654.

Original source ↗ · doi:10.1001/jama.288.16.2015 · PMID 12387654

What it supports: The individual-participant meta-analysis (30 studies; 5073 IHD events, 1113 stroke events): a 25% lower usual homocysteine (about 3 µmol/L) was associated with 11% lower IHD risk (OR 0.89, 95% CI 0.83–0.96) and 19% lower stroke risk (OR 0.81, 95% CI 0.69–0.95) in prospective studies.
What it does not show: Observational: associations were stronger in retrospective studies, and the authors concluded homocysteine is at most a modest independent predictor.

[16]Meta-analysis2002

Homocysteine and cardiovascular disease: evidence on causality from a meta-analysis

Wald DS, Law M, Morris JK. BMJ. 2002. PMID: 12446535.

Original source ↗ · doi:10.1136/bmj.325.7374.1202 · PMID 12446535

What it supports: The causality-framed companion meta-analysis of the homocysteine–cardiovascular association — the optimistic reading of the observational record, against which the trial era later pushed back.
What it does not show: Built on observational evidence; causal framing was later tested and largely not borne out by the large intervention trials in folate-replete populations.

[17]Meta-analysis2002

MTHFR 677C→T polymorphism and risk of coronary heart disease: a meta-analysis

Klerk M, Verhoef P, Clarke R, et al. JAMA. 2002. PMID: 12387655.

Original source ↗ · doi:10.1001/jama.288.16.2023 · PMID 12387655

What it supports: The classic Mendelian-randomization instrument: 677TT individuals had significantly higher coronary heart disease risk, “particularly in the setting of low folate status” — genotype as a natural experiment for lifelong slightly higher homocysteine.
What it does not show: A C677T finding; A1298C was not the instrument, and no equivalent A1298C instrument exists at this strength.

Homocysteine-lowering trials

[18]RCT2006

Lonn E, et al. (HOPE-2) N Engl J Med, 354:1567-1577 (2006)

Lonn E, et al. (HOPE-2). N Engl J Med. 2006;354:1567-1577. PMID: 16531613.

Original source ↗ · doi:10.1056/NEJMoa060900 · PMID 16531613

What it supports: Homocysteine lowering by B vitamins in 5522 patients with vascular disease or diabetes over 5 years: homocysteine fell (−2.4 µmol/L vs +0.8 on placebo) but the primary composite was null (RR 0.95, 95% CI 0.84–1.07). Fewer strokes (RR 0.75, 0.59–0.97) and more unstable-angina hospitalisations (RR 1.24, 1.04–1.49) sat inside the null.
What it does not show: Conducted in folate-replete, largely fortified populations; the trial lowered a biomarker without moving its primary clinical endpoint.

[19]RCT2006

Bønaa KH, et al. (NORVIT) N Engl J Med, 354:1578-1588 (2006)

Bønaa KH, et al. (NORVIT). N Engl J Med. 2006;354:1578-1588. PMID: 16531614.

Original source ↗ · doi:10.1056/NEJMoa055227 · PMID 16531614

What it supports: Post-myocardial-infarction B-vitamin lowering in 3749 patients: 27% homocysteine reduction with a null primary endpoint (RR 1.08, 95% CI 0.93–1.25) and a trend toward harm on triple therapy (RR 1.22, 1.00–1.50, P=0.05). The authors concluded that such treatment should not be used.
What it does not show: A post-MI population in a fortified era; results pushed the field away from routine homocysteine lowering.

[20]RCT2010

VITATOPS Trial Study Group Lancet Neurol, 9:855-865 (2010)

VITATOPS Trial Study Group. Lancet Neurol. 2010;9:855-865. PMID: 20688574.

Original source ↗ · doi:10.1016/S1474-4422(10)70187-3 · PMID 20688574

What it supports: In 8164 patients with recent stroke or TIA, B vitamins produced a composite relative risk of 0.91 (95% CI 0.82–1.00, P=0.05); the group stated the results do not support the use of B vitamins to prevent recurrent stroke.
What it does not show: A borderline P value on a composite endpoint in a folate-replete era; the conclusion stood against routine use.

[21]RCT2004

Toole JF, et al. (VISP) JAMA (2004)

Toole JF, et al. (VISP). JAMA. 2004. PMID: 14762035.

Original source ↗ · doi:10.1001/jama.291.5.565 · PMID 14762035

What it supports: The first of the big nulls: in ischemic stroke patients, B-vitamin homocysteine lowering produced no reduction in recurrent stroke, myocardial infarction or death.
What it does not show: A fortified-population trial; it set the pattern the later trials confirmed.

[22]RCT2010

Armitage JM, et al. (SEARCH Collaborative Group) JAMA (2010)

Armitage JM, et al. (SEARCH Collaborative Group). JAMA. 2010. PMID: 20571015.

Original source ↗ · doi:10.1001/jama.2010.840 · PMID 20571015

What it supports: The largest of the null trials: 12,064 myocardial-infarction survivors, with no benefit of homocysteine lowering on mortality or major morbidity.
What it does not show: Same fortified-era context as the other null trials; populations with low baseline folate were not the study ground.

[23]Meta-analysis of 8 RCTs2010

Clarke R, Halsey J, Lewington S, et al. Arch Intern Med (2010)

Clarke R, Halsey J, Lewington S, et al. Arch Intern Med. 2010. PMID: 20937919.

