GLP-1 for Opioid Use Disorder: Buprenorphine and Methadone Evidence

Opioid use disorder (OUD) affects roughly six million US adults, and three FDA-approved medications — buprenorphine, methadone, and extended-release naltrexone — cut overdose mortality in half when patients stay on them. The newer question is whether GLP-1 receptor agonists belong in the OUD toolkit. The signal is real but early: a retrospective EHR study of T2D patients with OUD found semaglutide was associated with lower opioid-overdose incidence vs other antidiabetic drugs (Wang 2024^[1]), a pharmacoepi analysis in Addiction found GLP-1 / GIP prescriptions were associated with reduced substance-related outcomes in OUD patients (Qeadan 2025^[2]), and preclinical rodent work shows exendin-4 and liraglutide suppress drug-seeking for heroin and fentanyl (Douton 2021^[6], Urbanik 2025^[7]). What we do not have yet is a randomized controlled trial in OUD patients. This article walks through what the evidence supports, the pharmacokinetic interactions with buprenorphine and methadone, and the practical addiction-medicine framework for patients on medication for opioid use disorder (MOUD) who are also considering GLP-1 therapy for obesity.

The honest summary

• No GLP-1 is FDA-approved for OUD. The on-label use is type 2 diabetes (Ozempic, Mounjaro) or chronic weight management (Wegovy, Zepbound). OUD evidence is limited to one retrospective EHR study (Wang 2024^[1]), one pharmacoepi study (Qeadan 2025^[2]), preclinical rodent reinstatement work, and extrapolation from the alcohol-use-disorder trials (Hendershot 2025^[4], Klausen 2022^[5], Quddos 2023^[3]).
• MOUD remains the standard. Buprenorphine and methadone reduce all-cause and overdose mortality by roughly 50% when patients are retained (Mattick 2014 Cochrane^[9]). Extended-release naltrexone (Vivitrol) is effective for highly motivated patients who can complete a detox window (Krupitsky 2012^[10]). A GLP-1 does not substitute for these — it would be added on, not instead of.
• No clinically significant DDI with buprenorphine or methadone has been flagged in the GLP-1 DDI literature. The Calvarysky 2024 systematic review (Drug Safety^[8]) catalogued GLP-1 receptor-agonist interactions with oral medications and did not identify buprenorphine or methadone as a high-concern class. The mechanistic concern — delayed gastric emptying altering oral methadone absorption — is theoretical and has not been documented as a clinical event in published case series.
• Patients stable on MOUD for at least three months are reasonable GLP-1 candidates if they meet standard obesity criteria, but the prescribing decision should sit with the addiction-medicine clinician, not the obesity-medicine clinician alone.

What the human OUD evidence actually shows

The strongest human signal comes from Wang and colleagues at Case Western (JAMA Network Open 2024^[1]), who used the TriNetX EHR network to identify 33,006 patients with type 2 diabetes and a comorbid OUD diagnosis between 2017 and 2023. Patients newly prescribed semaglutide were compared with new users of seven other antidiabetic drug classes via propensity matching. Across pairwise comparisons, semaglutide initiators had a lower incidence of opioid overdose at 12 months: hazard ratios ranged from 0.32 to 0.68 depending on the comparator, with confidence intervals consistently excluding 1.0. The study is retrospective and the OUD population is the small subset who happen to have T2D — not generalizable to nondiabetic OUD — but it is the single largest human OUD signal in the GLP-1 literature.

Qeadan and colleagues (Addiction 2025^[2]) extended the picture with the Oracle EHR Real-World Data set, examining patients with OUD or AUD who received GLP-1 or GIP receptor agonist prescriptions versus matched controls. The headline result: incidence of overdose was substantially lower among the GLP-1 / GIP exposed cohort, with effect sizes broadly consistent with the Wang analysis. Both studies are subject to the standard pharmacoepi caveats — channeling bias, adherence asymmetry, residual confounding by metabolic status — but the direction is consistent.

