Topiramate’s appetite effect points to a specific genetic target in binge eating behavior
Topiramate has long carried a double identity in clinical practice. It is an antiseizure drug and migraine preventive, yet it is also widely used for appetite suppression and weight loss, a pharmacologic profile that raises a difficult question in eating-disorder medicine: does the same biology that reduces intake also alter risk for anorexia nervosa or binge-eating behavior?
A drug-target Mendelian randomization analysis now points to a specific molecular pathway rather than a broad class effect. Genetically predicted inhibition of CA4 expression in blood was associated with lower binge-eating risk, while no robust evidence emerged for anorexia nervosa or for eight exploratory psychiatric outcomes. The pattern is narrow, but clinically meaningful. It suggests that one of topiramate’s biologic actions may intersect with circuits governing compulsive overeating without translating into a detectable causal signal for restrictive eating.
Genetic proxy signals were built around topiramate’s antiepileptic action
The analysis used summary-level data and focused on drug-target Mendelian randomization, a method designed to estimate whether genetically influenced changes in a drug target behave like the medication itself. Five genetic instruments consistent with topiramate’s antiepileptic action were validated: CA4 and CACNB3 expression in the brain, and CA4, GRIK1, and GRIK3 expression in blood. Those targets matter because topiramate does not act through a single receptor. Its pharmacology spans ion channels and glutamatergic signaling, which makes a target-level approach more informative than asking whether “topiramate exposure” as a whole influences complex behavior.
The outcome datasets were large. For anorexia nervosa, the analysis included 24,223 cases and 1,243,971 controls. For binge-eating behavior, there were 39,279 cases and 1,227,436 controls. The investigators also explored attention-deficit/hyperactivity disorder, anxiety disorders, alcohol use disorder, bipolar disorder, major depressive disorder, panic disorder, post-traumatic stress disorder, and schizophrenia. That breadth matters because weight-related symptoms often overlap with mood, anxiety, impulsivity, and substance-use phenotypes, and any credible signal has to survive that background noise.
CA4 inhibition in blood tracked with lower binge-eating risk
The clearest result came from CA4 in blood. After multiple-testing correction, genetically predicted inhibition of CA4 expression was associated with decreased binge-eating risk, with an odds ratio of 0.867 and a 95% confidence interval of 0.792 to 0.950
(p=0.002). The association remained significant in SMR-multi (p=0.0049), which strengthens confidence that the signal was not a statistical artifact of a single model specification.
A moderate colocalization result reinforced the same direction. The probability of a shared causal variant underlying CA4 expression and binge eating reached PPH4 = 0.58. That is not absolute proof of causality, but it is stronger than a vague association alone. In practical terms, the data fit a model in which changing the expression of a carbonic anhydrase target helps shift vulnerability toward binge-eating behavior. Carbonic anhydrases influence pH regulation and neuronal excitability, and those processes can shape reward signaling, satiety processing, and inhibitory control in ways that may be relevant to overeating phenotypes.
One result stood out for its restraint as much as for its signal.
The effect appeared in binge eating, not in anorexia nervosa.
No robust causal signal emerged for anorexia nervosa or the psychiatric panel
Despite topiramate’s real-world association with weight loss, the analysis found no robust evidence linking the drug-target proxies to anorexia nervosa. That distinction is important because anorexia nervosa and binge-eating behavior are not mirror images; they differ in reward processing, compulsivity, body-image disturbance, and metabolic adaptation. A medication can plausibly blunt compulsive intake without pushing the biology toward restrictive pathology, and this study supports that separation rather than collapsing the two disorders into a single appetite axis.
The exploratory psychiatric outcomes were also negative. No robust evidence was identified for attention-deficit/hyperactivity disorder, anxiety disorders, alcohol use disorder, bipolar disorder, major depressive disorder, panic disorder, post-traumatic stress disorder, or schizophrenia. Those null findings reduce the chance that the binge-eating association is merely a downstream reflection of a broad psychiatric liability signal. They also narrow the mechanistic frame: the most defensible interpretation is a specific interaction between topiramate-linked target biology and eating behavior.
Why this matters for appetite suppression and bedside interpretation
Topiramate’s clinical reputation rests on observed appetite suppression and weight loss, but mechanistic explanations have often remained diffuse. This preprint offers a more precise hypothesis: one of the drug’s targets, CA4, may help mediate effects on binge eating. The signal in blood may also reflect systemic biology that maps onto central nervous system pathways, or it may capture a biomarker-like proxy for the processes that matter in behavior. Either way, the result invites a more careful look at how topiramate changes eating patterns in patients whose primary problem is loss of control over intake rather than excessive caloric restriction.
The finding does not justify changing prescribing habits on its own. This is still preprint evidence, and Mendelian randomization estimates are only as good as the genetic instruments and assumptions behind them. Drug-target analyses can sharpen causal inference, but they do not substitute for clinical trials, dose-response data, or direct mechanistic experiments in humans. The practical next step is functional work that tests whether CA4-related biology influences satiety, reward, or inhibitory control in relevant neural and peripheral tissues.
For now, the study leaves clinicians with a narrower question than before. Topiramate’s appetite-related effects may not be a single blunt phenomenon; they may reflect target-specific biology that is more tightly linked to binge-eating behavior than to anorexia nervosa. That framing should guide the next round of translational studies, especially in patients whose eating pathology is driven by recurrent loss of control rather than by restriction.