Low IGF-1 Levels Linked to Higher Epilepsy Risk in UK Biobank

Lower concentrations of insulin-like growth factor 1 (IGF-1) in the blood are associated with a higher risk of developing epilepsy, according to a large prospective analysis of the UK Biobank published in Acta Epileptologica. The study followed 466,857 participants who had baseline serum IGF-1 measurements, and over more than 15 years of follow-up, 4,532 of them developed incident epilepsy. In the primary statistical model, each 1 nmol/L increase in serum IGF-1 was associated with a 2.2% lower hazard of incident epilepsy, with a hazard ratio of 0.978 (95% confidence interval 0.973–0.984). The authors emphasize that this is an observational association at the population level, not evidence that IGF-1 is a causal factor or a therapeutic target.

The research team, led by Jiahui Guo and colleagues, drew on the UK Biobank, a population-based resource that recruited more than 500,000 adults aged 40 to 69 years between 2006 and 2010 across 22 assessment centres in England, Scotland, and Wales. Participants completed baseline questionnaires, physical assessments, and biospecimen collection, with subsequent linkage to national health records. Serum IGF-1 was measured from blood samples collected at recruitment using a centralized chemiluminescent immunoassay platform, with standardized sample processing and external quality assurance procedures. Epilepsy diagnoses were identified from hospital records using ICD-10 codes G40–G41, aligned with the contemporary clinical framework of the International League Against Epilepsy, and follow-up accrued from recruitment until incident epilepsy, death, loss to follow-up, or October 30, 2023, whichever came first.

The cohort comprised three groups: 2,225 individuals with epilepsy at baseline, 4,532 who developed epilepsy during follow-up, and 460,100 who remained free of epilepsy throughout. Mean serum IGF-1 differed significantly across the groups, at 19.57 ± 6.43 nmol/L in the baseline epilepsy group, 20.38 ± 5.91 nmol/L in the incident epilepsy group, and 21.42 ± 5.69 nmol/L in the epilepsy-free group, with all pairwise differences significant after multiplicity correction.

Kaplan–Meier analysis based on the median baseline IGF-1 level showed significantly better epilepsy-free survival among participants with higher IGF-1 (log-rank P < 0.0001). In prespecified Cox proportional hazards models adjusted for age, sex, body mass index, smoking, alcohol use, sleep hours per day, and loneliness, the inverse association persisted. When IGF-1 was standardized, each 1 standard deviation increase was associated with an approximately 11–12% lower hazard (HR = 0.882, 95% CI 0.854–0.912). In quartile-based analyses, participants in the second, third, and fourth quartiles had lower risks than those in the lowest quartile, with hazard ratios of 0.783, 0.737, and 0.755 respectively, all statistically significant. Absolute risk estimates were directionally concordant: the incidence rate per 1,000 person-years fell from 0.937 in the lowest quartile to 0.604 in the highest, and the estimated 15-year cumulative incidence of epilepsy declined from 1.382% in the lowest quartile to 0.896% in the highest.

A striking feature of the results is the shape of the dose–response relationship. Restricted cubic spline analysis, with five knots placed across the IGF-1 distribution, revealed a significant L-shaped nonlinear association (P for nonlinearity < 0.001), with epilepsy risk rising steeply at lower IGF-1 concentrations and an apparent plateau above roughly 20 nmol/L. The quartile results likewise did not decline strictly monotonically from the third to the fourth quartile. The authors caution that the apparent inflection point near 20 nmol/L should not be interpreted as a clinically validated threshold for intervention, but rather as a feature of the observed exposure–response pattern in this cohort. The pattern suggests that lower circulating IGF-1 may mark a biological state in which epilepsy susceptibility rises more sharply, while at higher concentrations the apparent protective association levels off.

The investigators took several steps to address the possibility of reverse causation, in which preclinical disease might lower IGF-1 before diagnosis. Lagged Cox analyses excluding epilepsy events occurring within the first 2 and 5 years of follow-up preserved the inverse association, with hazard ratios per 1 nmol/L of 0.980 and 0.985 respectively in the primary model, and similar results in the sensitivity model, which additionally adjusted for ethnicity, education, deprivation, physical activity, and working hours in a reduced complete-case sample (HR = 0.979, 95% CI 0.968–0.990). Sensitivity analyses replacing binary alcohol coding with detailed drinking-frequency categories, and excluding daily or almost daily drinkers, left the IGF-1 association materially unchanged. Subgroup analyses showed the association was broadly consistent across sex, while an age interaction observed in the primary model—stronger among participants under 60—was not replicated in the sensitivity model and should be interpreted cautiously.

