Drug–Drug and Drug–Disease Interactions Across Antidiabetic Drug Classes: A Narrative Review and Practical Recommendations

Abstract

Background: The pharmacological management of type 2 diabetes mellitus has become increasingly complex due to expanding therapeutic options and the high prevalence of multimorbidity in affected patients. As a result, the risk of drug–drug and drug–disease interactions has grown significantly, with potential implications for glycemic control, safety, and treatment outcomes. Objective: This narrative review provides a comprehensive, class-based overview of clinically relevant interactions associated with antidiabetic medications, highlighting their mechanisms, clinical consequences, and practical management strategies. Methods: A targeted literature search was conducted using major medical databases to identify clinical studies, meta-analyses, pharmacovigilance reports, and evidence-based guidelines concerning interactions related to key antidiabetic drug classes. Interactions were categorized as pharmacokinetic, pharmacodynamic, or disease-related. Results: Significant variability exists across antidiabetic drug classes in terms of interaction profile and clinical relevance. Metformin presents interaction risks mainly through renal impairment or co-administration with drugs affecting lactate metabolism. Sulfonylureas and glinides are strongly associated with hypoglycemia-enhancing interactions. DPP-4 inhibitors generally exhibit a low interaction burden, whereas GLP-1 receptor agonists may interact through delayed gastric emptying. SGLT2 inhibitors require caution in patients with diuretics or conditions predisposing them to dehydration or ketoacidosis. Insulin remains highly sensitive to pharmacodynamic interactions with a broad spectrum of therapies. Underlying renal, hepatic, and cardiovascular conditions further modify the interaction risk. Conclusions: Understanding class-specific interaction profiles is essential for personalized and safe diabetes management. Careful medication review, close metabolic monitoring, and individualized dose adjustments can mitigate the risk of harmful interactions. Further research is needed to elucidate interactions in populations with advanced multimorbidity and polypharmacy.

1. Introduction

Diabetes mellitus is a complex metabolic pathology that negatively influences the health status, life expectancy and quality of life of patients and which also has a huge negative impact on healthcare systems. According to the International Diabetes Federation (IDF), 11.1% of the adult population (20–79 years old) lives with diabetes. About 90% of this is T2DM, and among the factors that contribute to the increase in incidence are urbanization, population aging, decreased physical activity, and obesity [1].

T2DM is characterized by hyperglycemia, but patients most often have comorbidities that require associated treatments. The most common concomitant diseases are dyslipidemia and hypertension, which together lead to the long-term complications of diabetes, through both microvascular (retinopathy, nephropathy, neuropathy) and macrovascular (cardiovascular disease, cerebrovascular disease, peripheral arterial disease) pathways [2,3,4].

Maintaining blood glucose levels within normal limits is essential to avoid complications, but, globally, patients with T2DM have very poor blood glucose control, with percentages ranging from 45% to 93% and with great variability between different countries and even within the population of the same country. Poor glycemic control is a major health problem and contributes to the development of diabetes complications. In the short term, hyperglycemic crises such as diabetic ketoacidosis or hyperglycemic hyperosmolar syndrome are the most common and serious acute complications. Therefore, improving the achievement and maintenance of glycemic targets is a priority for all specialists involved in the management of this disease [5,6].

The drug treatment of T2DM has changed significantly in recent years, with the emergence of new therapeutic classes with innovative mechanisms, which have entered the treatment guidelines due to the important benefits brought to the management of this disease. According to the IDF Global Clinical Practice Recommendations for Managing Type 2 Diabetes of 2025, metformin remains the first-choice monotherapy due to its worldwide availability, high efficacy, low risk of hypoglycemia or other adverse effects, optimal cost-effectiveness ratio, and long-term cardiovascular benefits [7,8,9]. In patients at risk or with cardiorenal complications, it is recommended to combine metformin with a sodium-glucose cotransporter-2 (SGLT2) inhibitor or glucagon-like peptide-1 (GLP-1) receptor agonist, the former being preferred in case of heart failure [10].

To significantly reduce the risk of cardiovascular mortality, patients are frequently treated with drug combinations that act synergistically in this regard. In addition, patients with diabetes frequently suffer from nervous system disorders, such as depression, which also require drug therapy [11]. With the increase in the number of drugs in therapeutic regimens, the risk of clinically significant drug–drug interactions (DDIs) also increases, with an increased possibility of adverse reactions. These interactions constitute an important, but often unrecognized, source of medical errors [12,13,14].

When a polypharmacological approach to a patient’s comorbidities is necessary, all possible interactions that may occur must be considered. A DDI is defined as an increase or decrease in the effect of a drug (the object of the interaction), under the influence of another substance (the precipitant of the interaction), which can be a drug, herb, or dietary supplement [14].

DDIs can be intentional, when a cumulative beneficial effect of the associated drugs is sought (e.g., antihypertensive, anti-infective, or antidiabetic), but also unwanted, when the associated drugs can lead to a decrease in the effect or the occurrence of adverse effects. The mechanisms of DDIs can be pharmacokinetic, with a change in the plasma concentration of the victim drug, by changing its metabolic processes at the level of absorption, distribution, metabolism, or elimination (ADME), or pharmacodynamic, with a change in the intensity of the drug effect, without changing the plasma concentration [15].

For DDIs to be considered, they must be clinically significant. Most drug combinations either do not produce any interactions at all or they are not clinically significant. There are several conditions for an interaction to be clinically significant: (1) the drug subject to the interaction must have a narrow therapeutic index (a small decrease in plasma concentration leads to loss of effect, a small increase in concentration leads to adverse effects); (2) a drug with an exponential relationship between plasma concentration and effect (a small change in concentration leads to an exponential change in the intensity of the effect); (3) drugs that undergo an intense first pass through the liver (in this case, an enzyme inhibitor or inducer will cause a significant change in plasma concentration and clinical efficacy); and (4) drugs eliminated in large part by renal excretion through active tubular secretion (in this case, another drug eliminated by the same active transporter will inhibit the elimination of the other drug) [16].

DDIs involving antidiabetic agents can have important clinical consequences, such as increased toxicity, decreased therapeutic efficacy, or difficulties in glycemic control. Therefore, healthcare professionals must be aware of the clinically significant interactions of these drugs, which may occur with commonly prescribed drugs such as antibiotics, cardiovascular medications, psychiatric medications, or even OTC drugs [17].

It is the role of healthcare professionals to try to prevent interactions from occurring or to keep them under control when they do occur. In this regard, a permanent review of all medications and supplements that patients are taking is necessary, patients must be educated to immediately report any sign or symptom of a potential interaction, and open communication with the patient is essential.

Given the abundance of information on antidiabetic drug interactions and the fact that some of these interactions are not clinically relevant, the aim of this article is to highlight the latest, most accurate, and clinically important data on these interactions. We hope that our article will become a valuable tool for prescribers to safely recommend these drugs to patients with polypathology and polymedication, in whom the risk of drug interactions is very high.

The present narrative review is intended for clinicians and pharmacists with pharmacology orientation (internists, diabetologists, endocrinologists, family physicians, clinical, and hospital pharmacists) who require a good understanding of interaction mechanisms in order to anticipate, recognize, and manage drug–drug and drug–disease interactions in patients with T2DM.

