GLP-1 Drug Interactions: What Medications Are Affected?
GLP-1 Safety and Side Effects
22 studies reviewed
A 2024 systematic review found that GLP-1 receptor agonists did not produce clinically significant changes in overall drug exposure for most co-administered oral medications.
Calvarysky et al., Drug Safety, 2024
Calvarysky et al., Drug Safety, 2024
View as imageTens of millions of people worldwide now take GLP-1 receptor agonists for type 2 diabetes or obesity, and many of those people also take other daily medications. A 2024 systematic review of 22 pharmacokinetic studies found that injectable GLP-1 receptor agonists did not produce clinically significant changes in overall drug exposure for most co-administered oral medications.[1] That is the headline finding. But headline findings obscure the exceptions, and in pharmacology, exceptions can matter. Tirzepatide reduces oral contraceptive absorption. Oral semaglutide increases levothyroxine exposure by 33%. Sulfonylureas and insulin combined with any GLP-1 drug raise hypoglycemia risk. This article maps every documented GLP-1 drug interaction, identifies which ones are clinically relevant, and explains why the primary mechanism, delayed gastric emptying, produces such variable effects across different medications. For a broader overview of GLP-1 adverse effects beyond drug interactions, see our guide to GLP-1 side effects.
Key Takeaways
- A 2024 systematic review of 22 studies found GLP-1 receptor agonists do not produce clinically significant changes in AUC for most co-administered oral drugs (Calvarysky et al., Drug Safety)
- Subcutaneous semaglutide at steady state (1.0 mg) did not affect the pharmacokinetics of metformin, warfarin, atorvastatin, or digoxin to a clinically relevant degree (Hausner et al., Clinical Pharmacokinetics, 2017)
- Oral semaglutide increased total thyroxine (T4) exposure by 33% when co-administered with levothyroxine, requiring thyroid function monitoring (Hauge et al., Expert Opinion on Drug Metabolism and Toxicology, 2021)
- Tirzepatide is the only GLP-1/GIP agonist shown to significantly reduce oral hormonal contraceptive absorption (AUC, Cmax, and Tmax all decreased) in clinical trial data (Skelley et al., JAPhA, 2024)
- GLP-1 receptor agonists do not cause hypoglycemia when used alone or with metformin, but the risk increases substantially when combined with sulfonylureas or insulin (Filippatos et al., Review of Diabetic Studies, 2014)
- GLP-1 drugs are metabolized by endopeptidases, not cytochrome P450 enzymes, which is why they avoid the metabolic drug interactions common with small-molecule drugs (Min et al., Drug Design, Development and Therapy, 2025)
Why GLP-1 Drugs Have Fewer Drug Interactions Than Expected
Most drug interactions occur when two medications compete for the same metabolic enzymes, primarily the cytochrome P450 (CYP) family in the liver. GLP-1 receptor agonists sidestep this problem entirely. As peptide-based drugs, they are broken down by endopeptidases throughout the body rather than processed through CYP enzymes or UDP-glucuronosyltransferases.[2] This fundamental biochemical difference explains why a 2025 comprehensive review concluded that no clinically significant enzyme- or transporter-mediated drug interactions have been reported for any approved GLP-1 receptor agonist.[2]
The structural modifications that give GLP-1 drugs their long half-lives, amino acid substitutions, fatty acid conjugation, and albumin or IgG Fc fusion, also contribute to their metabolic independence. Native GLP-1 is degraded by dipeptidyl peptidase-4 (DPP-4) within two minutes. The engineered analogs resist this degradation through different strategies: liraglutide uses fatty acid conjugation to bind albumin, dulaglutide fuses directly with an IgG Fc fragment, and semaglutide combines amino acid substitutions with fatty acid conjugation.[3] None of these elimination pathways overlap with standard hepatic drug metabolism.
The drug interactions that do exist with GLP-1 receptor agonists stem from a different mechanism entirely: their effect on gastric emptying. Understanding the differences in how each GLP-1 drug interacts with other medications requires understanding this mechanism. For a deeper look at the gastrointestinal effects, see our article on gastroparesis and GLP-1 drugs.