We do not yet have a prospective randomized trial in OUD patients. The closest randomized comparator is the alcohol-use-disorder literature. Hendershot 2025 (JAMA Psychiatry^[4]) randomized 48 adults with AUD to semaglutide 1.0 mg weekly or placebo over nine weeks and reported reductions in laboratory drinking and heavy drinking days. Klausen 2022 (JCI Insight^[5]) randomized 127 adults to exenatide once weekly or placebo over 26 weeks and found no overall reduction in heavy drinking days but a significant reduction in a prespecified obese subgroup. Quddos 2023 (Scientific Reports^[3]) reported secondary-outcome reductions in alcohol consumption among obesity-trial participants on semaglutide and tirzepatide. The cross-substance read-across is biologically plausible — all three drugs of abuse converge on the mesolimbic dopamine system — but it remains an extrapolation.

The preclinical mechanism: reward-pathway modulation

The mechanistic case rests on the overlap between GLP-1 receptor expression and the brain reward circuit. GLP-1 receptors are present in the nucleus accumbens, ventral tegmental area, and lateral septum — the same pathways engaged by opioids, alcohol, and other substances of abuse. Douton 2021 (Behavioral Pharmacology^[6]) showed that the GLP-1 agonist exendin-4 reduced cue-induced reinstatement of heroin-seeking behavior in rats after self-administration training. Urbanik 2025 (Behavioral Pharmacology^[7]) extended the finding to fentanyl, reporting that liraglutide attenuated cue- and drug-induced seeking in female rats with an estrus-phase interaction.

The translational signal is real but preclinical: rodent reinstatement models predict craving-related outcomes but not all-cause efficacy. The published rat work does not establish that GLP-1 therapy reduces heroin use or overdose in humans — only that the receptor-level mechanism is biologically plausible.

MOUD pharmacokinetics: what the GLP-1 might bump into

Buprenorphine and methadone have different absorption routes, which matters because GLP-1 receptor agonists slow gastric emptying and can blunt the peak concentration of oral medications taken alongside the injection.

Buprenorphine is a partial mu-opioid agonist delivered sublingually (Subutex, Suboxone film) or as a monthly subcutaneous depot (Sublocade). Sublingual bioavailability bypasses the stomach entirely, so the delayed-gastric-emptying mechanism does not apply. Sublocade is a depot injection — pharmacokinetics are entirely independent of GI transit. Metabolism is CYP3A4-mediated to the active metabolite norbuprenorphine; GLP-1 receptor agonists are not CYP inhibitors or inducers at clinical doses, so a metabolic interaction is not expected. The Calvarysky 2024 systematic review^[8] did not identify buprenorphine as a high-concern interaction class.

Methadone is a full mu-opioid agonist taken orally as a daily liquid or tablet, dispensed via federally regulated opioid treatment programs (OTPs). Methadone is metabolized primarily by CYP3A4 and CYP2B6 with substantial between-patient variability. The delayed-gastric-emptying question matters more here than for buprenorphine: a peak plasma concentration that arrives later than usual could modestly shift the dosing experience, particularly during GLP-1 dose escalation when GI effects are strongest. The published DDI literature does not document this as a clinical event, but the theoretical mechanism warrants slower GLP-1 titration in methadone patients and closer check-ins during the 0.25 to 1.0 mg escalation window.

Extended-release naltrexone (Vivitrol) is delivered as a monthly intramuscular injection — no oral absorption to interfere with. The pharmacokinetic question does not apply.

QTc, methadone, and the GLP-1 safety profile

Methadone carries a known QT-prolongation risk, especially at doses above 100 mg/day or with concurrent QT-prolonging medications. GLP-1 receptor agonists do not appear in the QT-risk classifications (CredibleMeds, FDA cardiac safety signal databases). The injection-class semaglutide and tirzepatide cardiovascular safety data (SUSTAIN-6 for sema, SURPASS for tirz) did not identify a QT or arrhythmia signal. For methadone patients on QT-prolonging concomitants, the clinical risk does not change meaningfully with addition of a GLP-1.