Original source ↗ · doi:10.1001/archinternmed.2010.348 · PMID 20937919

What it supports: The trial-level synthesis: across 8 RCTs and 37,485 participants, about 25% homocysteine reduction by folic acid had no significant effect on major vascular events (rate ratio 1.01, 95% CI 0.97–1.05), major coronary events (1.03, 0.97–1.10), stroke (0.96, 0.87–1.06), cancer or mortality.
What it does not show: Dominated by folate-replete populations; the analysis itself notes the results do not exclude benefit where folate status is low.

[24]Cochrane review2017

Martí-Carvajal AJ, et al. Cochrane Database Syst Rev (2017)

Martí-Carvajal AJ, et al. Cochrane Database Syst Rev. 2017. PMID: 28816346.

Original source ↗ · doi:10.1002/14651858.CD006612.pub5 · PMID 28816346

What it supports: The Cochrane synthesis: homocysteine-lowering interventions showed no effect on myocardial infarction or all-cause death, with only a small stroke effect favouring homocysteine lowering.
What it does not show: The small stroke signal sits against nulls on the harder cardiac endpoints; populations studied were largely folate-replete.

[25]RCT2015

Huo Y, et al. (CSPPT) JAMA, 313:1325-1335 (2015)

Huo Y, et al. (CSPPT). JAMA. 2015;313:1325-1335. PMID: 25771069.

Original source ↗ · doi:10.1001/jama.2015.2274 · PMID 25771069

What it supports: The notable exception: in 20,702 hypertensive Chinese adults — a largely non-fortified, folate-insufficient population, randomised within MTHFR C677T strata — enalapril plus folic acid versus enalapril alone reduced first stroke (2.7% vs 3.4%; HR 0.79, 95% CI 0.68–0.93) and first ischemic stroke (HR 0.76, 0.64–0.91) over a median 4.5 years.
What it does not show: Benefit appears contingent on baseline folate status; the trial stratified by C677T, not A1298C, and its setting is not the fortified world of the null trials.

[26]RCT sub-analysis2018

Qin X, et al. (CSPPT) Arterioscler Thromb Vasc Biol (2018)

Qin X, et al. Arterioscler Thromb Vasc Biol. 2018. PMID: 29371246.

Original source ↗ · doi:10.1161/ATVBAHA.117.310211 · PMID 29371246

What it supports: Within CSPPT, MTHFR genotype and baseline folate modified the homocysteine-lowering response — genotype matters most where the vitamin is scarcest.
What it does not show: A sub-analysis of one large trial in one population; about C677T strata, not A1298C.

[27]Cohort within RCT2017

CSPPT investigators Neurology (2017)

CSPPT investigators. Neurology. 2017. PMID: 29030456.

Original source ↗ · doi:10.1212/WNL.0000000000004648 · PMID 29030456

What it supports: Within CSPPT, the degree of homocysteine decline tracked stroke risk — the biomarker moved with the outcome inside the one large positive trial.
What it does not show: A cohort analysis inside a single trial; it does not resurrect the biomarker logic in the fortified populations where the large trials were null.

Riboflavin and the FAD cofactor

[28]Cohort (n=286)2002

McNulty H, et al. Am J Clin Nutr, 76:436-441 (2002)

McNulty H, et al. Am J Clin Nutr. 2002;76:436-441. PMID: 12145019.

Original source ↗ · doi:10.1093/ajcn/76.2.436 · PMID 12145019

What it supports: The riboflavin interaction: the high homocysteine of 677TT appeared only with low riboflavin status (mean tHcy 18.09 µmol/L in the lowest-riboflavin TT tertile, about twice the CC/CT level); adequate riboflavin rendered the TT group neutral with respect to homocysteine metabolism.
What it does not show: A C677T finding — the FAD-loss mechanism that motivates it does not apply to A1298C, which is not thermolabile.

[29]RCT (n=181)2010

Horigan G, et al. J Hypertens, 28:478-486 (2010)

Horigan G, et al. J Hypertens. 2010;28:478-486. PMID: 19952781.

Original source ↗ · doi:10.1097/HJH.0b013e328334c126 · PMID 19952781

What it supports: The genotype-guided blood-pressure trial: 16 weeks of riboflavin lowered blood pressure only in 677TT patients with premature cardiovascular disease (144/87 to 131/80 mmHg; P<0.05 systolic and diastolic); no response in CT or CC.
What it does not show: A C677T-stratified trial; there is no equivalent riboflavin-intervention trial for A1298C.

[30]RCT crossover follow-up (n=31)2012

Wilson CP, et al. Am J Clin Nutr, 95:766-772 (2012)

Wilson CP, et al. Am J Clin Nutr. 2012;95:766-772. PMID: 22277556.

Original source ↗ · doi:10.3945/ajcn.111.026245 · PMID 22277556

What it supports: Replication in the same 677TT patients four years later: riboflavin again lowered blood pressure (−9.2 ± 12.8 mmHg systolic, P=0.001; −6.0 ± 9.9 diastolic, P=0.003).
What it does not show: A small crossover follow-up in 677TT carriers only.

[31]RCT (n=91)2013

Wilson CP, et al. Hypertension, 61:1302-1308 (2013)

Wilson CP, et al. Hypertension. 2013;61:1302-1308. PMID: 23608654.

Original source ↗ · doi:10.1161/HYPERTENSIONAHA.111.01047 · PMID 23608654

What it supports: In treated hypertensive 677TT patients — over 60% of whom had failed to reach goal blood pressure on drugs alone — riboflavin added a systolic treatment effect of −5.6 ± 2.6 mmHg (P=0.033); diastolic was not significant.
What it does not show: A targeted 677TT hypertensive population; not a general-population finding.