The Gastric Emptying Mechanism: How GLP-1 Drugs Alter Drug Absorption
GLP-1 receptor agonists slow the rate at which the stomach empties food and medications into the small intestine. This is therapeutically useful: it reduces postprandial glucose spikes and helps limit triglyceride absorption.[4] But slower gastric emptying also means that co-administered oral medications spend more time in the stomach before reaching their absorption site in the intestine.
The pharmacokinetic consequence is predictable: delayed time to maximum concentration (Tmax) and sometimes reduced peak concentration (Cmax) for oral medications taken with GLP-1 drugs. However, total drug exposure (AUC, the area under the concentration-time curve) typically remains unchanged.[1] For most medications, this means the drug still gets fully absorbed; it just takes longer to reach peak blood levels.
An important caveat: tachyphylaxis occurs with prolonged GLP-1 receptor agonist treatment. The gastric emptying effect is strongest during the first weeks of treatment and diminishes with continued dosing.[4] This means drug interactions mediated by gastric emptying delay are most pronounced during dose initiation and escalation periods. For medications that require rapid onset or specific peak concentrations (pain relievers, certain antibiotics, rescue medications), this temporary delay could matter clinically even if total exposure is preserved.
Warfarin and Blood Thinners
Warfarin is the medication that generates the most concern when combined with GLP-1 drugs, because it has a narrow therapeutic index. Small changes in blood levels can mean the difference between effective anticoagulation and dangerous bleeding. The evidence, however, is reassuring.
In a dedicated pharmacokinetic trial, Hausner et al. (2017) gave 25 mg of warfarin to healthy subjects before and during steady-state subcutaneous semaglutide treatment (1.0 mg weekly). The AUC ratio for both S-warfarin and R-warfarin remained within the pre-specified bioequivalence interval of 0.80 to 1.25. Cmax was not affected to a clinically relevant degree. The international normalized ratio (INR) response was also unchanged.[5]
The 2024 systematic review confirmed this finding across multiple GLP-1 receptor agonists: warfarin showed reduced Cmax and delayed Tmax consistent with slower gastric emptying, but overall drug exposure was unaffected.[1] Current evidence does not support routine warfarin dose adjustments when starting a GLP-1 drug. That said, standard INR monitoring should continue unchanged, and heightened vigilance during dose escalation remains prudent for any narrow therapeutic index drug.
Direct oral anticoagulants (DOACs) like apixaban, rivarelbaan, and dabigatran have received less formal pharmacokinetic study with GLP-1 drugs. The 2025 comprehensive review noted that dabigatran's AUC increased by up to 205% in one interaction study context, a potentially clinically significant change for a narrow therapeutic index anticoagulant.[2] However, this finding requires replication and has not been specifically established in a GLP-1-specific interaction trial. Until more data are available, patients on DOACs starting GLP-1 therapy should have their anticoagulation status monitored during dose escalation.
Oral Contraceptives: The Tirzepatide Exception
This is the most clinically consequential drug interaction in the GLP-1 class, and it applies specifically to tirzepatide (Mounjaro/Zepbound), not to GLP-1 receptor agonists as a whole.
A 2024 literature review by Skelley et al. analyzed six clinical trials examining interactions between incretin agents and oral hormonal contraceptives. Five studies involving GLP-1 receptor agonists (semaglutide, liraglutide, exenatide, dulaglutide) showed no statistically or clinically significant impact on oral contraceptive pharmacokinetics. The sixth study, involving tirzepatide, showed a statistically significant reduction in AUC, Cmax, and Tmax of the oral contraceptive.[6]
The mechanism behind this difference is tirzepatide's dual receptor agonism. As both a GLP-1 and GIP receptor agonist, tirzepatide produces a more pronounced delay in gastric emptying than single-receptor GLP-1 drugs, particularly during the initial dose escalation phase. After subsequent doses, tachyphylaxis develops and the gastric emptying effect partially attenuates.[6] This is why the interaction is most relevant during the first weeks of treatment and after each dose increase.
The manufacturer's prescribing information for tirzepatide recommends that patients using oral contraceptives either switch to a non-oral method or add a barrier method for at least four weeks after starting tirzepatide and for four weeks after each dose escalation step.