2. Materials and Methods

A targeted literature search was conducted between October 2025 and March 2026 in PubMed, Scopus, and the Cochrane Library, complemented by direct consultation of the regulatory documents of the European Medicines Agency (EMA), the United States Food and Drug Administration (FDA), and the Romanian National Agency for Medicines and Medical Devices (ANMDMR, for nationally authorized SmPCs). Pharmacovigilance signal communications of the EMA (EudraVigilance) and the FDA (FAERS/Drug Safety Communications) were consulted for post-marketing safety information.

Search terms combined the names of antidiabetic drug classes and individual agents (“metformin”, “sulfonylureas”, “glinides”, “DPP-4 inhibitors”, “GLP-1 receptor agonists”, “SGLT2 inhibitors”, “thiazolidinediones”, “alpha-glucosidase inhibitors”, “insulin”, “pramlintide”) with the modifiers “drug–drug interaction”, “drug–disease interaction”, “pharmacokinetic interaction”, “pharmacodynamic interaction”, “lactic acidosis”, “hypoglycemia”, “ketoacidosis”, and “polypharmacy”.

Selection of the cited primary literature additionally favored sources with explicit methodological strength (peer-reviewed studies, prospective designs, and meta-analyses for outcome claims) while regulatory documents were used as authoritative sources for recognized DDIs and drug–disease interactions.

Sources were included if they reported a defined pharmacokinetic or pharmacodynamic mechanism of interaction, with a clinical outcome attributable to an interaction, or an officially approved interaction warning, contraindication, or dose-adjustment recommendation in current regulatory product information.

Single case reports were excluded unless the interaction described was not documented in larger studies and represented a clinically significant signal subsequently incorporated into regulatory product information.

3. Results and Discussion

3.1. Biguanide Drugs—Metformin

Metformin is a biguanide antidiabetic drug which acts on glucose homeostasis by reducing glucose production in the liver and the intestinal absorption of glucose. Though it also improves its peripheral uptake and use, therefore improving insulin sensitivity, it also increases GLP-1 secretion and has immunomodulatory effects [18,19].

Regarding its role in T2DM management, only in the guidelines published in the last few years has metformin been replaced as a first line treatment of T2DM with GLP-1 agonists or SGLT2 inhibitors in patients with high cardiovascular risk, but it still remains the most frequently prescribed oral antidiabetic drug in Europe [20,21,22,23].

3.1.1. Pharmacokinetic Interactions Leading to Increased or Decreased Metformin Serum Levels

There are several important transporters involved in metformin hepatic uptake and/or renal clearance, including OCT1, OCT2, MATE1, and MATE2-K. Therefore, drugs inhibiting these proteins will reduce the uptake and/or elimination of metformin, increasing the risk of lactic acidosis and possibly other relevant adverse reactions, especially given the fact that metformin is excreted unchanged in the urine [24]. The severity of the interactions depends on the inhibitory potency, the dose of the inhibitor and/or metformin, as well as relevant genetic polymorphisms. High risk interactions include those with cimetidine, fexinidazole, ranolazine, risdiplam, tafenoquine, and vimseltinib, which typically require therapy modification or a reduction of the dose of metformin; for example, the current recommendation is that, during the concurrent use of ranolazine 1000 mg twice daily, the metformin daily dose should be limited to 1700 mg [25,26]. Moderate risk interactions, which typically require careful monitoring (with metformin AUCs being increased by as much as between 10% to a height of 150–200%), include those with abemaciclib, arimoclomol, bictegravir, cephalexin, copanlisib, crizotinib, dolutegravir, fedratinib, gilteritinib, givinostat, ondansetron, ribociclib, topiramate, trilaciclib, trimethoprim, vandetanib, and verapamil [14,24,26].

In addition, beside the drugs which might increase metformin serum concentrations, it is also worth mentioning the interactions which are associated with decreased metformin serum concentrations through reduced absorption. More specifically, patiromer, because of its tendency to bind to metformin in the gastrointestinal tract, should be administered at least 3 h before/3 h after metformin [27].

3.1.2. Pharmacokinetic Interactions Leading to Increased or Decreased Serum Levels of Co-Administered Drugs

Besides being a substrate for several transporters involved in hepatic uptake and renal clearance, metformin can also influence the serum levels of several drugs. Though such interactions are typically considered to be of moderate risk, most of them require monitoring, especially for narrow therapeutic index medication. Relevant examples include topiramate and vitamin K antagonists (VKAs), for which the systemic exposure was decreased when combined with metformin. Initiating metformin in patients treated with VKAs was found to reduce VKA AUC and half-life and/or INR, depending on the VKA drug. However, the studies which have highlighted these interactions were only undertaken with phenprocoumon or warfarin, not acenocoumarol [28,29]. Nevertheless, regardless of the VKA used, careful INR monitoring is warranted during concomitant treatment with VKAs and metformin [30].

3.1.3. Pharmacodynamic Interactions Leading to an Increased Risk of Lactic Acidosis

Though rare, with a reported incidence of approximately between 5 and 50 cases per 100,000 patient-years, lactic acidosis is one of the most severe and clinically relevant adverse reactions of metformin, with risk factors such as renal and cardiac failure, acute coronary syndrome, septic and cardiogenic shock, or use of alcohol or of other drugs which might cause this condition [31,32,33]. The current mortality of metformin-induced lactic acidosis (MALA) is estimated at around 25% [34]. The mechanisms for MALA include impaired oxidative phosphorylation, reduced metabolism of pyruvate and increased conversion to lactate [35]. As the risk of metformin-induced lactic acidosis is significantly higher in altered kidney function, metformin use is contraindicated if estimated glomerular filtration rate (eGFR) is lower than 30 mL/min/1.73 m², while dose adjustment is recommended if eGFR is lower than 45 mL/min/1.73 m², but at least equal to 30 mL/min/1.73 m²; the contraindication in eGFR being lower than 30 mL/min/1.73 m² is consistent across the different updated guidelines and public documents from medicines agencies [10,20,21,33]. Nonetheless, due to similar reasons, metformin can be used in patients with stable heart failure, while in those hospitalized for acute heart failure it is contraindicated.

Besides the relevant clinical risk factors of lactic acidosis, it is of utmost importance to also consider other possible pharmacological agents, which might precipitate lactic acidosis in metformin patients. For example, drugs which have, per se, an increased probability of lactic acidosis are carbonic anhydrase inhibitors, such as acetazolamide, methazolamide, topiramate or zonisamide. In addition, pharmacological agents associated with higher risk of metformin-induced lactic acidosis due to kidney injury include iodinated contrast agents (e.g., iobitridol, iodixanol, iohexol, iomeprol, iopamidol, iopromide, and ioversol) and nonsteroidal anti-inflammatory drugs (NSAIDs) (both selective and non-selective) [35,36]. The majority of these drugs have moderate risk interactions with metformin, with the notable exception of iodinated contrast agents, which typically require, depending on the guideline, withholding metformin for a few days before and after the procedure, sometimes only for specific patients. For example, the latest diabetes and cardiovascular diseases guideline developed by the European Society of Cardiology and the European Association for the Study of Diabetes recommends that only for those with chronic kidney disease should metformin be stopped immediately before a procedure requiring contrast agents (i.e., angiography or percutaneous coronary intervention) and be discontinued for 48 h or until the renal function has returned to baseline [20].