[32]Cohort (n=6076)2020

Ward M, et al. BMC Med, 18:318 (2020)

Ward M, et al. BMC Med. 2020;18:318. PMID: 33172445.

Original source ↗ · doi:10.1186/s12916-020-01780-x · PMID 33172445

What it supports: The JINGO project cohort: 677TT (12% of Irish adults) carried a hypertension odds ratio of 1.42 (95% CI 1.07–1.90), rising to 3.00 (1.34–6.68) when TT combined with riboflavin deficiency — an effect independent of homocysteine, with poorer blood-pressure control on medication in TT (30% vs 45% in CC, P<0.027).
What it does not show: Observational; a C677T-and-riboflavin story, not an A1298C one.

[33]Cohort + RCT sub-study2020

Rooney M, et al. Biochimie, 173:91-99 (2020)

Rooney M, et al. Biochimie. 2020;173:91-99. PMID: 32330571.

Original source ↗ · doi:10.1016/j.biochi.2020.04.004 · PMID 32330571

What it supports: The methylation-cycle readout in humans: 677TT adults showed lower plasma SAM (74.7 ± 21.0 vs 85.2 ± 22.6 nmol/L, P=0.013) and a lower SAM:SAH ratio (1.66 ± 0.55 vs 1.85 ± 0.51, P=0.043) than CC, with higher homocysteine; riboflavin raised SAM and cystathionine in TT adults.
What it does not show: Measured for the stronger C677T variant; equivalent A1298C-specific methylation-cycle measurements are not part of the record.

[34]RCT2020

Amenyah SD, et al. Biochimie, 173:17-26 (2020)

Amenyah SD, et al. Biochimie. 2020;173:17-26. PMID: 32334045.

Original source ↗ · doi:10.1016/j.biochi.2020.04.007 · PMID 32334045

What it supports: Riboflavin supplementation altered global and MTHFR-locus DNA methylation in 677TT adults — the methyl-donor cycle measurably touched the epigenome in the genotype with the weakest enzyme.
What it does not show: A C677T-stratified intervention; DNA-methylation endpoints are mechanistic readouts, not clinical outcomes.

[35]RCT2021

Amenyah SD, et al. Int J Cardiol (2021)

Amenyah SD, et al. Int J Cardiol. 2021. PMID: 32920065.

Original source ↗ · doi:10.1016/j.ijcard.2020.09.011 · PMID 32920065

What it supports: DNA methylation of hypertension-related genes responded to riboflavin in 677TT adults — the epigenetic thread of the genotype-guided blood-pressure story.
What it does not show: Volume and pages were not captured in the 2026-08-09 verification pass; a C677T-stratified mechanistic trial.

[36]Cochrane review2025

Bradbury KE, et al. Cochrane Database Syst Rev, 10:CD015464 (2025)

Bradbury KE, et al. Cochrane Database Syst Rev. 2025;10:CD015464. PMID: 41123035.

Original source ↗ · doi:10.1002/14651858.CD015464.pub2 · PMID 41123035

What it supports: The sceptical counterweight: across 4 RCTs and 374 participants, the Cochrane review of riboflavin for blood pressure found the evidence “very uncertain” (systolic MD −1.94 mmHg, 95% CI −5.74 to 1.86, P=0.32, very low certainty; diastolic −3.03, −5.97 to −0.09, very low certainty), with most studies at high risk of bias and a call for large, well-conducted trials.
What it does not show: The genotype-specific trials are small; the Cochrane verdict governs how strongly the riboflavin–blood-pressure story may be told.

[37]Narrative review2025

Duffy B, et al. BMJ Nutr Prev Health (2025)

Duffy B, et al. BMJ Nutr Prev Health. 2025. PMID: 41768514.

Original source ↗ · doi:10.1136/bmjnph-2025-001212 · PMID 41768514

What it supports: The field’s most developed gene–nutrient interaction reviewed: randomised trials show riboflavin lowers blood pressure specifically in MTHFR 677TT individuals, with extension into hypertensive disorders of pregnancy proposed.
What it does not show: A C677T story, useful here as the contrast showing what a mature genotype-guided intervention looks like; not A1298C evidence.

Folate vitamers — 5-MTHF and folic acid

[38]In-vitro (human liver, n=6)2009

Bailey SW, Ayling JE. PNAS, 106:15424-15429 (2009)

Bailey SW, Ayling JE. Proc Natl Acad Sci USA. 2009;106:15424-15429. PMID: 19706381.

Original source ↗ · doi:10.1073/pnas.0902072106 · PMID 19706381

What it supports: The DHFR bottleneck: human hepatic dihydrofolate reductase reduces folic acid at under 2% of the rat rate at physiological pH, with about five-fold inter-individual variation — the kinetic reason unmetabolised folic acid can appear in plasma.
What it does not show: In-vitro enzymology on a small human-liver panel; it establishes capacity, not a clinical consequence.

[39]Cohort (Framingham Offspring)2008

Kalmbach RD, et al. Am J Clin Nutr, 88:763-768 (2008)

Kalmbach RD, et al. Am J Clin Nutr. 2008;88:763-768. PMID: 18779294.

Original source ↗ · doi:10.1093/ajcn/88.3.763 · PMID 18779294

What it supports: The UMFA observation: after US folic-acid fortification, median plasma unmetabolised folic acid doubled in non-supplement users (0.25 to 0.50 nmol/L, P<0.001) and the prevalence of high circulating UMFA rose from 9.4% to 19.1%. The authors’ own framing: the biochemical and physiologic consequences of this are unknown.
What it does not show: An exposure observation, not an outcome: whether chronic low-level UMFA harms anyone remains unresolved.

