GLP-1 for Binge Eating Disorder: Vyvanse Comparison and Stacking Evidence

Binge eating disorder (BED) is the most common eating disorder in the United States, affecting roughly 2.8% of adults across the lifespan (Hudson 2007, NCS-R^[4]). About half of adults with BED also live with obesity, which is why the question of GLP-1 receptor agonists for BED keeps coming up in primary care and obesity-medicine clinics. The honest answer is that lisdexamfetamine (Vyvanse) is still the only FDA-approved drug for moderate-to-severe BED — on the strength of the McElroy 2015 RCT^[1] and the Hudson 2017 randomized-withdrawal maintenance trial^[2]— while GLP-1s have a growing but still systematic-review level evidence base (Aoun 2024^[5], Balantekin 2024^[6]). This article walks through the BED diagnostic criteria, the Vyvanse trial data, what the GLP-1 BED literature actually shows, and the practical psychiatric pathway when both BED and obesity are on the chart.

The honest summary

• Vyvanse is the only FDA-approved BED drug. The McElroy 2015 JAMA Psychiatry RCT^[1] randomized adults with moderate-to-severe BED to lisdexamfetamine 50 mg, 70 mg, or placebo; the 70 mg arm reduced weekly binge days by roughly 4 vs ~3 on placebo and produced a meaningfully higher rate of 4-week cessation. The Hudson 2017 randomized-withdrawal trial^[2] showed a hazard ratio for relapse of roughly 0.30 on continued LDX vs placebo over 26 weeks.
• GLP-1s reduce binge episodes per published case series and systematic reviews, but no FDA-registration RCT exists for BED. Aoun 2024^[5] systematically reviewed the GLP-1 + binge-eating literature and found consistent reductions in binge frequency across small open trials and case series; Carminati 2026^[8] reached the same conclusion for semaglutide specifically.
• CBT is the psychological gold standard. The Hilbert 2019 meta-analysis^[9] of psychological and medical treatments for BED found cognitive behavioral therapy produced the largest and most durable effects on binge abstinence; pharmacotherapy works on binge frequency and weight but is generally not as durable without an ongoing behavioral component.
• Stacking Vyvanse + a GLP-1 is plausible for BED-plus-obesity but is off-label, requires cardiovascular and hydration monitoring, and is best coordinated between a psychiatrist and an obesity-medicine clinician (Himmerich 2024^[7]).

What BED actually is

DSM-5 binge eating disorder requires recurrent episodes of eating an objectively large amount of food in a discrete period with a sense of loss of control, occurring on average at least once a week for three months, and associated with marked distress. Critically, BED differs from bulimia nervosa in the absence of regular compensatory behaviors (purging, restricting, excessive exercise). The Hudson 2007 NCS-R survey^[4]estimated 12-month US prevalence at 1.2% and lifetime prevalence at 2.8% — making it more common than anorexia nervosa and bulimia nervosa combined. About half of adults with BED meet criteria for obesity at presentation, and BED predicts worse weight-loss outcomes for both diet-based and surgical interventions when untreated.

The Vyvanse evidence: McElroy 2015 and Hudson 2017

Lisdexamfetamine dimesylate (Vyvanse) is a prodrug stimulant approved by the FDA in 2015 for moderate-to-severe BED in adults at 50–70 mg/day. The pivotal evidence comes from a pair of trials.

McElroy 2015 (JAMA Psychiatry)^[1] randomized 514 adults with moderate-to-severe BED to LDX 30, 50, or 70 mg/day vs placebo for 11 weeks. The 50 and 70 mg arms separated from placebo on the primary endpoint — change in weekly binge days — with the 70 mg arm producing roughly 4 fewer binge days per week vs roughly 3 on placebo. The 4-week cessation rate (no binge days for 4 consecutive weeks at endpoint) was about 36% on 50 mg, 50% on 70 mg, and 21% on placebo. The 30 mg arm did not separate meaningfully from placebo, which is why the FDA label starts at 30 mg as a titration dose and targets 50–70 mg therapeutically.