3.1.4. Pharmacodynamic Interaction with 18F-Fludeoxyglucose

In addition, another high risk pharmacodynamic interaction involves that with 18F-fludeoxyglucose (FDG-F18), possibly through the increased uptake of FDG-F18 in the intestines and/or colon, leading to false negative diagnostic positron emission tomography–computer tomography (PET-CT) tests, therefore masking an underlying malignancy. Discontinuing metformin for 48 h prior to the investigation is currently recommended [37,38].

3.2. Sulfonylureas

Sulfonylureas (SUs) have been used as an important treatment for type 2 diabetes mellitus since the 1950s, beginning with tolbutamide [39], followed by first-generation agents such as chlorpropamide, acetohexamide, and tolazamide. Second-generation sulfonylureas, including glyburide (glibenclamide) and glipizide, were introduced in the United States in 1984 [40]. Later, glimepiride, a third-generation sulfonylurea [41], became available in 1995 [42]. Sometimes glimepiride is considered a second-generation sulfonylurea [41,43]. However, in the last decades, SUs use has decreased, being replaced by safer alternatives [44].

SUs stimulate insulin secretion by closing pancreatic β-cell KATP channels composed of Kir6.2 and sulfonylurea receptor (SUR) subunits, leading to membrane depolarization, calcium influx, and insulin release. Excellent glycemic control remained after 10 years of sulfonylurea therapy in all examined patients with mutations in genes encoding KATP channel subunits [45]. SUs and glinide drugs, which bind to SUR1, close the channel through a pathway independent of ATP and are now the primary therapy for neonatal diabetes mellitus caused by mutations in the genes encoding KATP channel subunit [46].

The primary use-limiting side effect of SUs is hypoglycemia, although they are also associated with weight gain [47]. Generally, a hemoglobin A1C (HbA1c) reduction 1.4% (glimepride) [48] or 1.50–1.82% (glipizide) [49] can be expected in a responsive patient with T2DM.

Globally, SUs remain an important option, either as monotherapy or more commonly as a combination with GLP-1 RA, metformin, pioglitazone, or SGLT2 inhibitors [10]. Some patients with neonatal diabetes or MODY-3 genetic diabetes may experience sustained, long-term benefits from SUs [50].

However, SUs are contraindicated in type 1 diabetes, pregnancy, breastfeeding, and severe hepatic or renal impairment. Currently the EMA has authorized the following: glibenclamide (glyburide), glipizide, gliquidone, gliclazide and glimepiride [51]. Because these are the most clinically used SUs, we will present interactions related to them.

3.2.1. Pharmacokinetic DDIs of SUs

The pharmacokinetic profile of SUs is similar, but there are some differences between the drugs currently used. The pharmacokinetic parameters of the second-generation sulfonylureas can be seen in

Table 1. Comparative pharmacokinetics of sulfonylureas.

SUs	Latency (h)	Duration (h)	t1/2 (h)	Bound to Plasma Albumin	Metabolization	Elimination
Second generation
Glibenclamide (glyburide—USA) [52,53]	1–4	10–24	4–13	99%	Liver (2 active metabolites and 3 inactive metabolites) CYP3A4, CYP2C9, CYP2C19	Bile (60%) and urine (40%)
Glipizide [54,55]	0.5	10–24	2–4	98–99%	Liver 90% (inactive metabolites) CYP2C9 transported into hepatocytes primarily via the organic anion-transporting polypeptide 1B3 (OATP1B3)	Urine (<10% as unchanged drug; 80% as metabolites); feces (10%)
Gliquidone [56,57]	2.25	8–10	5.7–9.4	99%	Liver 99% (inactive metabolites)	Feces (86%) and urine (5%)
Gliclazide (not in USA) [58]	4–6	6–24	12–20	95%	Liver 99% (inactive metabolites) CYP2C19, CYP2C9	Urine (60–70%) and feces (10–20%)
Second or third generation
Glimepiride [59,60,61]	2–3	10–24	5–9	99%	Liver (2 inactive metabolites) CYP2C9	Urine (58%) and feces (35%)

.

From the pharmacokinetic profile of SUs presented in Table 1, we can state that SUs are rapidly absorbed after oral administration; SUs in plasma are largely (90–99%) bound to albumins, which may can contribute to drug interactions, but they are not easily detached by acidic medicinal products; and that the liver metabolizes all SUs, and the metabolites are excreted in the urine.

Gliclazide and glimepiride are substrates of OATP1B1 and glibenclamide and glipizide are substrates of OATP1B3. Chen Y., et al. have confirmed the interaction between these SUs and rosuvastatin. No transport was observed for gliquidone, suggesting that it is not a substrate of either transporter [62].

In individuals carrying CYP2C9 variants, studies show increased efficacy for several sulfonylureas (glibenclamide, gliclazide), with no negative clinical impact, so no treatment adjustment is needed. For glimepiride, although there is a slight increase in hypoglycemia risk, the overall improvement in efficacy is considered more important. Therefore, no therapeutic changes are recommended for these gene–drug interactions [63].

Elexacaftor, tezacaftor, and ivacaftor may increase serum concentrations of the CYP2C9 substrates [64,65] gliclazide, glimepiride, glipizide, gliquidone, and glyburide. One should consider monitoring for increased effects and toxicities of these CYP2C9 substrates.

Concomitant administration of glyburide and bosentan is contraindicated due to the increased incidence of elevated liver enzymes. Glyburide may decrease serum concentrations of bosentan and bosentan may decrease serum concentrations of glyburide [66]. Concomitant use of glyburide and bosentan should be avoided.

The most important adverse reactions of SUs are hypoglycemia, which in severe cases may progress to coma, and weight gain, typically ranging from 1 to 3 kg, as glycemic control improves. Short-acting SUs (glipizide) are indeed safer than long-acting glimepiride in older adults [67].

The hypoglycemic effect of sulfonylureas may be enhanced by various mechanisms (decreased hepatic metabolism or renal excretion, displacement from protein-binding sites). Some drugs (sulfonamides, clofibrate, gemfibrozil, warfarin, and salicylates) displace the sulfonylureas from binding proteins, thereby increasing SUs plasma concentration and the risk of hypoglicemia [68].

3.2.2. Pharmacodynamic DDIs

Rarely, patients treated with these drugs develop an alcohol-induced flush similar to that caused by disulfiram or hyponatremia. Although there has been longstanding controversy over the cardiovascular safety of sulfonylureas, recent comparative clinical trials indicate that this class of drugs has no more risk of cardiovascular events than other commonly used glucose-lowering agents [69,70].

Other pharmacodynamic DDIs of SUs that are common to all antidiabetic agents will be discussed in Section 3.11 and Section 3.12.

3.2.3. Drug–Disease Interactions

SUs in hepatic and renal impairment

SUs should be administered with caution to patients with either renal or hepatic insufficiency. Kidney failure does not affect its elimination as long as creatinine clearance remains above 30 mL/min. Glipizide, gliclazide, and gliquidone are preferred in CKD, while glyburide and glimepiride are to be avoided [71].