[40]PK crossover RCT (n=24)2009

Prinz-Langenohl R, et al. Br J Pharmacol, 158:2014-2021 (2009)

Prinz-Langenohl R, et al. Br J Pharmacol. 2009;158:2014-2021. PMID: 19917061.

Original source ↗ · doi:10.1111/j.1476-5381.2009.00492.x · PMID 19917061

What it supports: Single-dose pharmacokinetics in women: [6S]-5-MTHF gave higher AUC and Cmax of plasma folate than equimolar folic acid in both 677CC and TT women, and UMFA appeared regularly after folic acid but rarely after 5-MTHF.
What it does not show: Single-dose pharmacokinetics, not clinical outcomes.

[41]PK crossover RCT (n=24)2020

Obeid R, et al. Nutrients, 12:3623 (2020)

Obeid R, et al. Nutrients. 2020;12:3623. PMID: 33255787.

Original source ↗ · doi:10.3390/nu12123623 · PMID 33255787

What it supports: Modern pharmacokinetic confirmation: (6S)-5-MTHF sodium versus folic acid gave AUC0–8h of 126.0 vs 56.0 nmol/L·h and Cmax of 36.8 vs 11.1 nmol/L (P<0.001); transient UMFA (2.5–4.7 nmol/L) appeared only after folic acid. Sodium and calcium 5-MTHF salts were pharmacokinetically similar.
What it does not show: Short-window pharmacokinetics; biomarker exposure, not disease endpoints.

[42]RCT (n=144, 24 wk)2006

Lamers Y, et al. Am J Clin Nutr, 84:156-161 (2006)

Lamers Y, et al. Am J Clin Nutr. 2006;84:156-161. PMID: 16825690.

Original source ↗ · doi:10.1093/ajcn/84.1.156 · PMID 16825690

What it supports: Red-blood-cell folate — the long-haul status marker — rose more with [6S]-5-MTHF than with folic acid over 24 weeks in women (P<0.001).
What it does not show: A biomarker endpoint; no clinical outcome was measured.

[43]RCT2004

Lamers Y, et al. Am J Clin Nutr, 79:473-478 (2004)

Lamers Y, et al. Am J Clin Nutr. 2004;79:473-478. PMID: 14985224.

Original source ↗ · doi:10.1093/ajcn/79.3.473 · PMID 14985224

What it supports: The equivalence result the debate must absorb: [6S]-5-MTHF and folic acid lowered plasma homocysteine equally.
What it does not show: Homocysteine is a biomarker; equal biomarker lowering is not evidence of equal clinical outcomes.

[44]RCT (n=160)2002

Fohr IP, et al. Am J Clin Nutr, 75:275-282 (2002)

Fohr IP, et al. Am J Clin Nutr. 2002;75:275-282. PMID: 11815318.

Original source ↗ · doi:10.1093/ajcn/75.2.275 · PMID 11815318

What it supports: The genotype-dependence result: the homocysteine response to either folate form was MTHFR-genotype-dependent; folic acid cut homocysteine by at least 13% in all C677T genotypes, TT women benefited most from either form, and in CC/CT women folic acid outperformed racemic 5-MTHF at the doses used.
What it does not show: Racemic 5-MTHF at then-current doses; a biomarker trial, stratified by C677T rather than A1298C.

[45]Narrative review2014

Folate, folic acid and 5-methyltetrahydrofolate are not the same thing

Scaglione F, Panzavolta G. Xenobiotica. 2014;44:480-488. PMID: 24494987.

Original source ↗ · doi:10.3109/00498254.2013.845705 · PMID 24494987

What it supports: The position review for the 5-MTHF case: argues that 5-MTHF bypasses MTHFR-polymorphism effects, avoids masking B12-deficiency haematology and avoids unmetabolised folic acid.
What it does not show: Argumentation, not trial-derived outcome data; no outcomes RCT has shown superior clinical results for 5-MTHF over folic acid.

[46]Feasibility RCT (22 couples)2026

Ledowsky C, Scarf V, Rogers K, Steel A. Nutr Res (2026)

Ledowsky C, Scarf V, Rogers K, Steel A. Nutr Res. 2026. PMID: 41544303.

Original source ↗ · doi:10.1016/j.nutres.2025.12.008 · PMID 41544303

What it supports: The formulation question reaches trial stage: in a double-blind feasibility RCT of 5-MTHF versus folic acid prenatal vitamins in couples with recurrent pregnancy loss, acceptability (86–94%) and adherence (above 78%) were high, unmetabolised folic acid fell on 5-MTHF and rose on folic acid — and a notable finding was 5-MTHF degradation in retained samples, a formulation-stability warning for the whole vitamer debate.
What it does not show: A feasibility trial; it establishes that a full trial can run, not which vitamer wins outcomes.

Relevance to Panacea research · The instability finding is the point of contact with Panacea’s preservation research: the molecule at the centre of the methylation cycle is labile in real formulations, and stability is an engineering question in its own right.

Pregnancy and development

[47]Meta-analysis (22 studies)2022

Association between Fetal MTHFR A1298C (rs1801131) Polymorphism and Neural Tube Defects Risk: A Systematic Review and Meta-Analysis

Soleimani-Jadidi S, et al. Fetal Pediatr Pathol. 2022. PMID: 32536231.