The overdose-tolerance question after stop

One indirect mechanism warrants explicit discussion. Patients who reduce or stop opioid use during GLP-1 therapy — even modestly — will lose tolerance. If they relapse to prior-dose opioid use, overdose risk is elevated for the same reason it is elevated after any abstinence period. The Wang 2024 EHR data^[1] showed reduced overdose incidence on semaglutide, so the net effect appears protective; but clinical counseling should include the same reduced-tolerance-after-abstinence warning that addiction medicine routinely provides to MOUD patients considering taper.

Magnitude: reduction in opioid use days at 12 weeks

The practical framework

For a patient stable on MOUD for at least three months who is also a candidate for GLP-1 therapy on obesity criteria, the practical sequence is:

1. Addiction medicine consultation first. The MOUD prescriber should sign off on the addition of a GLP-1. For patients in an OTP (methadone), the OTP medical director is usually the right consultant.
2. Confirm MOUD stability. At least three months on a steady dose, no recent relapses, no escalating craving. Patients in active early treatment should defer GLP-1 initiation.
3. Slow GLP-1 escalation. Stay at each semaglutide or tirzepatide dose step for at least four weeks rather than rushing the titration. GI side effects peak during escalation and overlap with the early-MOUD discomfort patients are most sensitive to.
4. Monitor sleep, mood, and craving at each visit. Document baseline craving severity (a 0-10 scale or a brief instrument like the Opioid Craving Scale) and re-measure monthly. Any sustained increase warrants MOUD-side reassessment, not just GLP-1 dose adjustment.
5. Avoid initiating during pregnancy. MOUD (buprenorphine or methadone) is continued through pregnancy per ACOG and SAMHSA guidance; GLP-1 receptor agonists are discontinued. The combination question becomes moot for the duration of pregnancy.
6. Counsel about reduced-tolerance overdose risk if opioid use does decrease on the GLP-1. This is the same counseling addiction medicine already provides during MOUD taper.

Insurance, cost, and provider routes

MOUD is covered by Medicaid in every state under the federal SUPPORT Act and by most commercial plans. Buprenorphine sublingual generics run about $30/month; Sublocade is approximately $1,500/dose and is typically billed under the medical benefit. Methadone for OUD is dispensed only through federally certified OTPs and is generally bundled into the program fee. Extended-release naltrexone (Vivitrol) lists at roughly $1,500/dose monthly. GLP-1 therapy for obesity follows the standard coverage maze — commercial PA pathways, Medicare non-coverage for obesity-only indications, and state-by-state Medicaid variability. Patients on MOUD are often on Medicaid and face the same GLP-1 access barriers as any other Medicaid enrollee.

Provider routing is the key practical point. Obesity-medicine telehealth providers do not, as a rule, manage MOUD; addiction-medicine providers do not, as a rule, prescribe GLP-1 therapy for obesity. The patient in the middle needs either a dual-trained clinician (uncommon) or coordinated care between two providers (more common). Asking the GLP-1 provider whether they will accept a patient on Sublocade or methadone before paying for an intake is reasonable.

Related research and tools

• GLP-1 + naltrexone for alcohol use disorder — the cross-substance evidence base most relevant to OUD extrapolation
• GLP-1 for alcohol use disorder — the AUD RCT literature (Hendershot 2025, Klausen 2022) in depth
• GLP-1 + bupropion stacking — the parallel question for nicotine and reward-pathway stacking
• GLP-1 with lithium and mood stabilizers — another psychiatry-comorbidity decision tree
• Naltrexone (Contrave) for weight loss — the obesity-medicine use of an opioid antagonist

Important disclaimer. This article is educational and does not constitute medical advice. Opioid use disorder is a chronic, relapsing condition that requires specialist care. No GLP-1 receptor agonist is FDA-approved for OUD, and the published evidence does not support discontinuing MOUD in favor of GLP-1 therapy under any circumstances. Patients on buprenorphine, methadone, or extended-release naltrexone considering GLP-1 therapy for obesity should coordinate care with their addiction-medicine prescriber. Pregnancy considerations differ — MOUD is continued, GLP-1 is discontinued — per ACOG and SAMHSA guidance. 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 randomized controlled trial of a GLP-1 receptor agonist in opioid use disorder is published.