A 2015 study specifically examined semaglutide and found no reduction in the bioavailability of combined oral contraceptives containing ethinylestradiol and levonorgestrel.[2] The media narrative around "Ozempic babies" conflates all GLP-1 drugs, but the pharmacokinetic data point specifically to tirzepatide.
Levothyroxine and Thyroid Medications
This interaction is specific to oral semaglutide (Rybelsus), not to injectable semaglutide (Ozempic/Wegovy) or other injectable GLP-1 drugs. The mechanism involves the SNAC (sodium N-[8-(2-hydroxybenzoyl) amino] caprylate) absorption enhancer in the oral formulation.
Hauge et al. (2021) conducted an open-label crossover trial in 45 healthy subjects, comparing levothyroxine (600 mcg single dose) given alone versus co-administered with oral semaglutide 14 mg at steady state. Total thyroxine (T4) exposure increased by 33% with co-administration. SNAC alone did not increase T4 exposure, indicating the interaction is driven by the combined oral semaglutide formulation rather than the absorption enhancer in isolation.[7]
The clinical implication: patients taking both oral semaglutide and levothyroxine should have thyroid function tests monitored, particularly after starting oral semaglutide or changing the dose. The dosing guidance for oral semaglutide (take first thing in the morning with up to 120 mL of water, then wait at least 30 minutes before eating or taking other medications) helps mitigate this interaction but does not eliminate it.
This interaction does not apply to injectable semaglutide. For patients on levothyroxine who need a GLP-1 drug, the injectable formulations avoid this pharmacokinetic complication entirely.
Insulin and Sulfonylureas: The Hypoglycemia Risk
Unlike the pharmacokinetic interactions discussed above, the interaction between GLP-1 drugs and insulin or sulfonylureas is pharmacodynamic: the drugs amplify each other's glucose-lowering effects, increasing hypoglycemia risk.
GLP-1 receptor agonists enhance insulin secretion in a glucose-dependent manner, meaning they stimulate insulin release primarily when blood glucose is elevated. This mechanism was expected to produce minimal hypoglycemia risk.[8] When used as monotherapy or combined with metformin, this expectation holds: hypoglycemia rates are low.[9]
The problem emerges with sulfonylureas and insulin. Sulfonylureas stimulate insulin release regardless of blood glucose levels. When combined with a GLP-1 drug that also enhances insulin secretion (even if glucose-dependently), the combined insulin output can drive blood glucose dangerously low. The same logic applies to exogenous insulin: adding a GLP-1 drug that further enhances insulin secretion on top of injected insulin creates additive glucose-lowering.[9]
Clinical guidelines recommend reducing sulfonylurea doses or reducing insulin doses by approximately 25% when initiating a GLP-1 receptor agonist. The FDA evaluated potential hypoglycemia safety signals across all 12 marketed GLP-1 agonists, with some evidence suggesting glucose-dependent action may not be as absolute as initially thought.[8]
For more on GLP-1 adverse effects in general, see our guide to GLP-1 side effects.
Metformin: A Safe Combination
Metformin is the most commonly co-administered drug with GLP-1 receptor agonists, and the evidence consistently shows no clinically meaningful interaction.
Hausner et al. (2017) tested metformin 500 mg twice daily for 3.5 days in healthy subjects before and during steady-state semaglutide treatment. The AUC ratio for metformin remained within the bioequivalence interval (0.80 to 1.25), and Cmax was not significantly altered.[5] No increased hypoglycemia was observed with the combination.
This finding is consistent across the GLP-1 class. The 2024 systematic review found no clinically significant changes in metformin pharmacokinetics with any GLP-1 receptor agonist.[1] No dose adjustment of either drug is required. The combination of GLP-1 agonists and metformin operates through complementary mechanisms: metformin reduces hepatic glucose output and improves insulin sensitivity, while GLP-1 drugs enhance glucose-dependent insulin secretion and slow gastric emptying.