Original source ↗ · doi:10.1080/15513815.2020.1764682 · PMID 32536231

What it supports: The largest A1298C–NTD synthesis is null: across 22 case-control studies (3,224 NTD fetuses vs 3,295 controls), no evidence for a significant association between fetal MTHFR A1298C and NTD risk — overall, and in every ethnicity, country and NTD-type stratum.
What it does not show: Case-control pooling; the null concerns the fetal genotype and A1298C specifically — maternal C677T remains an accepted modest NTD risk factor.

[48]Meta-analysis (124 articles)2023

Wen Y, He H, Zhao K. J Assist Reprod Genet (2023)

Wen Y, He H, Zhao K. J Assist Reprod Genet. 2023. PMID: 37248348.

Original source ↗ · doi:10.1007/s10815-023-02823-x · PMID 37248348

What it supports: The positive side of the recurrent-pregnancy-loss conflict: across 124 articles (17,278 RPL patients, 16,021 controls), MTHFR A1298C was significantly associated with RPL (dominant OR 1.66, 95% CI 1.26–2.18; recessive OR 1.79, 1.42–2.26), alongside C677T and other thrombophilia variants.
What it does not show: Reported beside its contradiction: the 2019 umbrella review found no dependable A1298C–RPL correlation; signals concentrate in Asian datasets and recessive models.

[49]Umbrella review (11 SR/MAs)2019

Polymorphisms of MTHFR in recurrent pregnancy loss: an overview of systematic reviews and meta-analyses

Du B, Shi X, Yin C. J Assist Reprod Genet. 2019. PMID: 31254142.

Original source ↗ · doi:10.1007/s10815-019-01473-2 · PMID 31254142

What it supports: The contradicting synthesis: across eleven systematic reviews and meta-analyses, no significant correlation was found between A1298C and recurrent pregnancy loss (only the heterozygous model reached significance, OR 1.62, 95% CI 1.17–2.25); C677T was the consistent signal (homozygous OR 2.08).
What it does not show: Umbrella reviews inherit the quality of the pooled reviews; the field’s signals are ethnicity- and model-dependent.

[50]Meta-analysis2015

Folate metabolism gene polymorphisms MTHFR C677T and A1298C and risk for preeclampsia: a meta-analysis

Wu X, et al. J Assist Reprod Genet. 2015. PMID: 25758986.

Original source ↗ · doi:10.1007/s10815-014-0408-8 · PMID 25758986

What it supports: Preeclampsia at meta level: C677T was associated with preeclampsia risk (T vs C OR 1.157, 95% CI 1.057–1.266); for A1298C (1,103 cases, 988 controls) the authors could not demonstrate an increased risk (p = 0.667).
What it does not show: A null for A1298C with modest numbers; small-study signals in either direction exist beneath the meta level.

[51]Case-control2024

Nasri K, et al. Birth Defects Res (2024)

Nasri K, et al. Birth Defects Res. 2024. PMID: 38716581.

Original source ↗ · doi:10.1002/bdr2.2333 · PMID 38716581

What it supports: The gene–nutrient frontier in NTD research: fatty-acid profiles interacted with MTHFR and MTRR genotypes in Tunisian NTD families — the field has shifted from single-variant association to interaction structure.
What it does not show: A single case-control study; the authors state further research is required.

Brain and mind

[52]Meta-analysis (HUGE review)2007

MTHFR genetic polymorphisms and psychiatric disorders: a HuGE review

Gilbody S, Lewis S, Lightfoot T. Am J Epidemiol. 2007. PMID: 17074966.

Original source ↗ · doi:10.1093/aje/kwj347 · PMID 17074966

What it supports: The psychiatric baseline: C677T TT vs CC in unipolar depression gave OR 1.36 (95% CI 1.11–1.67; 1,280 cases, 10,429 controls, I²=0%). A1298C was included in the planned primary comparisons, and no A1298C association was demonstrated.
What it does not show: The psychiatric signal belongs to C677T; A1298C was examined and came out null.

[53]Meta-analysis2022

Association between variants of MTHFR genes and psychiatric disorders: A meta-analysis

Zhang YX, et al. Front Psychiatry. 2022. PMID: 36061291.

Original source ↗ · doi:10.3389/fpsyt.2022.976428 · PMID 36061291

What it supports: Fifteen years later, the pattern held: C677T was significantly related to schizophrenia and major depression, while the MTHFR A1298C polymorphism was only marginally linked to major depression.
What it does not show: “Marginal” is not an established association; heterogeneity across psychiatric genetics is high.

[54]RCT (NCT02709668)2016

Mech AW, Farah A. J Clin Psychiatry (2016)

Mech AW, Farah A. J Clin Psychiatry. 2016. PMID: 27035272.

Original source ↗ · doi:10.4088/JCP.15m10166 · PMID 27035272

What it supports: The one small positive intervention: in a double-blind, placebo-controlled trial enrolling MTHFR C677T- or A1298C-positive depressed patients, a reduced-B-vitamin and micronutrient combination separated from placebo by week 2; 42% of the treatment arm reached remission by week 8, and improvement correlated with homocysteine reduction.
What it does not show: A single small trial; independent replication is lacking, and the enrolment pooled both variants.

[55]Systematic review2026

Systematic Review of MTHFR 677C>T and 1298A>C Variants and Treatment-Resistant Depression

Brum Moraes J, et al. Complex Psychiatry. 2026. PMID: 41737528.

