Background
Metformin stands as the first-line pharmacological treatment for type 2 diabetes mellitus worldwide, prescribed to millions of patients due to its efficacy, safety profile, and cost-effectiveness. However, long-term metformin use has been consistently associated with vitamin B12 deficiency, a complication that can lead to serious neurological manifestations including peripheral neuropathy, cognitive impairment, and megaloblastic anemia. Previous research has identified clinical risk factors such as duration of metformin therapy, dosage, and baseline B12 levels, yet the underlying genetic predisposition to this adverse effect remained poorly understood.
The clinical burden of metformin-induced B12 deficiency is substantial. Studies suggest that approximately 10-30% of patients on long-term metformin therapy develop biochemical B12 deficiency, with many experiencing subclinical depletion that may progress to symptomatic disease. The variability in individual susceptibility has long puzzled clinicians, with some patients developing deficiency within a few years of treatment while others remain replete despite decades of use. This heterogeneity has led researchers to hypothesize that genetic factors may play a significant role in determining individual vulnerability to metformin’s effect on B12 absorption.
Study Design
The investigation employed a genome-wide association study (GWAS) approach to identify genetic variants associated with metformin-induced vitamin B12 deficiency. The discovery cohort comprised individuals from the UK Biobank, a large-scale biomedical database containing genetic and health information from approximately 500,000 participants in the United Kingdom. From this cohort, researchers identified 487 individuals with documented metformin-induced B12 deficiency and 6,686 metformin-tolerant control individuals who had not developed deficiency despite comparable metformin exposure.
Genome-wide association analysis was conducted using logistic regression under an additive genetic model. To ensure the robustness of findings, replication was undertaken in three independent cohorts: a Scottish cohort, the Diabetes Prevention Program Outcomes Study (DPPOS), and a separate cohort from Liverpool. Notably, the Liverpool cohort provided additional mechanistic insights through measurement of plasma metformin levels, allowing researchers to examine whether the identified genetic variant affected metformin pharmacokinetics.
Key Findings
The analysis revealed a genome-wide significant association between a non-synonymous single nucleotide polymorphism (SNP) in the cubilin gene (CUBN, rs1801222/p.S253F) and metformin-induced vitamin B12 deficiency. The cubilin gene encodes a protein essential for the intestinal absorption of the vitamin B12-intrinsic factor complex, making it a biologically plausible candidate gene for this adverse effect.
The association demonstrated a clear allele dose-response relationship under the additive genetic model (adjusted p=1.86×10⁻¹⁰). Compared with individuals carrying the GG genotype (reference), those with the AG genotype showed an odds ratio of 1.56 (95% CI 1.36, 1.79), while individuals with the AA genotype had a substantially elevated odds ratio of 2.43 (95% CI 1.85, 3.20). This stepwise increase in risk with each additional A allele provides compelling evidence for a causal relationship between the CUBN variant and metformin-induced B12 deficiency.
The clinical magnitude of this genetic effect was striking. In non-metformin-exposed individuals, vitamin B12 deficiency occurred at a baseline rate of 0.84-1.20% regardless of rs1801222 genotype, confirming that the genetic variant has minimal impact on B12 metabolism in the absence of metformin. However, a profound interaction with metformin use was observed. Among metformin-treated patients, B12 deficiency rates increased progressively with genotype: 6.02% in GG carriers, 7.96% in GA carriers, and 12.84% in AA carriers.
Perhaps most clinically relevant were the findings regarding time to B12 deficiency onset. Kaplan-Meier analysis of time from metformin initiation to B12 deficiency diagnosis revealed that 10% of AA genotype carriers developed deficiency by 11 years of treatment, compared with 21 years for the GG genotype group. This represents a nearly twofold acceleration in deficiency development, suggesting that AA carriers may require earlier and more frequent monitoring.
The replication cohorts provided consistent support for these findings. Both the Scottish cohort and the DPPOS demonstrated significant associations with the rs1801222 variant, confirming the generalizability of the discovery across diverse populations. Importantly, the Liverpool cohort’s metformin level measurements suggested that the genetic effect was not mediated by altered metformin pharmacokinetics, indicating that the CUBN variant likely affects B12 absorption through a mechanism specific to metformin’s interference with cubilin-mediated uptake.
Expert Commentary
The identification of the CUBN rs1801222 variant as a major determinant of metformin-induced B12 deficiency represents a paradigm shift in understanding this common adverse effect. From a clinical perspective, this finding suggests that genetic testing could potentially be employed to stratify patients by risk at the time of metformin initiation, allowing for personalized monitoring strategies.
The biological plausibility of this association is noteworthy. Cubilin serves as the primary receptor for the B12-intrinsic factor complex at the apical membrane of enterocytes in the distal ileum. Metformin has been shown to interfere with calcium-dependent B12 absorption, and cubilin-mediated endocytosis represents the pathway through which this absorption occurs. The rs1801222 variant results in a serine-to-phenylalanine substitution at position 253, potentially altering protein function or stability in a manner that exacerbates metformin’s inhibitory effect.
Several limitations merit consideration. First, the study relied on clinical diagnosis codes and laboratory values to define B12 deficiency, which may underestimate the true prevalence of subclinical deficiency. Second, while the association was robustly replicated, the absolute risk difference between genotypes, while statistically significant, remains modest in absolute terms. Third, the clinical utility of routine genetic testing for this variant would require careful health economic evaluation and consideration of implementation challenges in primary care settings.
Current clinical guidelines recommend periodic B12 monitoring for patients on metformin, though implementation remains inconsistent. The American Diabetes Association Standards of Care suggest assessing B12 levels periodically in patients with diabetes on metformin, particularly those with peripheral neuropathy or other risk factors. The identification of a genetic risk marker may prompt reconsideration of monitoring strategies, with more intensive surveillance potentially targeted toward high-risk genotypes.
Conclusion
This landmark study establishes the CUBN rs1801222 polymorphism as a significant genetic risk factor for metformin-induced vitamin B12 deficiency. The clear allele dose-response relationship, consistent replication across multiple cohorts, and biologically plausible mechanism collectively support the clinical validity of this finding. For patients initiating metformin therapy, carriers of the AA genotype may benefit from earlier and more frequent B12 monitoring, potentially preventing the neurological sequelae of prolonged deficiency. Future research should focus on validating these findings in prospective clinical trials and developing evidence-based guidelines for genotype-guided monitoring strategies. Until such guidelines are established, clinicians should remain vigilant for signs and symptoms of B12 deficiency in all patients on long-term metformin therapy, with heightened awareness for those with known genetic risk factors.
Funding
This research was supported by the UK Biobank and associated research grants. The Diabetes Prevention Program (DPP) was supported by the National Institutes of Health. Full funding disclosures are available in the original publication.
References
Baldwin FD, Bedair KF, Jorgensen AL, et al. Identification of a genetic risk factor for metformin-induced vitamin B12 deficiency. Diabetologia. 2026;69(4):953-965. PMID: 41537778.
American Diabetes Association. Standards of Care in Diabetes—2024. Diabetes Care. 2024;47(Suppl 1):S1-S456.
de Jager J, Kooy A, Lehert P, et al. Long term treatment with metformin in patients with type 2 diabetes and risk of vitamin B-12 deficiency: randomised placebo controlled trial. BMJ. 2010;340:c2181.