Too Much of a Good Thing? Cancer Risk and High Dose Folic Acid in Epilepsy

Abstract

High-dose Folic Acid Use and Cancer Risk in Women Who Have Given Birth: A Register-Based Cohort Study

Vegrim HM, Dreier JW, Igland J, Alvestad S, Gilhus NE, Gissler M, Leinonen MK, Tomson T, Zoega H, Christensen J, Bjork MH. Epilepsia. 2025; 66(1):75–88. doi: 10.1111/epi.18146. Epub 2024 Nov 14.

Objective: This study was undertaken to study whether high-dose folic acid (>1 mg daily) use is associated with an increased risk of cancer in all women who have given birth and in women with epilepsy. High-dose folic acid supplementation during pregnancy has been linked to increased cancer risk in children born to mothers with epilepsy. Methods: We identified women with their first pregnancy in Denmark (1997-2017), Norway (2005-2017), and Sweden (2006-2017) using medical birth registers, linking individual data across nationwide health registers and statistical agencies. Exposure was defined as filled prescriptions for high-dose folic acid, considered time-varyingly. The primary outcome was the first malignant cancer diagnosis. Hazard ratios (HRs) of cancer after high-dose folic acid exposure were estimated using Cox proportional hazard models with 95% confidence intervals (CIs), adjusted for confounders including antiseizure medication (ASM) use, and stratified by maternal epilepsy diagnosis. A 6-month time lag was applied, as cancer is unlikely to develop immediately. Results: With up to 21 years of follow-up, we identified 1 465 785 women who gave birth, including 64 485 (4.4%) exposed to high-dose folic acid. In the exposed group, 755 cancer cases were observed (208 per 100 000 person-years, 95% CI = 193.8-223.5), compared with 18 702 cases in the unexposed group (164 per 100 000 person-years, 95% CI = 161.5-166.2), yielding a 20% increased cancer risk overall (adjusted HR = 1.2, 95% CI = 1.1-1.2). This risk was attenuated after the 6-month lag analysis (adjusted HR = 1.1, 95% CI = 1.04-1.2). The risk for non-Hodgkin lymphoma was increased in all analyses (n = 28, adjusted HR = 2.0, 95% CI = 1.3-2.9). The association between high-dose folic acid use and overall cancer risk was similar in those with epilepsy regardless of ASM use (adjusted HR = 1.3, 95% CI = 1.0-1.8). Significance: High-dose folic acid use was associated with increased overall cancer risk in women who have given birth, with a consistent association with non-Hodgkin lymphoma, including those with epilepsy, regardless of ASM use.

Commentary

Folic acid supplementation for people with childbearing potential and epilepsy has been recommended via American Academy of Neurology (AAN) practice parameters for nearly 3 decades.^1-3 The American College of Obstetrics and Gynecology and international groups including the United Kingdom National Institute for Health and Care Excellence (NICE) recommend high dose folic acid supplementation for childbearing people with epilepsy.^4,5 Further, recent evidence in epilepsy demonstrated periconception folic acid likely improves global IQ of children ⁶ and linked lack of preconception folic acid with increased odds of autistic traits. ⁷ This data supported a new 2024 AAN level A guideline recommendation to prescribe folic acid periconceptually and during pregnancy among people with epilepsy, to potentially improve neurodevelopmental outcomes. ¹

While the evidence of potential neurodevelopmental benefits of folic acid for people of childbearing potential with epilepsy has developed in recent years, data on folic acid dose and benefits remains limited. The Norwegian Mother and Child Cohort Study may suggest a benefit of higher dose, as lower plasma folic acid concentrations during gestational weeks 17-19 were associated with greater autistic traits in children at age 3 years. ⁷ Evidence exists for benefit of high dose (4 mg) folic acid in preventing neural tube defects in the general population, ⁸ especially in settings without folic acid fortification, ¹ and guideline committees have highlighted a potential increased risk for neural tube defects among people with epilepsy as a justification for recommending high dose folic acid (4-5 mg daily) in the absence of specific evidence for benefit in epilepsy.^4,5 However, the recent AAN guideline systematic review found folic acid supplementation in epilepsy likely does not reduce major congenital malformations, and there was insufficient evidence to determine effects of folic acid dose on malformations or neurodevelopmental outcomes. ¹

How much folic acid to prescribe childbearing people with epilepsy is an important dilemma for clinicians, and while the recent AAN guideline emphasized the continued lack of sufficient data for folic acid dosing recommendations, recent data on harms is raising concern about potential risks of high dose folic acid that clinicians will need to weigh against the benefits. New data on potential harm of high dose folic acid includes the 2022 population-based analysis of cancer risk among children born to mothers with epilepsy by Vegrim et al that found a hazard ratio of 2.7 (95% confidence interval, CI 1.2-6.3) for childhood cancer among children born to mothers with epilepsy prescribed 1-5 mg folic acid. ⁹ Some other data in older, non-epilepsy patients suggested a possible link between high dose folic acid and cancer risk, but risk in childbearing individuals with epilepsy had not been assessed.