Original source ↗ · doi:10.1159/000548757 · PMID 41737528

What it supports: The 2026 synthesis for treatment-resistant depression: evidence suggests a role for both variants, but heterogeneity among studies limits definitive conclusions; L-methylfolate as an adjunct in carriers is an open research question, not established practice.
What it does not show: The review itself calls for standardised definitions and integration studies before conclusions can be drawn.

[56]Meta-analysis (26 studies)2019

Effects of MTHFR C677T and A1298C Polymorphisms on Migraine Susceptibility: A Meta-Analysis of 26 Studies

Liu L, et al. Headache. 2019. PMID: 31045246.

Original source ↗ · doi:10.1111/head.13540 · PMID 31045246

What it supports: Migraine: the 677T allele was a risk factor for migraine with aura in Caucasians; pooling the five A1298C studies gave 1298CC vs AA OR 1.78 (95% CI 1.03–3.07) overall and OR 2.83 (1.30–6.16) for migraine without aura — a positive but fragile signal on few studies and wide intervals.
What it does not show: Few A1298C studies, wide intervals; a later meta-analysis found A1298C reached significance only in codominant and recessive models.

[57]Meta-analysis2013

Pu D, Shen Y, Wu J. Autism Res (2013)

Pu D, Shen Y, Wu J. Autism Res. 2013. PMID: 23653228.

Original source ↗ · doi:10.1002/aur.1300 · PMID 23653228

What it supports: The early positive: a 2013 meta-analysis suggested MTHFR associations with autism spectrum disorder — the starting point the larger later syntheses would test.
What it does not show: Superseded for A1298C by the larger 2020 and 2026 syntheses, which found no A1298C–ASD association.

[58]Meta-analysis2020

Li Y, et al. BMC Pediatr (2020)

Li Y, et al. BMC Pediatr. 2020. PMID: 32972375.

Original source ↗ · doi:10.1186/s12887-020-02330-3 · PMID 32972375

What it supports: The refutation: C677T is an autism-susceptibility factor in the pooled data, but the MTHFR A1298C polymorphism is not associated with ASD susceptibility.
What it does not show: Meta-analytic pooling of case-control studies; the A1298C null was later confirmed again in 2026.

[59]Meta-analysis2026

Pan Y, McDill B, Mooney M. Brain Sci (2026)

Pan Y, McDill B, Mooney M. Brain Sci. 2026. PMID: 41594814.

Original source ↗ · doi:10.3390/brainsci16010093 · PMID 41594814

What it supports: The 2026 confirmation, ethnicity-stratified: a statistical association between ASD and C677T, but none between ASD and the MTHFR A1298C variant.
What it does not show: The field’s A1298C verdict is now twice-replicated at meta level — an association that did not survive larger synthesis.

Cancer

[60]Meta-analysis (265 studies)2016

Association between the MTHFR A1298C polymorphism and risk of cancer: evidence from 265 case-control studies

Zhu XL, et al. Mol Genet Genomics. 2016. PMID: 26156333.

Original source ↗ · doi:10.1007/s00438-015-1082-y · PMID 26156333

What it supports: The largest A1298C–cancer synthesis (81,040 cases, 114,975 controls): direction depended on site and ethnicity — increased cervical cancer (AC vs AA OR 1.48, 95% CI 1.13–1.92) and lymphoma risk (CC vs AA OR 1.75, 1.21–2.53) in Asians, but decreased colorectal cancer risk in Asians (CC vs AA OR 0.77, 0.60–1.00).
What it does not show: Signals are site- and ethnicity-specific; no consistent A1298C cancer-risk elevation exists across sites, and the opposing directions have no demonstrated molecular explanation.

[61]Meta-analysis2025

Evaluating MTHFR gene polymorphisms and colorectal cancer risk through meta-analysis

Wang YW, et al. PLoS One. 2025. PMID: 40668818.

Original source ↗ · doi:10.1371/journal.pone.0305517 · PMID 40668818

What it supports: The protective-direction candidate: the analysis reports that MTHFR A1298C may play a protective role in the development of colorectal cancer — with the authors’ explicit caveat that further studies are required.
What it does not show: The protective signal is ethnicity-restricted and not settled; a biologically plausible folate-partitioning mechanism awaits confirmatory work.

[62]Meta-analysis2022

Lack of Impact of the A1298C MTHFR on the Risk of Childhood Acute Lymphoblastic Leukemia: Evidence from a Meta-analysis

Frikha R. Indian J Hematol Blood Transfus. 2022. PMID: 35496972.

Original source ↗ · doi:10.1007/s12288-021-01453-6 · PMID 35496972

What it supports: A clean null: no association between A1298C and childhood acute lymphoblastic leukemia in any genetic model (OR about 0.91, p above 0.05); the author concludes A1298C is unlikely to be a major risk gene for childhood ALL.
What it does not show: A single-disease synthesis; it rules out a large effect in childhood ALL, nothing broader.

[63]Meta-analysis2021

Rezaee M, et al. Biochem Genet (2021)

Rezaee M, et al. Biochem Genet. 2021. PMID: 33387129.

Original source ↗ · doi:10.1007/s10528-020-10020-z · PMID 33387129

What it supports: Breast cancer in Asian datasets: A1298C associated in allelic (OR 1.15, 95% CI 1.04–1.28) and homozygous (OR 1.38, 1.15–1.66) models, strongest in Western Asia.
What it does not show: Regional case-control literature is mixed in both directions; the signal is ethnicity-restricted.

Pharmacogenetics

[64]Cohort + meta-analysis2013

Owen SA, et al. Pharmacogenomics J (2013)

Owen SA, et al. Pharmacogenomics J. 2013. PMID: 21931346.