SUs in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency

Patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency may be at an increased risk of sulfonylurea-induced hemolytic anemia; however, cases have also been described in patients without G6PD deficiency during postmarketing surveillance. One should use this with caution and consider a non-sulfonylurea alternative in patients with G6PD deficiency [57].

3.3. Meglitinides

Meglitinides (repaglinide, nateglinide) are non-sulfonylurea KATP channel modulators. They are oral insulin secretagogues that stimulate insulin release by closing KATP channels in pancreatic β cells, although not by direct binding to the SUR [72]. Meglitinides activity is dependent on the presence of functioning β cells and glucose. In contrast to sulfonylurea insulin secretagogues, meglitinides have no effect on insulin release in the absence of glucose. As such, meglitinides are more effective at reducing postprandial blood glucose levels than fasting blood glucose levels and require a longer duration of therapy (approximately one month) before decreases in fasting blood glucose are observed [73].

3.3.1. Pharmacokinetic DDIs of Meglitinides

Repaglinide is absorbed rapidly from the GI tract, and peak blood levels are obtained within 1 h with a t1/2 of approximately 1 h. Repaglinide is metabolized primarily by the liver (CYP3A4 and CYP2C8 isoenzymes) to inactive metabolites (Table 2). Because a small proportion (~10%) is metabolized by the kidney, dosing of the drug in patients with renal insufficiency also should be performed cautiously [74]. Repaglinide has more clinically significant metabolic interactions because of its metabolism through CYP2C8 and CYP3A4 [75].

Because of its rapid onset, repaglinide is indicated for use in controlling postprandial glucose excursions. The drug should be taken just before each meal in doses of 0.25–4 mg (maximum 16 mg/day). Hypoglycemia is a risk if the meal is delayed or skipped or contains insufficient carbohydrates [76]. It can be used in patients with renal impairment and in the elderly. Repaglinide is used as monotherapy or as combination therapy. There is no sulfur in its structure, so repaglinide may be used in patients with T2DM with a severe sulfur or sulfonylurea allergy [77]. Repaglinide is preferred over nateglinide in the meglitinide class for patients with decreased kidney function [74].

Table 2. Comparative pharmacokinetics of meglitinides.

Parameter	Repaglinide [75]	Nateglinide [78]
Absorption	Rapid	Rapid
Absolut bioavailability	56% ± 9%	73%
Volume of distribution	31 L	10 L
Protein binding, plasma	>98% to albumin and α1-acid glycoprotein	98%, primarily to albumin
Metabolism	Hepatic via CYP3A4 and CYP2C8 isoenzymes; glucuronidation to inactive metabolites; a substrate for active hepatic uptake transporter (organic anion transporting protein OATP1B1)	Hepatic via CYP2C9 (70%) and CYP3A4 (30%) to metabolites; the isoprene minor metabolite possesses potency similar to that of the nateglinide
Half-life elimination	~1 h	1.5 h
Time to peak, plasma	1 h	1 h
Maximal biological effect	3–3.5 h	4 h
Excretion	Feces (~90%, <2% as unchanged drug); urine (~8%, 0.1% as unchanged drug)	Urine (83%, 16% as unchanged drug); feces (10%)

Table 2. Comparative pharmacokinetics of meglitinides.

Parameter	Repaglinide [75]	Nateglinide [78]
Absorption	Rapid	Rapid
Absolut bioavailability	56% ± 9%	73%
Volume of distribution	31 L	10 L
Protein binding, plasma	>98% to albumin and α1-acid glycoprotein	98%, primarily to albumin
Metabolism	Hepatic via CYP3A4 and CYP2C8 isoenzymes; glucuronidation to inactive metabolites; a substrate for active hepatic uptake transporter (organic anion transporting protein OATP1B1)	Hepatic via CYP2C9 (70%) and CYP3A4 (30%) to metabolites; the isoprene minor metabolite possesses potency similar to that of the nateglinide
Half-life elimination	~1 h	1.5 h
Time to peak, plasma	1 h	1 h
Maximal biological effect	3–3.5 h	4 h
Excretion	Feces (~90%, <2% as unchanged drug); urine (~8%, 0.1% as unchanged drug)	Urine (83%, 16% as unchanged drug); feces (10%)

Nateglinide is most effective when administered at a dose of 120 mg, three times daily, 1 to 10 min before a meal. It is metabolized primarily by hepatic CYPs (2C9, 70%; 3A4, 30%) and should be used cautiously in patients with hepatic insufficiency. About 16% of an administered dose is excreted by the kidney as unchanged drug (Table 2).

Nateglinide therapy may produce fewer episodes of hypoglycemia than other currently available oral insulin secretagogues, including repaglinide [79]. A major difference between the two meglitinides is that nateglinide has an active metabolite, whereas repaglinide metabolites are inactive (Table 2).

Nateglinide should be used with caution in patients with moderate-to-severe hepatic impairment and in patients with severe renal impairment due to risk of hypoglycemia. It may be necessary to discontinue nateglinide and administer insulin if the patient is exposed to stress (e.g., fever, trauma, infection, surgery) [78].

Meglitinides share several pharmacokinetic features such as rapid absorption, short half-life, and hepatic metabolism, making them suitable for postprandial glucose control. However, important differences exist: nateglinide is primarily metabolized by CYP2C9 and produces an active metabolite and repaglinide is metabolized by CYP3A4 and CYP2C8 and has more clinically significant drug interactions. Repaglinide interactions often require dose limitation, therapy modification, or avoidance of the combination [76], while nateglinide interactions typically require monitoring only.

Understanding these similarities and differences is essential for individualizing therapy, minimizing drug interactions, and reducing the risk of hypoglycemia in patients with T2DM.

DDIs for repaglinide

Nonsteroidal anti-inflammatory drugs, salicylates, sulfonamides, and other highly protein-bound drugs may potentiate the hypoglycemic effects of repaglinide [61].

Concomitant therapy repaglinide with gemfibrozil should be avoided (strong CYP2C8 inhibitor). This combination is associated with considerably enhanced and prolonged blood glucose-lowering effects of repaglinide [80].

Concomitant use of clopidogrel and repaglinide should be avoided. Hypoglycemia was observed during coadministration. The mechanism of this interaction is that the glucuronide metabolite of clopidogrel inhibits CYP2C8, an enzyme responsible for repaglinide metabolism [81].

Similar precautions apply to other CYP2C8 inhibitors, including linzagolix [82], cannabidiol, deferasirox, leflunomide, pirtobrutinib, selpercatinib, and teriflunomide, as well as weaker inhibitors such as trimethoprim and pazopanib [83]. Cyclosporine also increases repaglinide exposure through inhibition of OATP1B1-mediated hepatic uptake and CYP3A4 metabolism, requiring limitation of the repaglinide dose to a maximum of 6 mg/day [84].

Repaglinide efficacy may be reduced by CYP3A4 strong inducers (such as apalutamide, carbamazepine, encorafenib, enzalutamide, fosphenytoin, lumacaftor/ivacaftor, mitotane, and phenytoin) and moderate inducers (including bexarotene (systemic), bosentan, cenobamate, dabrafenib, dipyrone, efavirenz, elagolix, estradiol/norethindrone, eslicarbazepine, etravirine, fexinidazole, lorlatinib, mitapivat, modafinil, nafcillin, pacritinib, pexidartinib, phenobarbital, primidone, repotrectinib, rifabutin, rifapentine, sotorasib, and St John’s wort) [83].