The cognitive findings were notably weaker than the epilepsy-risk results. In cross-sectional analyses of the 2,225 participants with epilepsy at baseline, higher IGF-1 was nominally associated with faster reaction time (β = –1.588, P = 0.027) and a more favorable prospective memory category (OR = 1.033, P = 0.031) in fully adjusted models, but neither association survived Benjamini–Hochberg false discovery rate correction across the five cognitive domains tested, with both yielding q values of 0.078. No significant fully adjusted associations were observed for visuospatial memory, numeric memory, or fluid intelligence. Exploratory longitudinal cross-lagged analyses in the epilepsy subgroup included only 177 participants with complete data for IGF-1 and reaction time and found no significant bidirectional associations, limiting any inference about cognitive trajectories in epilepsy.

In the full cohort, however, higher baseline IGF-1 was associated with better subsequent performance in several cognitive domains: reaction time (β = –0.049, P < 0.001), visuospatial memory (β = –0.030, P < 0.001), and fluid intelligence (β = 0.032, P = 0.010), with no significant association for prospective memory. Baseline cognitive performance did not significantly predict follow-up IGF-1 in any available domain, and analyses involving numeric memory were not feasible because of insufficient longitudinal data. The authors stress that these full-cohort findings are not epilepsy-specific and should be read as population-level, hypothesis-generating observations rather than evidence that IGF-1 influences cognitive outcomes in people with epilepsy.

Biologically, the result is plausible. IGF-1 is a pleiotropic peptide hormone produced predominantly in the liver under growth hormone regulation but also expressed in the brain, and its signaling has been implicated in neurodevelopment, neuronal survival, synaptic plasticity, myelination, and neuroimmune regulation. Recent work showing that microglia regulate prenatal human GABAergic neurogenesis through IGF-1 signaling is particularly relevant to epilepsy, in which inhibitory–excitatory imbalance and altered network maturation are central themes. Nevertheless, the authors are explicit that the study does not establish reduced circulating IGF-1 as a direct cause of epileptogenesis, nor does it identify the specific pathways involved, especially since circulating IGF-1 is itself influenced by age, metabolic state, endocrine physiology, and systemic illness.

Several limitations temper the conclusions. Epilepsy was ascertained at the population level through health records rather than detailed syndrome-specific clinical phenotyping, so epilepsy etiology, seizure frequency, EEG findings, imaging substrates, and antiseizure medication exposure could not be accounted for. The models did not include detailed vascular, metabolic, psychiatric, neurological, or medication-related covariates, leaving residual confounding possible, and serum IGF-1 was measured at a single baseline time point. The cohort was drawn predominantly from middle-aged adults of European ancestry, limiting generalizability. The authors also note that the counterintuitive apparent protective association of alcohol with epilepsy most likely reflects residual confounding or healthy-user effects rather than a genuine protective effect. Overall, they conclude that serum IGF-1 should be regarded as a robust epidemiological correlate of incident epilepsy and a candidate biomarker warranting further investigation, with future studies requiring richer clinical phenotyping, repeated biomarker measurements, and adequately powered epilepsy-specific longitudinal designs to clarify any clinical utility.

Subject of Research: The association between serum insulin-like growth factor-1 and epilepsy risk and cognitive function in the UK Biobank

Article Title: Serum insulin-like growth factor-1 and its association with epilepsy risk and cognitive function

Article References: Guo, J., Qiu, Z., Chen, B., Li, H., Su, P., Yuan, Z., Wu, D., & Fang, J. (2026). Serum insulin-like growth factor-1 and its association with epilepsy risk and cognitive function. Acta Epileptologica, 8(1), Article 36. https://doi.org/10.1186/s42494-026-00274-2

Image Credits: AI Generated

DOI: 10.1186/s42494-026-00274-2

Keywords: IGF-1, epilepsy, UK Biobank, cohort study, seizure risk, cognitive function, biomarker, neurology, epidemiology, hazard ratio, prospective study, brain health