Original source ↗ · doi:10.1038/tpj.2011.42 · PMID 21931346

What it supports: The methotrexate null: pooled random-effects odds ratios showed no association of C677T or A1298C with methotrexate efficacy or toxicity in rheumatoid arthritis (A1298C efficacy OR 0.81, 95% CI 0.53–1.24; toxicity OR 1.19, 0.80–1.78); the authors concluded neither SNP has predictive utility in RA.
What it does not show: A null contested by a later meta-analysis that found a protective-direction signal for 1298CC — the pharmacogenetic literature is internally contradictory.

[65]Meta-analysis (12 studies)2014

Song GG, Bae SC, Lee YH. Clin Rheumatol (2014)

Song GG, Bae SC, Lee YH. Clin Rheumatol. 2014. PMID: 24794492.

Original source ↗ · doi:10.1007/s10067-014-2645-8 · PMID 24794492

What it supports: The divergent result: across 12 studies and 2,288 RA patients, 677TT associated with overall methotrexate toxicity (OR 1.615) while 1298CC associated with fewer adverse effects (OR 0.501, 95% CI 0.284–0.886) — a protective-direction signal for A1298C.
What it does not show: Contradicted by the Owen 2013 null; the field has not resolved which reading is correct.

[66]Systematic review / meta-analysis2024

Liu B, et al. World J Surg Oncol (2024)

Liu B, et al. World J Surg Oncol. 2024. PMID: 38212758.

Original source ↗ · doi:10.1186/s12957-023-03287-0 · PMID 38212758

What it supports: High-dose methotrexate in osteosarcoma: C677T may raise adverse-reaction risk, while existing studies have not found a significant correlation between MTHFR A1298C and adverse reactions — another A1298C null.
What it does not show: A null within one oncology protocol family; not a general statement about antifolate toxicity.

[67]Systematic review2026

The impact of gene polymorphisms on the response of methotrexate-based treatments

Esperón P, Vital M, Giletti A. Eur J Clin Pharmacol. 2026. PMID: 42104114.

Original source ↗ · doi:10.1007/s00228-026-04059-x · PMID 42104114

What it supports: The 2026 practice-level verdict: the MTHFR 677T allele and the 677T–1298A haplotype are the variants most consistently linked to methotrexate toxicity — but the evidence is of moderate quality, and CPIC and DPWG guidelines currently do not advise routine methotrexate dose adjustment based on single-gene variants; a pharmacogenetics-guided dosing guideline remains an elusive goal.
What it does not show: The haplotype signal keys on 677T–1298A; single-variant A1298C guidance is not part of any dosing recommendation framework.

Guidelines and broad reviews

[68]Guideline (ACMG)2013

Hickey SE, et al. (ACMG) Genet Med, 15:153-156 (2013)

Hickey SE, et al. (ACMG). Genet Med. 2013;15:153-156. PMID: 23288205.

Original source ↗ · doi:10.1038/gim.2012.165 · PMID 23288205

What it supports: The professional-guidance anchor: the American College of Medical Genetics and Genomics states that MTHFR polymorphism testing — for either common variant — has minimal clinical utility and should not be part of routine thrombophilia evaluation.
What it does not show: Practice guidance, not a primary study; it governs ordering behaviour, not research questions.

[69]Narrative review2015

Moll S, Varga EA. Circulation, 132:e6-e9 (2015)

Moll S, Varga EA. Circulation. 2015;132:e6-e9. PMID: 26149435.

Original source ↗ · doi:10.1161/CIRCULATIONAHA.114.013311 · PMID 26149435

What it supports: The pathway-logic anchor: 5-methylTHF donates its methyl group to homocysteine via methionine synthase (B12-dependent), methionine feeds SAM — the universal methyl donor — and demethylation yields SAH, whose hydrolysis regenerates homocysteine; the SAM:SAH ratio indexes methylation capacity.
What it does not show: A clinical review of established biochemistry; pathway logic, not variant-specific effect sizes.

[70]Systematic / narrative review2026

Sun T, et al. Autoimmun Rev (2026)

Sun T, et al. Autoimmun Rev. 2026. PMID: 41083014.

Original source ↗ · doi:10.1016/j.autrev.2025.103939 · PMID 41083014

What it supports: The broad 2026 reframing: C677T and A1298C across eight autoimmune disorders, with population-specific associations and mechanistic framing spanning inflammation, endothelial dysfunction and epigenetics.
What it does not show: Population-specific associations across heterogeneous diseases; a map of questions more than of answers.

[71]Case report2026

Saravanan YS, et al. Ewha Med J (2026)

Saravanan YS, et al. Ewha Med J. 2026. PMID: 41821223.

Original source ↗ · doi:10.12771/emj.2025.00941 · PMID 41821223

What it supports: Compound heterozygosity (677/1298) appearing in a case report of recurrent thrombosis — the kind of anecdote that keeps the compound genotype in the clinical conversation.
What it does not show: A case report is anecdotal by design; it generates hypotheses and proves nothing.

[72]Case-control2026

Magrì, et al. Int Dent J (2026)

Magrì, et al. Int Dent J. 2026. PMID: 42546423.

Original source ↗ · PMID 42546423

What it supports: Representative of the continuing 2024–2026 A1298C association literature: a null association with recurrent aphthous stomatitis — small hypothesis-generating case-control work that changes nothing in the larger picture.
What it does not show: A small single-domain case-control study; most of the recent A1298C association literature looks like this.