Conversely, strong CYP3A4 inhibitors, such as adagrasib, atazanavir, ceritinib, clarithromycin, cobicistat, darunavir, idelalisib, itraconazole, ketoconazole (systemic), levoketoconazole, lonafarnib, lopinavir, mifepristone, nefazodone, nelfinavir, nirmatrelvir/ritonavir, ombitasvir/paritaprevir/ritonavir (with or without dasabuvir), posaconazole, relacorilant, ritonavir, and voriconazole, as well as letermovir (via inhibition of CYP3A4 and OATP1B1/1B3) [85] may increase plasma concentrations of repaglinide, thereby elevating the risk of hypoglycemia and requiring close monitoring and possible dose reduction [83]. Therefore, careful clinical monitoring and individualized dose adjustments are essential to ensure both the safety and efficacy of therapy.

If repaglinide is combined with an inhibitor of the OATP1B1/1B3 substrate, such as belumosudil [86], bulevirtide [87] ceftobiprole, darolutamide, eltrombopag [88], encorafenib, leniolisib [89], teriflunomide, and voclosporin, serum concentrations of repaglinide may increase, raising the risk of hypoglycemia. Coadministration should be avoided or patients should be monitored for increased drug concentrations and associated toxicities.

DDIs specific to nateglinide

Strong CYP3A4 inducers, such as apalutamide, carbamazepine, encorafenib, enzalutamide, fosphenytoin, lumacaftor/ivacaftor, mitotane, phenytoin, and rifampin, may decrease serum concentrations of nateglinide [90], potentially reducing its therapeutic effect; therefore, dose increases of nateglinide may be required.

Conversely, moderate CYP2C9 inhibitors, including adagrasib, asciminib, fluconazole, mifepristone, nitisinone, ivacaftor, vanzacaftor, tezacaftor, and deutivacaftor, may enhance the effects of nateglinide, increasing the risk of hypoglycemia [64,65]. Accordingly, patients should be closely monitored for signs of excessive glucose lowering.

Additionally, combination with elexacaftor/ivacaftor/tezacaftor may increase the risk of nateglinide-related toxicities, particularly hypoglycemia [64,65]. This interaction is likely mediated by inhibition of OATP1B1/1B3 transporters and ivacaftor-related inhibition of CYP2C9, a key enzyme involved in nateglinide metabolism [91].

3.3.2. Pharmacodynamic DDIs

Pharmacodynamic drug interactions shared across all antidiabetic agents will be addressed in Section 3.11 and Section 3.12.

3.3.3. Drug–Disease Interactions

Because nateglinide is mainly renally eliminated, it has to be used with caution in patients with severe renal impairment.

Repaglinide is contraindicated in the following situations: diabetes mellitus type 1, C-peptide negative; diabetic ketoacidosis, with or without coma; and severe hepatic function disorder with concomitant use of gemfibrozil [92].

3.4. Dipeptidyl Peptidase-4 Inhibitors

Dipeptidyl peptidase-4 (DPP-4) inhibitors (“gliptins”) are oral antidiabetic agents used in the treatment of T2DM.

DPP-4 is an exopeptidase that rapidly inactivates the incretin hormones GLP-1 and GIP, thereby reducing insulin secretion and impairing glycemic control. By inhibiting DPP-4, gliptins increase postprandial GLP-1 concentrations by two- to threefold, resulting in glucose-dependent stimulation of insulin secretion and suppression of glucagon release, without intrinsic hypoglycemic activity [93]. They have little or no clinically meaningful effect on food intake [94].

This class includes sitagliptin, saxagliptin, linagliptin, and alogliptin, which are approved in both the United States and Europe, while vildagliptin is authorized only in Europe. Several DPP-4 inhibitors are available only in Asian markets, such as gemigliptin, teneligliptin, anagliptin, evogliptin, omarigliptin, and trelagliptin [93].

Major clinical benefits of DPP-4 inhibitors include a modest but clinically meaningful improvement in glycemic control, with reductions in HbA1c of approximately 0.5–1.0%, together with lower fasting and postprandial glucose levels. Their glucose-dependent mechanism of action is associated with a low risk of hypoglycemia, particularly when compared with sulfonylureas. In addition, they are generally weight-neutral and do not promote weight gain [95,96].

DPP-4 inhibitors are a generally safe and well-tolerated antidiabetic drug class. They do not seem to confer any significant cardiovascular benefit for patients with T2DM and do not seem to be associated with a significant risk for any major cardiac arrhythmias, except for atrial flutter [97].

3.4.1. Drug Interactions Involving CYP Enzymes and Drug Transporters

With the exception of saxagliptin, which undergoes metabolism predominantly via hepatic cytochrome P450 (CYP) 3A4/5, DPP-4 inhibitors are generally not dependent on CYP-mediated pathways. Only a small fraction of alogliptin and sitagliptin (approximately 10–20%) is metabolized by hepatic enzymes, yielding metabolites with minimal or no pharmacological activity [98].

Saxagliptin differs from the other DPP-4 inhibitors in that approximately half of the absorbed parent drug is metabolized by CYP3A4/5 to 5-hydroxysaxagliptin, an active metabolite that is also a selective and reversible DPP-4 inhibitor, although with about half the potency of saxagliptin. Consequently, coadministration with strong CYP3A4/5 inhibitors can significantly alter exposure to both saxagliptin and its active metabolite, making dose adjustment necessary [98,99].

The interaction between saxagliptin and CYP3A4 substrates or inhibitors—simvastatin (substrate), diltiazem (moderate inhibitor), and ketoconazole (strong inhibitor)—was investigated in healthy subjects. Saxagliptin AUC increased only slightly with simvastatin, more substantially with diltiazem, and most prominently with ketoconazole, while exposure to its CYP3A4/5-generated active metabolite, 5-hydroxysaxagliptin, decreased in parallel. Despite these pharmacokinetic changes, saxagliptin was generally safe and well tolerated, and no clinically meaningful interaction was considered likely with simvastatin or diltiazem. Accordingly, dose adjustment is not required when saxagliptin is coadministered with a CYP3A4 substrate or a moderate CYP3A4 inhibitor. In contrast, coadministration with a strong CYP3A4 inhibitor such as ketoconazole significantly increases saxagliptin exposure and therefore warrants reduction of saxagliptin to the lowest clinical dose [100,101].

Rifampicin, a potent CYP3A4 inducer, reduced saxagliptin exposure, but increased active metabolite concentration, and DPP-4 inhibition was not meaningfully affected. Therefore, the interaction was not considered clinically relevant, and no dose adjustment was required [102].

These findings support the need for dose reduction when saxagliptin is coadministered with strong CYP3A4/5 inhibitors.

No DPP-4 inhibitor has been shown to inhibit or to induce hepatic CYP-mediated drug metabolism [98,103].