Statins, Digoxin, and Other Common Medications
The dedicated semaglutide interaction trial by Hausner et al. (2017) also examined atorvastatin (40 mg single dose) and digoxin (0.5 mg single dose). Neither showed clinically relevant changes in AUC when co-administered with semaglutide at steady state. Cmax values were also unaffected.[5]
The systematic review extended these findings to statins as a class, ACE inhibitors, and acetaminophen: all showed the expected pattern of slightly delayed Tmax and sometimes reduced Cmax, but no clinically significant changes in total exposure.[1]
One case report described a potential interaction between liraglutide and clonazepam. A 23-year-old woman who had successfully used clonazepam for flight anxiety experienced a panic attack after taking liraglutide shortly before clonazepam, suggesting that delayed gastric emptying may have impaired the rapid onset needed for acute anxiolysis.[10] This is a single case report and cannot establish causation, but it illustrates an important principle: for medications that depend on rapid absorption and peak concentration for their effect (rescue medications, acute pain relievers, fast-acting sedatives), the delayed gastric emptying from GLP-1 drugs could matter even when total exposure is preserved.
Oral vs. Injectable GLP-1 Drugs: Different Interaction Profiles
The distinction between oral and injectable GLP-1 formulations matters for drug interactions. Oral semaglutide (Rybelsus) has a unique interaction profile because of its absorption enhancer, SNAC, and its strict dosing requirements.
Oral semaglutide has a bioavailability of only 0.8% under recommended conditions: taken with up to 120 mL of water on an empty stomach, followed by at least 30 minutes of fasting before eating or taking other medications.[11] Within-subject daily variability in absorption is 137%, though this smooths to 33% at steady state with daily dosing.
Co-administration of five placebo tablets with oral semaglutide reduced semaglutide exposure by 34%.[7] This means taking other pills at the same time as oral semaglutide can reduce the GLP-1 drug's own effectiveness. The dosing guidance (take oral semaglutide first, wait at least 30 minutes before other medications) exists specifically to avoid this problem.
Injectable GLP-1 drugs bypass the gastrointestinal tract entirely for their own absorption, so their blood levels are not affected by co-administered oral medications. The only interaction pathway for injectable formulations is their effect on other drugs via gastric emptying delay, not the reverse.
The 2025 comprehensive review noted that pharmacokinetic modeling approaches, including physiologically based pharmacokinetic (PBPK) models, can predict the extent of gastric-emptying-mediated interactions. Thirty models have been developed to date for GLP-1 receptor agonists.[2]
Anesthesia and Procedural Considerations
Delayed gastric emptying creates a specific concern for patients undergoing procedures requiring sedation or general anesthesia. Retained gastric contents increase the risk of pulmonary aspiration during intubation.
Jalleh et al. (2024) reviewed the evidence on retained gastric contents in patients taking GLP-1 receptor agonists. Their findings were nuanced: retained gastric contents at the time of upper GI endoscopy were found more frequently in GLP-1 RA users, but were rarely associated with actual pulmonary aspiration. Several important factors complicate simple recommendations: the long half-lives of these drugs mean that brief withholding periods may be insufficient; the gastric emptying effect occurs even at physiological GLP-1 concentrations; tachyphylaxis develops with prolonged treatment; and individuals with already slow gastric emptying before starting treatment show less additional effect.[4]
Current guidance from multiple anesthesiology societies suggests extended fasting periods for solid foods before procedures, point-of-care gastric ultrasound to check for retained contents, and consideration of prokinetic medications like erythromycin. Whether to withhold GLP-1 drugs before procedures and for how long remains debated, with limited evidence to support specific protocols.
The challenge is pharmacokinetic: semaglutide has a half-life of approximately one week, and tirzepatide's is approximately five days. Even withholding these drugs for several days before a procedure may not substantially reduce gastric emptying delay. The American Society of Anesthesiologists released guidance in 2023 suggesting GLP-1 drugs be withheld on the day of surgery for daily formulations and for one week for weekly formulations, but acknowledged this was based on expert opinion rather than clinical evidence. Patients with type 2 diabetes who already have some degree of gastropathy may face compounded risk, though Jalleh et al. noted that the effect of diabetes itself on retained gastric contents has received insufficient study.[4]
For more detail on the gastrointestinal motility effects, see our article on gastroparesis and GLP-1 drugs.
