Prenatal Sterol-Inhibiting Drugs Linked to Increased Autism Risk

Cholesterol Metabolism and the Developing Fetal Brain

The mevalonate and cholesterol biosynthetic pathways are fundamental to the structural and functional development of the central nervous system, serving as essential components for synaptogenesis (the formation of synapses between neurons) and myelin formation [1]. While genetic mutations in these pathways are known to cause severe neurologic and psychiatric phenotypes, there is growing concern regarding the impact of exogenous chemical interference during critical gestational windows [2]. Sterol metabolism is increasingly recognized as a biological factor in the pathogenesis of neurodevelopmental conditions, including autism spectrum disorders [3]. Despite this, many common medications used to treat cardiovascular and psychiatric conditions act by inhibiting post-lanosterol biosynthesis (the enzymatic steps that convert lanosterol into cholesterol), often without established safety guidelines for use during pregnancy [2]. Emerging metabolomic evidence from animal models further suggests that disrupted steroid hormone and cholesterol pathways in the hippocampus, a region critical for memory and social behavior, correlate with autistic-like behavioral traits [4]. A large-scale clinical analysis now provides critical data on how these common pharmacological exposures may influence neurodevelopmental outcomes in humans.

Large-Scale Cohort Analysis of 6 Million Births

To evaluate the potential neurodevelopmental impact of prenatal medication exposure, the researchers conducted a retrospective cohort study using the Epic Cosmos database, a platform that aggregates linked child and maternal health records from diverse clinical settings. The study population included a total of 6,135,213 children with linked maternal health records, providing sufficient statistical power to detect subtle associations between gestational drug exposure and subsequent clinical outcomes. The cohort comprised births occurring between 2014 and 2023, with longitudinal follow-up data extending through December 2025 to ensure adequate time for the emergence and documentation of developmental diagnoses. Within this population of over six million children, the researchers identified 234,971 individuals with a diagnosis of autism spectrum disorder, establishing a baseline incidence of 3.8% for the total cohort. Among the children who received an autism spectrum disorder diagnosis, 35,152 (15.0%) were born to mothers who had been prescribed at least one sterol biosynthesis inhibiting medication during pregnancy. By utilizing Cox proportional hazard modeling (a statistical method used to estimate the risk of a specific event occurring over time while accounting for various factors), the authors were able to adjust for potential confounders and quantify the specific risk associated with these common pharmacological agents.

Identifying High-Risk Pharmacological Exposures

The researchers focused their investigation on fifteen sterol biosynthesis inhibiting medications, a class of drugs that interfere with the complex enzymatic pathway responsible for producing cholesterol and other essential sterols. Because cholesterol is a critical structural component of cell membranes and a precursor to signaling molecules in the developing brain, the study analyzed the relationship between prenatal exposure to these agents and the subsequent incidence of autism spectrum disorder in offspring. The medications evaluated in this cohort included a diverse range of commonly prescribed agents across multiple therapeutic classes: the antipsychotics aripiprazole and haloperidol; the antidepressants and anxiolytics bupropion, buspirone, fluoxetine, sertraline, and trazodone; the lipid-lowering statins atorvastatin, pravastatin, rosuvastatin, and simvastatin; and the beta-blockers metoprolol, nebivolol, and propranolol. To quantify the association between these pharmacological exposures and neurodevelopmental outcomes, the authors employed Cox proportional hazard modeling. This statistical method is used to determine the risk of a specific outcome, such as an autism diagnosis, occurring over time while accounting for the duration of follow-up and various maternal health factors. This analytical approach allowed the researchers to adjust for potential confounders, such as maternal age or pre-existing conditions, and calculate the specific hazard ratios associated with maternal use of these sterol-inhibiting drugs. By examining these specific medications, which collectively account for over 400 million prescriptions annually in the United States, the study aimed to identify whether the disruption of fetal sterol synthesis represents a modifiable risk factor for autism spectrum disorder.