Active research fronts 2024–2026

[73]Case-control2025

Association of MTHFR C677T and A1298C Polymorphisms with First-Episode Myocardial Ischemia

Badea IA, et al. Genes. 2025. PMID: 40869906.

Original source ↗ · doi:10.3390/genes16080858 · PMID 40869906

What it supports: A current A1298C-specific cardiovascular data point: homozygous C677T and A1298C genotypes were both associated with first-episode myocardial ischemia, especially over age 50.
What it does not show: A small single-center study — hypothesis-generating, not confirmatory; no large A1298C-specific CVD meta-analysis shows a robust association.

[74]Cohort / interventional analysis2025

Association of MTHFR C677T, MTHFR A1298C, and MTRR A66G Polymorphisms with Hyperhomocysteinemia and Its Modulation by Vitamin B12 and Folate in Chinese Population with Hypertension

Siddiqi SM, et al. J Nutr. 2025. PMID: 39299473.

Original source ↗ · doi:10.1016/j.tjnut.2024.09.003 · PMID 39299473

What it supports: The overriding variable: adequate folate and B12 status markedly blunted the genotypic effect on homocysteine in hypertensive Chinese adults — genotype effects were small when vitamin status was adequate.
What it does not show: One population, one design (ChiCTR2100051983); it confirms the pattern rather than settling it.

[75]Case-control + bioinformatics2025

Maurya AK, et al. Mol Neurobiol (2025)

Maurya AK, et al. Mol Neurobiol. 2025. PMID: 41417174.

Original source ↗ · doi:10.1007/s12035-025-05621-x · PMID 41417174

What it supports: A dementia-genetics data point: in an Indian cohort (120 Alzheimer’s-disease cases, 120 controls), both C677T and A1298C associated with AD under dominant and over-dominant models (ORs 3.41–4.09), with higher homocysteine and lower folate in cases.
What it does not show: A single moderate-sized study; the effect sizes are large relative to the dementia-genetics literature and need replication.

[76]Retrospective cohort2025

Xiong X, et al. Front Nutr (2025)

Xiong X, et al. Front Nutr. 2025. PMID: 41473191.

Original source ↗ · doi:10.3389/fnut.2025.1729915 · PMID 41473191

What it supports: The gene–treatment angle in pregnancy: folic acid supplementation stratified by MTHFR and MTRR genotype in a Chinese retrospective cohort; folic acid may mitigate preeclampsia risk via PlGF levels.
What it does not show: Retrospective; an interaction hypothesis, not a demonstrated genotype-guided protocol.

[77]Pilot case-control2026

Badulescu OV, et al. Int J Mol Sci (2026)

Badulescu OV, et al. Int J Mol Sci. 2026. PMID: 41977299.

Original source ↗ · doi:10.3390/ijms27073112 · PMID 41977299

What it supports: The multigenic direction in recurrent pregnancy loss: a pilot case-control on multigenic thrombophilic risk including A1298C — single variants giving way to panels.
What it does not show: A pilot; the RPL literature for A1298C remains genuinely unsettled.

[78]Clinical registry — recruiting2024

China Stroke Primary Prevention Trial 2 for Participants With Hypertension and MTHFR 677 TT Genotype (CSPPT-2)

ClinicalTrials.gov. NCT04974151: China Stroke Primary Prevention Trial 2 for Participants With Hypertension and MTHFR 677 TT Genotype. Record consulted 2026-08-09.

Original source ↗

What it supports: The largest genotype-stratified stroke-prevention programme is recruiting: the 677TT arm of CSPPT-2 (amlodipine with folic acid or 5-MTHF comparison arms) — the field’s decisive test of genotype-guided folate intervention.
What it does not show: Stratified by C677T, not A1298C; a registry record documents design and status, not results.

[79]Clinical registry — recruiting2024

CSPPT2-CC/CT (H-type hypertension, MTHFR 677 CC/CT)

ClinicalTrials.gov. NCT04974138: CSPPT2-CC/CT (H-type hypertension, MTHFR 677 CC/CT with elevated homocysteine and low folate). Record consulted 2026-08-09.

Original source ↗

What it supports: The companion CSPPT-2 arm for 677 CC/CT participants with elevated homocysteine and low folate — the second half of the genotype-stratified prevention programme.
What it does not show: Recruiting; no results yet, and the stratification key is C677T.

[80]Clinical registry — recruiting2023

Genotype-personalized B-vitamin therapy in hyperhomocysteinemia

ClinicalTrials.gov. NCT06264570: genotype-personalized B-TMG (betaine) and B-SAM supplementation versus placebo in hyperhomocysteinemia. Record consulted 2026-08-09.

Original source ↗

What it supports: The personalised-nutrition front: genotype-personalised B-vitamin and methyl-donor (betaine, SAM) supplementation versus placebo in hyperhomocysteinemia — the methyl-donor cycle as an interventional target.
What it does not show: Recruiting; a biomarker-directed intervention trial, not an outcomes trial.

[81]Clinical registry — recruiting2026

Maternal Methyl-Nutrient Status and Infant Neurodevelopment Study

ClinicalTrials.gov. NCT07568561: Maternal Methyl-Nutrient Status and Infant Neurodevelopment Study (methylation, homocysteine, TMAO outcomes). Record consulted 2026-08-09.

Original source ↗

What it supports: The developmental front: maternal methyl-nutrient status tracked against infant neurodevelopment, with methylation and homocysteine among the measured outcomes.
What it does not show: An observational cohort; recruiting, with results years away.