Emerging Questions: Weight Loss and Drug Metabolism
The 2025 comprehensive review by Min et al. raised an underexplored concern: GLP-1 receptor agonists produce significant weight loss, and weight loss itself can alter drug pharmacokinetics through multiple pathways. Reduced fat mass changes the volume of distribution for lipophilic drugs. Changes in body composition may alter CYP450 enzyme activity. Improved kidney function (higher glomerular filtration rate) from weight loss could increase renal clearance of drugs eliminated through the kidneys.[2]
These indirect pharmacokinetic effects are poorly studied. Most drug interaction trials for GLP-1 drugs were conducted in healthy volunteers over short periods, a design that cannot capture the cumulative effects of 15 to 20% body weight reduction over months or years. For patients on narrow therapeutic index drugs (warfarin, lithium, certain anti-epileptics, immunosuppressants), the gradual pharmacokinetic shifts from substantial weight loss may be more clinically relevant than the direct gastric emptying effects.
The 2026 safety review by Jalleh et al. in the Journal of Clinical Investigation reinforced that improved pharmacovigilance and more standardized assessment of adverse events in clinical trials are needed as GLP-1 drug use continues to expand.[12] The drug interaction landscape for these medications is still evolving.
A related but distinct question involves immunosuppressant drugs. Organ transplant recipients taking tacrolimus, cyclosporine, or mycophenolate are increasingly prescribed GLP-1 drugs for post-transplant diabetes or obesity. These immunosuppressants have narrow therapeutic windows, and even modest pharmacokinetic changes from altered absorption, shifted body composition, or improved renal function could affect graft survival. No dedicated pharmacokinetic interaction studies have been published for GLP-1 drugs with immunosuppressants, making this one of the most consequential data gaps in the field.
The emergence of dual and triple receptor agonists (targeting GLP-1, GIP, and glucagon receptors simultaneously) will add further complexity. Tirzepatide already demonstrated a distinct interaction profile from GLP-1-only drugs with oral contraceptives. As new multi-agonist peptides enter the market, each will require its own drug interaction characterization rather than relying on assumptions from the GLP-1 class. For broader context on the cost implications of managing these complex medication regimens, see our article on the cost-effectiveness of GLP-1s.
Drug Interaction Summary by Medication Class
| Medication | Effect with GLP-1 Drugs | Clinical Action |
|---|---|---|
| Warfarin | Delayed Tmax, AUC unchanged | Continue standard INR monitoring |
| Oral contraceptives | No effect (GLP-1 RAs); reduced absorption (tirzepatide) | Backup contraception with tirzepatide during dose escalation |
| Levothyroxine | 33% increased T4 exposure (oral semaglutide only) | Monitor thyroid function; injectable GLP-1s avoid this interaction |
| Sulfonylureas | Additive glucose-lowering, hypoglycemia risk | Reduce sulfonylurea dose when starting GLP-1 drug |
| Insulin | Additive glucose-lowering, hypoglycemia risk | Reduce insulin dose by ~25% when starting GLP-1 drug |
| Metformin | No clinically significant interaction | No dose adjustment needed |
| Statins | Delayed Tmax, AUC unchanged | No dose adjustment needed |
| Digoxin | Delayed Tmax, AUC unchanged | No dose adjustment needed |
| ACE inhibitors | Delayed Tmax, AUC unchanged | No dose adjustment needed |
| Acetaminophen | Delayed Tmax, AUC unchanged | No dose adjustment needed |
For related safety considerations, see our articles on GLP-1 agonists and pancreatitis, GLP-1s and gallbladder problems, thyroid cancer risk, retinopathy, nausea management, and injection site reactions.
The Bottom Line
The evidence consistently shows that GLP-1 receptor agonists produce few clinically significant drug interactions with most co-administered medications. The exceptions, tirzepatide's effect on oral contraceptives, oral semaglutide's effect on levothyroxine, and the pharmacodynamic hypoglycemia risk with insulin and sulfonylureas, are well-characterized and manageable with appropriate monitoring and dose adjustments. The indirect effects of GLP-1-mediated weight loss on drug pharmacokinetics remain an open research question that may become more clinically relevant as these medications are used long-term for obesity.