To address this gap, Vegrim et al conducted a new population-based analysis examining cancer risk among women who have given birth. ¹⁰ This medical register-based cohort study included all women with a first birth (at 22 weeks or more of gestation) during 11-21 year timeframes in Denmark, Norway, and Sweden and observed for a first cancer diagnosis during follow-up. High-dose folic acid was defined as filling prescription(s) for 1 mg or 5 mg formulations of folic acid (as recorded in national prescription registers) from 2 years before 22 weeks of a first pregnancy to end of follow-up. Overall, the sample included 1,465,785 women, with 64,485 having received a high dose folic acid prescription. Median follow-up was 5 years for those who received high dose folic acid and 7 years for those who did not, and longest follow-up was 21 years among the cohort.

The main finding was a 20% increase in cancer risk during follow-up among those who had filled high dose folic acid prescriptions (adjusted hazard ratio 1.2, 95% CI 1.1-1.2), which was slightly lower but still significant when the first 6 months of follow-up were removed (as cancer would not be expected to develop that soon after exposure; hazard ratio 1.1, 95% CI 1.04-1.2). Specific cancers were also analyzed using rigorous techniques to account for multiple comparisons, and the findings demonstrated a statistically significant 2-fold increased risk of non-Hodgkin lymphoma but no significant risk for other cancers. Cumulative overall dosing of prescription folic acid during follow-up was analyzed, with quartiles of cumulative exposure examined separately for cancer risk. No dose-response relationship was identified, with adjusted hazard ratios of 1.3 for the 2 lowest quartiles (up to 299 mg and 300-999 mg), yet only 1.1 and 1.0 for 1000-1999 mg and ≥2000 mg respectively. The analysis did not explore for associations separately among individuals prescribed the 1 mg dosage form versus 5 mg dosage form.

While the population-based design, large sample, completeness of prescription and health records available in these national registers, and analysis approach are study strengths, there are multiple limitations. These include the relatively short-term median follow-up, which at only 5-7 years may be too short to fully capture outcomes of cancer diagnoses, reliance on prescription dosages and dispensing records without examining over-the-counter forms or more direct measures of actual intake, and lack of daily dose-based analysis (eg, 1 mg vs 5 mg, considering that while 5 mg is well accepted to be “high dose,” many clinicians would not necessarily consider 1 mg to be “high dose”). Further, the lack of a dose-response association in the cumulative dose analysis weakens the potential case for a causal association of high dose folic acid consumption and cancer risk.

These limitations and the importance of the clinical dilemma regarding folic acid dose selection highlight the critical need for future research in this area, which has been a documented priority by AAN guideline panels on this topic for decades.^1,2 Future, longer-term research to determine if these findings are replicable, estimate risk, and more closely examine discrete daily dosage is needed, including data on 0.4 mg, 0.8 mg, and more data on 1 mg daily dosing. Further, data from countries with folic acid fortification is needed, as the Scandinavian countries investigated by Vegrim et al do not have folic acid fortification. Ideally, future work would include high-quality randomized designs examining benefits and short-term risks of different folic acid doses in epilepsy, paired with additional high quality prospective long-term studies to further evaluate for potential long-term harms.

In the meanwhile, clinicians should now consider whether to modify their prescribing practice in response to Vegrim et al's findings of a small increase in cancer risk with folic acid prescribed at 1 or 5 mg, ¹⁰ and the similar association with cancer risk in offspring published in 2022. ⁹ Regardless of the individual clinician's decision, these new data certainly weaken a prior common rationalization for prescribing 4-5 mg folic acid daily (because there is no harm and there might be a benefit). In response, clinicians may become more reluctant to prescribe high folic acid doses, especially to individuals not taking antiseizure medications with high risk for poor neurodevelopmental outcomes, such as valproic acid.