Quantifying the Dose-Dependent Risk

The researchers determined that maternal use of these medications significantly correlates with neurodevelopmental outcomes in offspring. After adjusting for potential confounders, exposure to at least one sterol biosynthesis inhibiting medication during pregnancy was associated with a 1.47-fold (95% CI 1.45-1.49) increased risk of autism spectrum disorder. This finding suggests a substantial association between prenatal sterol disruption and subsequent diagnosis. To put this into clinical perspective, the study identified 234,971 children with an autism diagnosis within the cohort. Among these children, 35,152 (15.0%) had mothers who were prescribed at least one sterol biosynthesis inhibiting medication during pregnancy, indicating that a notable portion of the clinical population may have had prenatal exposure to these agents. The data also revealed a clear dose-response relationship, where the risk of a diagnosis increased with the number of different sterol-inhibiting agents used concurrently. Each additional sterol biosynthesis inhibiting medication co-prescribed was associated with a 1.33 (95% CI 1.32-1.34) times increased risk of autism spectrum disorder. This cumulative effect was most pronounced in cases of polypharmacy, where multiple medications from the fifteen identified classes were used. Specifically, the co-prescription of 4 or more sterol biosynthesis inhibiting medications was associated with a 2.33-fold risk of autism spectrum disorder. For clinicians, these figures highlight that while single-agent exposure carries a measurable risk, the hazard increases significantly as the pharmacological burden on the fetal cholesterol biosynthetic pathway grows, potentially impacting the structural integrity of developing neural circuits.

Clinical Implications of Rising Prescription Trends

The clinical relevance of these findings is underscored by a sharp upward trend in the use of these agents among the pregnant population. According to the longitudinal data from the Epic Cosmos database, the utilization of sterol biosynthesis inhibiting medications by pregnant women increased from 4.6% in 2014 to 16.8% in 2023. This nearly fourfold rise in prenatal exposure suggests that a growing proportion of pregnancies are now managed with medications that may interfere with essential fetal cholesterol synthesis. Given that these fifteen specific medications, which include widely used antidepressants, statins, and beta-blockers, account for over 400 million prescriptions annually in the United States, the potential for unintended neurodevelopmental impact is a significant public health consideration. In light of the dose-dependent risk identified in this cohort of over 6 million births, the researchers conclude that sterol biosynthesis inhibiting medications may be potentially harmful to the developing fetus. The authors recommend that clinicians carefully consider these neurodevelopmental risks and conduct a thorough risk-benefit analysis before prescribing these medications during pregnancy. While managing maternal conditions such as depression, hypertension, or hyperlipidemia remains a clinical priority, these data suggest that the cumulative pharmacological burden on the cholesterol biosynthetic pathway should be a factor in prenatal treatment planning to mitigate the risk of autism spectrum disorder in offspring, raising the prospect that future guidelines may need to account for the total sterol-inhibiting load of a patient's regimen.

References

1. Segatto M, Tonini C, Pfrieger FW, Trezza V, Pallottini V. Loss of Mevalonate/Cholesterol Homeostasis in the Brain: A Focus on Autism Spectrum Disorder and Rett Syndrome. International Journal of Molecular Sciences. 2019. doi:10.3390/ijms20133317

2. Peeples ES, Mirnics K, Korade Ž. Chemical Inhibition of Sterol Biosynthesis. Biomolecules. 2024. doi:10.3390/biom14040410

3. Aitken KJ. Intersubjectivity, Affective Neuroscience, and the Neurobiology of Autistic Spectrum Disorders: A systematic review. The Keio Journal of Medicine. 2008. doi:10.2302/kjm.57.15

4. Toczyłowska B, Ziemińska E, Senator P, Łazarewicz JW. Hippocampal Metabolite Profiles in Two Rat Models of Autism: NMR-Based Metabolomics Studies. Molecular Neurobiology. 2020. doi:10.1007/s12035-020-01935-0