Several DPP-4 inhibitors are substrates of drug transporters, including sitagliptin, saxagliptin, linagliptin, and alogliptin. Saxagliptin and linagliptin are substrates of P-gp, and alogliptin is handled in part by OATP [102,104,105]. Among these, sitagliptin, a substrate of P-gp and OAT3, is one of the best characterized, with transporter-mediated renal excretion being particularly well studied. Sitagliptin is also a substrate for intestinal P-gp, but coadministration with cyclosporine, a potent P-glycoprotein inhibitor increased sitagliptin exposure in a non-clinically significant way [106].

According to these pharmacokinetic data, dose adjustment is generally not required when gliptins are coadministered with other drugs in patients with T2DM. An exception is saxagliptin, for which a lower daily dose is recommended when used together with strong CYP3A4/5 inhibitors.

3.4.2. Risk of Angioedema with DPP-4 Inhibitors and Angiotensin-Converting Enzyme Inhibitors

DPP-4 is one of the key enzymes involved in the degradation of bradykinin and substance P. Inhibition of DPP-4 reduces the breakdown of these peptides, leading to their accumulation. Elevated levels of bradykinin, a potent vasodilator, increase vascular permeability and may result in angioedema [107]. Postmarketing reports have identified angioedema as a rare but clinically relevant adverse effect of DPP-4 inhibitors, sometimes occurring early after treatment initiation, even after the first dose. These events may be part of broader hypersensitivity reactions, which can also include anaphylactoid responses and severe cutaneous reactions such as Stevens–Johnson syndrome. In clinical trials, saxagliptin was associated with a slightly higher incidence of hypersensitivity-related events, including angioedema, compared with placebo [108].

Pharmacovigilance studies have shown that the risk of angioedema is increased in patients receiving DPP-4 inhibitors together with angiotensin-converting enzyme (ACE) inhibitors, particularly with the alogliptin/lisinopril combination, and especially in women and older adults. A slight increase has also been observed with the concomitant use of DPP-4 inhibitors and angiotensin receptor blockers, which likewise act on the renin–angiotensin–aldosterone system [109].

Additionally, clinical cases have reported severe angioedema in patients with a prior history of ACE inhibitor–induced angioedema who were started on DPP-4 inhibitor therapy [110].

Available data suggest a clinically relevant interaction, so this combination should be used with caution, especially in patients at higher risk, with appropriate clinical monitoring when DPP-4 inhibitors are prescribed together with ACE inhibitors.

3.4.3. DPP-4 Inhibitors Adjustment in Renal and Hepatic Impairment

DPP-4 inhibitors are widely used in patients with T2DM, including those with chronic kidney disease (CKD) or hepatic impairment, but require different dose adjustment strategies depending on their primary route of elimination.

Most DPP-4 inhibitors (sitagliptin, saxagliptin, alogliptin, and vildagliptin) are predominantly eliminated via the kidneys and require dose reduction according to eGFR or creatinine clearance to avoid accumulation (

Table 3. Regulatory recommendations for the use of DPP-4 inhibitors across stages of renal and hepatic impairment [113,114,115,116,117,118,119,120,121].

DPP4 Inhibitors	Regulatory Agency	Renal Insufficiency 1				Hepatic Insufficiency
		Mild (CrCl 60–89)	Moderate (CrCl 30–59)	Severe (CrCl 15–29)	ESRD (CrCl < 15)	Mild/Moderate	Severe
Sitagliptin	EMA	100 mg/day	CrCl 45–59: 100 mg/day CrCl 30–44: 50 mg/day	25 mg/day	25 mg/day	100 mg/day	Limited data; not recommended
	FDA	100 mg/day	CrCl 45–59: 100 mg/day CrCl 30–44: 50 mg/day	25 mg/day	25 mg/day	100 mg/day	Limited data; not recommended
Saxagliptin	EMA	5 mg/day	CrCl 45–59: 5 mg/day CrCl 30–44: 2.5 mg/day	2.5 mg/day	Not recommended	5 mg/day	Not recommended
	FDA	5 mg/day	CrCl 45–59: 5 mg/day CrCl 30–44: 2.5 mg/day	2.5 mg/day	2.5 mg/day	5 mg/day	5 mg/day
Vildagliptin	EMA	50 mgx2/day	CrCl 51–59: 50 mgx2/zi CrCl 30–50: 50 mg/day	50 mg/day	50 mg/day	Not recommended	Not recommended
	FDA	-	-	-	-	-	-
Alogliptin	EMA	25 mg/day	CrCl 51–59: 25 mgx2/zi CrCl 30–50: 12.5 mg/day	6.25 mg/day	6.25 mg/day	25 mg/day	Limited data; not recommended
	FDA	25 mg/day	12.5 mg/day	6.25 mg/day	6.25 mg/day	25 mg/day	Limited data; not recommended
Linagliptin	EMA	5 mg/day	5 mg/day	5 mg/day	5 mg/day	5 mg/day	Limited data
	FDA	5 mg/day	5 mg/day	5 mg/day	5 mg/day	5 mg/day	5 mg/day

¹ CrCl values are expressed in mL/min; ESRD—end-stage renal disease.

). In contrast, linagliptin is eliminated almost exclusively via the biliary/intestinal route and does not require dose adjustment in renal impairment, as it does not accumulate even in patients with advanced chronic kidney disease [111,112].

Although dose adjustment of most DPP-4 inhibitors in renal impairment is well known, it is still not consistently applied in everyday practice. In a large retrospective cohort of 82,332 patients with T2DM and chronic kidney disease, inappropriate dosing remained common: about 40% of patients received a non-adjusted dose during 2009–2011, and although this proportion decreased over time, it was still 24.4% in 2015. More importantly, inappropriate dosing was associated with poorer outcomes, including a higher risk of emergency department visits, severe hypoglycemia, and death. These findings show that renal dose adjustment is not only a formal recommendation, but an important measure for improving the safety of DPP-4 inhibitor therapy [122].

In patients with mild or moderate hepatic impairment, DPP-4 inhibitors can generally be used without dose adjustment. This applies to sitagliptin, saxagliptin, and alogliptin, while linagliptin appears to be the least affected by liver dysfunction. In severe hepatic impairment, available data are limited for most agents, so their use is usually not recommended or should be considered with caution (Table 3).

3.5. Glucagon-like Peptide-1 Receptor Agonists 1 Receptor Agonists

Glucagon-like peptide-1 receptor agonists (GLP-1 RAs), including the dual agonist tirzepatide, which also targets the glucose-dependent insulinotropic polypeptide (GIP) receptor, are peptide-derived analogues that were initially developed for the treatment of T2DM [123]. In addition to highly effective glycemic control, with significant reductions in HbA1c for all representatives, GLP-1 RAs also induce consistent weight loss. Liraglutide, semaglutide, and tirzepatide are approved for this indication, with the latter producing substantial weight reductions of up to approximately 20% of initial body weight in some obesity studies [124].

GLP-1RAs offer benefits beyond glycemic control, including cardiovascular, renal, neuroprotective, and hepatic effects [125]. Evidence shows reduced major cardiovascular events and mortality in T2DM, with semaglutide also lowering cardiovascular risk in overweight or obese patients without diabetes [126]. These agents slow kidney disease progression, support neuroprotection in neurodegenerative disorders, and semaglutide is approved for chronic kidney disease and metabolic dysfunction-associated steatohepatitis (MASH) based on clinical trial outcomes [127,128].