Hudson 2017 (JAMA Psychiatry)^[2] used a randomized-withdrawal design: 418 participants completed a 12-week open-label LDX phase, and the responders were re-randomized to continue LDX or switch to placebo for a 26-week double-blind phase. Relapse was defined as ≥2 binge days per week for two consecutive weeks plus a 2-point increase in CGI-Severity. The hazard ratio for relapse on continued LDX vs placebo was roughly 0.30 (about a 70% risk reduction), with Kaplan-Meier relapse rates of about 3.7% on LDX vs 32.1% on placebo at week 26. This is the trial that anchors the practical recommendation to continue LDX rather than stopping after acute response.

Gasior 2017^[3] added a 12-month open-label safety/tolerability extension confirming the safety signal: the dominant adverse events were dry mouth, headache, insomnia, and modest increases in heart rate (~6–7 bpm) and systolic blood pressure (~2–3 mmHg). The CV signal is why baseline ECG plus quarterly HR/BP monitoring is the standard.

The GLP-1 evidence: smaller, but consistent

The GLP-1 receptor agonist literature for BED is at an earlier stage. There is no FDA-registration RCT for any GLP-1 in BED, and the published trials are small, open-label, or retrospective.

Aoun 2024 (J Clin Transl Endocrinol)^[5] systematically reviewed the GLP-1 + binge-eating literature across BED and bulimia nervosa. The included studies used liraglutide, exenatide, dulaglutide, and semaglutide, with consistent reductions in binge frequency, food-craving scores, and body weight in the BED cohorts. The reviewers explicitly flag the absence of any large randomized BED trial. Balantekin 2024 (J Endocrinology)^[6]reviewed the mechanistic pathway: GLP-1 receptors on the nucleus tractus solitarius and the ventral tegmental area suppress reward-driven feeding in animal models, with translational support from human functional-imaging studies showing reduced striatal response to highly palatable food cues on liraglutide and semaglutide. Carminati 2026 (Int Clin Psychopharmacol)^[8] systematically reviewed semaglutide's psychiatric effects and concluded the binge-reduction signal is consistent but the evidence base is still grade C; it does not yet support a label change.

Himmerich 2024 (CNS Drugs)^[7]reviewed pharmacological treatment of BED across LDX, topiramate, fluoxetine, naltrexone-bupropion, and GLP-1 agonists. The author group's practical conclusion: LDX remains first-line on the strength of the registration trials; GLP-1s are an emerging option, particularly when obesity is the dominant clinical concern, and they may be reasonable to layer on top of LDX or to substitute when stimulant exposure is undesirable (cardiovascular comorbidity, history of substance use disorder, pregnancy planning).

Why GLP-1s plausibly work for BED: the mechanism

Binge eating is, mechanistically, a problem of reward-driven food intake exceeding the homeostatic satiety signal. The GLP-1 receptor agonist mechanism touches both sides of that equation. Centrally, GLP-1 receptors in the arcuate nucleus and the nucleus tractus solitarius amplify post-meal satiety; peripherally, they slow gastric emptying so a normal portion produces a longer-lasting fullness signal. The reward-pathway effect is the more interesting one for BED: GLP-1 receptor agonism in the ventral tegmental area and nucleus accumbens reduces the dopaminergic response to highly palatable food cues (Balantekin 2024^[6]). That is why patients describe “food noise” quieting on GLP-1s — a phenomenological match to what BED patients describe as the loss-of-control intrusive thoughts that precede a binge.

Vyvanse works through a different lever. Lisdexamfetamine is a prodrug that releases d-amphetamine slowly across the day; amphetamine increases dopamine and norepinephrine in the prefrontal cortex and striatum. The behavioral correlate is reduced reward valuation for food cues plus improved impulse-control resources in the prefrontal circuits that normally veto a binge. The two mechanisms are non-redundant, which is the theoretical basis for stacking.

Magnitude: binge-frequency reduction by intervention

The practical psychiatric pathway

For a patient who presents with both BED and obesity, the sequencing matters. The pathway we see most consistently applied by psychiatry-trained obesity-medicine clinicians is:

1. Confirm the BED diagnosis. DSM-5 criteria with a structured tool such as the Eating Disorder Examination-Questionnaire (EDE-Q) or the Binge Eating Scale. Rule out bulimia nervosa (compensatory behaviors) and night-eating syndrome.
2. Start CBT. Manualized CBT-Enhanced (CBT-E) or the Fairburn protocol is the gold-standard psychological treatment per Hilbert 2019^[9]. Sixteen to twenty sessions is the typical course; group formats and self-help workbooks are second-line.
3. For moderate-to-severe BED: add lisdexamfetamine 50–70 mg/day, titrated from 30 mg over 2–4 weeks. Baseline ECG; HR and BP at each follow-up. Continue indefinitely if responding (Hudson 2017^[2] showed that stopping LDX produces high relapse rates).
4. If obesity is the dominant concern after binge control: add a GLP-1 (semaglutide 2.4 mg or tirzepatide titrated to 10–15 mg) under obesity-medicine supervision. Monitor combined HR/BP, since stimulant + GLP-1 dehydration risk is additive in the early weeks.
5. Alternative path: If stimulant exposure is undesirable (cardiovascular comorbidity, substance-use history, pregnancy planning), GLP-1 monotherapy plus CBT is a reasonable off-label first-line approach — with the caveat that the BED evidence base is smaller than for LDX.
6. Monitor: weekly binge-day count, weight, resting HR and BP, mood (PHQ-9 every 3 months), and electrolytes/hydration in the first 8 weeks of any new combination.

Contraindications and edge cases

• Active anorexia nervosa: GLP-1 receptor agonists are contraindicated. Suppressing appetite in a low-weight patient with restrictive eating is harmful.
• Active bulimia nervosa: generally contraindicated until the purging cycle is in remission; GLP-1-induced nausea can amplify purging behaviors.
• Bariatric surgery + active BED: most bariatric programs require demonstrated binge-control for 6–12 months pre-op. Post-op binge eating predicts weight regain.
• Pregnancy: stop both lisdexamfetamine and GLP-1 (per current FDA pregnancy guidance); continue CBT.
• Compulsive overeating without DSM-5 BED: GLP-1 monotherapy is reasonable and well-tolerated; LDX is off-label and not justified.
• Cardiovascular disease: stimulant contraindicated with structural heart disease, arrhythmia, or uncontrolled hypertension; GLP-1 monotherapy is the appropriate route.

Cost and access

Branded Vyvanse runs roughly $400/month at most US pharmacies. Generic lisdexamfetamine (launched 2023) is available at roughly $200/month and is widely covered by commercial insurance and most state Medicaid plans for the BED indication. GLP-1 coverage for the obesity indication is highly plan-dependent; the BED indication does not currently unlock GLP-1 coverage on any major formulary. Patients combining LDX and a GLP-1 typically pay the stimulant out-of-pocket via insurance and the GLP-1 either through insurance (when obesity-coded) or through a compounded telehealth route.

Related research and tools

• GLP-1 + ADHD stimulants (Adderall / Vyvanse) — the cardiovascular and hydration monitoring detail when stacking stimulants on a GLP-1
• GLP-1 non-responder phenotypes — how to recognize a reward-driven non-responder and when to add a stimulant or switch class
• GLP-1 cognitive effects and brain fog — what the central GLP-1 signaling literature actually says about attention and executive function
• GLP-1 anhedonia and emotional blunting — the reward-pathway effect that helps with BED can also blunt non-food rewards
• Food noise, cortisol, and GLP-1 — the intrusive-thought phenomenology that overlaps with binge urges
• GLP-1 drug interaction checker — check lisdexamfetamine + GLP-1 combination flags

Important disclaimer. This article is educational and does not constitute medical advice. Binge eating disorder is a serious psychiatric condition; diagnosis and treatment should be coordinated with a licensed psychiatrist or eating-disorder specialist, ideally with a registered dietitian and a CBT-trained therapist on the team. Lisdexamfetamine is a DEA Schedule II stimulant with abuse potential and a meaningful cardiovascular signal; it is contraindicated with structural heart disease, certain arrhythmias, and uncontrolled hypertension. GLP-1 receptor agonists for binge eating disorder are off-label; the regulatory standard for BED remains lisdexamfetamine. Anyone considering combination therapy should do so under coordinated psychiatry and obesity-medicine care. PMIDs were verified live against the PubMed E-utilities API on 2026-05-29.

Last verified: 2026-05-29. Next review: every 6 months, or sooner if a registration-grade RCT of semaglutide or tirzepatide in BED is published.