These drugs mimic the action of incretin hormones (GLP-1 and GIP), enhancing glucose-dependent insulin secretion, suppressing glucagon release during hyperglycemia. They also target receptors in the brain and gastrointestinal tract, reducing appetite, slowing gastric emptying, and postponing glucose absorption [129,130].

3.5.1. GLP-1RAs and Drug Absorption: Role of Gastric Emptying

GLP-1 RAs slow gastric emptying, which delays the movement of food and oral medications from the stomach into the small intestine. As the small intestine is the primary site of drug absorption, this slower transit can affect the timing and, in some cases, the extent of absorption of orally administered medications [131].

These changes may have clinical consequences for medicines with a narrow therapeutic index or for those requiring the rapid achievement of effective concentrations, including digoxin, warfarin, and oral contraceptives. This type of interaction has also been investigated for medicines commonly co-administered in patients with T2DM, such as statins, antihypertensive agents, and analgesics, particularly paracetamol/acetaminophen, which is often used as a marker of gastric emptying. Exenatide is a short-acting exendin-4-based peptide (typically given twice daily) that is mainly eliminated by the kidneys [132]. Exenatide tends to slow the absorption of several oral drugs, with lower Cmax and delayed Tmax, while AUC usually remains unchanged or is only slightly affected. For warfarin, digoxin, lovastatin, lisinopril, and paracetamol/acetaminophen, these changes were not considered clinically relevant, so dose adjustment is generally not needed [133,134,135,136,137].

However, for drugs whose efficacy depends on rapid absorption, such as oral contraceptives, it is preferable to administer them at least one hour before exenatide, to avoid a possible delay or reduction in therapeutic response [138].

Lixisenatide, similar to exenatide, is a short-acting exendin-4-derived peptide, mainly eliminated by the kidneys, so renal function is an important consideration in clinical use [139]. It mainly decreased Cmax and prolonged Tmax, while overall exposure remained largely preserved for drugs such as paracetamol and ramipril [140,141,142].

In the case of warfarin, although Tmax was markedly delayed, this had no relevant pharmacodynamic impact, as INR was not meaningfully affected [143].

Liraglutide is a long-acting, once-daily injectable peptide based on human GLP-1, with sustained effects due to albumin binding and protection from DPP-4 degradation. Compared with short-acting agents, it has a weaker and less timing-dependent effect on gastric emptying, so clinically relevant absorption interactions with oral drugs are less likely. It is mainly cleared through proteolytic metabolism, rather than through a single renal elimination pathway [144,145].

Liraglutide slightly delayed the absorption of several oral drugs, with reduced Cmax and delayed Tmax for atorvastatin, lisinopril and digoxin, while griseofulvin showed an increased Cmax. However, AUC changed only minimally for all four drugs and was not considered clinically relevant [146].

For oral contraceptives, unlike short-acting GLP-1 RAs such as exenatide, liraglutide caused only a modest reduction in Cmax and a slight delay in Tmax, without a relevant reduction in overall bioavailability [147].

Dulaglutide is a human-analog GLP-1RA administered once weekly. It has 90% homology with human GLP-1 RA and is covalently linked to a modified human IgG4-Fc heavy chain by a small peptide linker. Due to its large Fc-linked molecular design, dulaglutide undergoes nonspecific protein metabolism (catabolism to amino acids) and has minimal renal clearance [148]. Evidence from clinical studies and pharmacokinetic modelling showed that dulaglutide generally reduced Cmax and delayed Tmax for several oral drugs, including digoxin, warfarin, atorvastatin, combined oral contraceptives, lisinopril, metformin, metoprolol, paracetamol, and sitagliptin, in line with its effect on gastric emptying, while overall exposure remained largely unchanged. The main exception was atorvastatin, for which AUC decreased modestly. For warfarin, INR changes were minimal and not clinically meaningful. Overall, both the observed clinical data and pharmacokinetic estimates indicate that dulaglutide does not produce clinically relevant interactions with oral small-molecule drugs [149,150].

Semaglutide is a human-analog GLP-1RA with C18 acylation and a half-life of about 7 days, enabling once-weekly administration; it is available both as an injectable formulation and as the first oral GLP-1RA formulation. Semaglutide is eliminated via general proteolytic metabolism, with metabolites excreted mainly in urine and secondarily in feces [151].

Semaglutide did not meaningfully affect the bioavailability of combined oral contraceptives, with ethinylestradiol exposure preserved and levonorgestrel exposure slightly increased. In studies with metformin, warfarin, atorvastatin, and digoxin, Cmax and overall exposure were not changed to a clinically relevant extent, and warfarin’s INR response was unaffected. Pharmacokinetic modelling showed some slowing of absorption for paracetamol and atorvastatin, but these changes were also considered clinically insignificant [152,153,154].

For two substances, potentially significant interactions with semaglutide were reported. Alectinib is an ALK inhibitor used in ALK-positive non-small cell lung cancer and has been associated with early, persistent weight gain. For this reason, its potential interaction with semaglutide was investigated. In a small crossover study, coadministration significantly reduced alectinib exposure, with lower AUC, Cmax, and Cmin, and fewer patients remaining above the proposed efficacy threshold. These findings suggest a clinically relevant interaction, so concomitant use should be approached cautiously, with monitoring of alectinib plasma concentrations when possible [155].

Additionally, a case series reports three patients on stable lithium therapy who started semaglutide: in two cases, lithium concentrations rose markedly and caused toxicity despite stable renal function and no medication changes, while in the third, preventive lithium dose reductions helped limit toxicity but levels still increased more than expected. Possible contributors include dehydration and delayed gastric emptying [156].

Oral semaglutide contains the GLP-1 analog semaglutide and sodium N-(8-[2-hydroxybenzoyl] amino) caprylate (SNAC), an absorption enhancer that helps semaglutide be taken up after oral administration.

This combination was generally not associated with clinically relevant interactions with most co-administered drugs. It did not meaningfully affect lisinopril, warfarin, digoxin, omeprazole, or combined oral contraceptives, while modest increases in exposure were observed for metformin, furosemide, and rosuvastatin, without clinical relevance [157,158,159].

However, levothyroxine exposure increased by about 33%, suggesting that thyroid parameters should be monitored when the two drugs are used together [160].

Tirzepatide is a once-weekly dual GIP/GLP-1 RA, the first in its class, with higher potency at the GIP receptor and superior efficacy when compared with single GLP-1RAs (greater HbA1c reduction and greater weight loss). It is an acylated peptide with a C20 fatty-acid chain, a half-life of about 5 days, and elimination mainly via proteolytic metabolism and renal excretion of peptide metabolites, with no urinary excretion of the intact molecule [161].

Like other GLP-1 RAs, this dual agonist also delayed gastric emptying, as shown by changes in paracetamol absorption, but did not affect overall exposure [162].

However, tirzepatide may meaningfully affect oral contraceptive absorption. A single 5 mg dose reduced both the AUC and Cmax of ethinyl estradiol and norgestimate, while delaying Tmax, suggesting a potentially relevant interaction for contraceptive efficacy [163].

In conclusion, administration of GLP-1RAs, including tirzepatide, slows gastric emptying, with direct consequences for the absorption kinetics of concomitantly administered drugs. In most clinical studies, the typical effect consists of a reduction in Cmax and a delay in Tmax, while the AUC generally remains unchanged or is only minimally affected. This suggests that overall systemic drug exposure is not significantly altered and, in most cases, these changes are not of major clinical relevance. However, delayed absorption may have implications for drugs for which a rapid onset of action is essential, like oral contraceptives, potentially reducing initial therapeutic effectiveness.

3.5.2. Effects of GLP-1RAs on CYP Enzymes and Drug Transporters

GLP-1 RAs and tirzepatide are large peptide-based agents that are not dependent on hepatic cytochrome P450 metabolism, being predominantly degraded through proteolytic pathways. As a result, the potential of conventional drug–drug interactions driven by CYP enzyme inhibition or induction is considered low [164,165]. Evidence for clinically relevant enzyme-mediated drug–drug interactions involving GLP-1 RAs is limited. In vitro interaction available for liraglutide, semaglutide, and tirzepatide showed minimal inhibition or induction of CYP enzymes [164,166].

Available in vitro data suggest that semaglutide and tirzepatide have a low potential for clinically relevant transporter-mediated interactions like OATP, OAT, OCT, P-gp, BCRP, and MATE [164].

3.5.3. Interaction Between GLP-1 RAs and General Anesthetics

GLP-1 RAs do not classically interact with general anesthetics (GAs), by altering their pharmacokinetic or pharmacodynamic parameters. The interaction is indirect, through changes in gastric physiology and glucose metabolism.

The most relevant mechanism is the delay in gastric emptying, which may increase the risk of residual gastric contents at the time of anesthesia induction. This is particularly important in the context of general anesthesia, where delayed gastric emptying may increase the risk of regurgitation and pulmonary aspiration [167].

A meta-analysis of several randomized and nonrandomized studies showed that GLP-1 RAs are associated with a higher incidence of pre-procedural gastrointestinal symptoms compared with controls. In addition, patients receiving these agents had a significantly increased likelihood of elevated residual gastric content despite adhering to standard fasting recommendations. Semaglutide use was associated with an increased risk of elevated residual gastric content, with approximately a fivefold higher likelihood compared with controls. Rare cases of pulmonary aspiration were reported even after adequate fasting, highlighting potential clinical relevance in the perioperative settings [168].

In addition, GLP-1 RAs modulate glucose homeostasis by enhancing insulin secretion and suppressing glucagon release in a glucose-dependent manner. While this profile is generally associated with a low risk of hypoglycemia, perioperative factors such as reduced oral intake, stress response, and concomitant antidiabetic therapy may alter glycemic control. As a result, both hyperglycemia and hypoglycemia remain potential concerns in the perioperative period [169].

Current recommendations suggest that perioperative use of GLP-1 RAs should be guided by shared decision-making between the patient and the care team, taking into account both the metabolic benefits and the individual risk profile. The risk of delayed gastric emptying and aspiration appears to be higher in certain situations, particularly during the dose-escalation phase, at higher doses, with weekly formulations, and in patients presenting with gastrointestinal symptoms such as nausea, vomiting, or dyspepsia. In patients without increased risk, GLP-1 RAs may generally be continued. When risk is considered unacceptable, the decision to withhold treatment should be individualized, balancing aspiration risk against the potential metabolic consequences of discontinuation, such as hyperglycemia. If treatment interruption is deemed necessary, current practice is to withhold daily formulations on the day of surgery and weekly formulations for one week before surgery, while still reassessing all patients on the day of the procedure for symptoms suggestive of delayed gastric emptying [170].

3.5.4. GLP-1 RA Adjustment in Renal Impairment

Most GLP-1 RAs, as well as tirzepatide, have a favorable profile in renal impairment, with no dose adjustment required across mild to moderate and even severe stages for the majority of agents (

Table 4. Regulatory recommendations for the use of GLP-1 RAs across stages of renal impairment [171,172,173,174,175,176,177,178,179,180,181,182,183,184].

GLP-1RA	Regulatory Agency	Mild to Moderate (CrCl 30–89 *)	Severe (CrCl 15–29 *)	ESRD (CrCl < 15 *)
Exenatide IR	EMA	CrCl 50–80: No adjustment CrCl 30–50: Cautious escalation (5→10 mcg)	Not recommended	Not recommended
	FDA	No adjustment	Not recommended	Not recommended
Exenatide ER	EMA	No adjustment	Not recommended	Not recommended
	FDA	CrCl ≥ 46: No adjustment CrCl < 45: Not recommended	Not recommended	Not recommended
Lixisenatide	EMA	No adjustment	Limited data; not recommended	Limited data; not recommended
	FDA	No adjustment	No adjustment; use with caution	Limited data; not recommended
Liraglutide	EMA	No adjustment	No adjustment.	Limited data; not recommended
	FDA	No adjustment	No adjustment.	Limited data; use with caution
Dulaglutide	EMA	No adjustment	No adjustment	Limited data; not recommended
	FDA	No adjustment	No adjustment	No adjustment
Semaglutide	EMA	No adjustment	No adjustment	Limited data; use with caution
	FDA	No adjustment	No adjustment	No adjustment
Tirzepatide	EMA	No adjustment	No adjustment	No adjustment; use with caution
	FDA	No adjustment	No adjustment	No adjustment

* CrCl values are expressed in mL/min; ESRD—end-stage renal disease.

). This can be explained by their mainly proteolytic metabolism and limited renal clearance, which reduces the impact of declining kidney function on systemic exposure.

The main exception is exenatide, particularly the extended-release formulation, which shows the most restrictive recommendations. This reflects its greater dependence on renal elimination, leading to accumulation in patients with reduced renal function and explaining why its use is not recommended below certain CrCl thresholds. For lixisenatide, recommendations are more cautious, especially in severe renal impairment and end-stage renal disease (ESRD), where data are limited.

In contrast, liraglutide, dulaglutide, semaglutide, and tirzepatide demonstrate consistent guidance across both EMA and FDA, generally supporting use without dose adjustment, although caution or limited experience is noted in end-stage renal disease.

Differences between EMA and FDA recommendations are relatively minor and mainly reflect variations in the interpretation of limited clinical data, particularly in ESRD.

3.6. Sodium–Glucose Cotransporter 2 Inhibitors

Sodium–glucose cotransporter 2 (SGLT2) inhibitors are oral antidiabetic agents that inhibit glucose and sodium reabsorption in the proximal renal tubule. As a result, glycosuria and natriuresis appear, with a reduction of HbA1c by approximately 0.5% to 1% in patients with T2DM. Their glucose-lowering effect is independent of insulin and is associated with weight loss [185] and a moderate decrease in systolic blood pressure [186]. SGLT2 inhibitors manifest favorable hemodynamic, metabolic, and renoprotective effects [187,188,189].

Large randomized clinical trials have demonstrated the cardiovascular and renal benefits of SGLT2 inhibitors. They reduce heart failure hospitalizations [190] and cardiovascular mortality in selected populations with T2DM [191], and slow the progression of chronic kidney disease in patients with and without diabetes [192]. SGLT2 inhibitors have become cornerstone cardio-renal therapies due to their pleiotropic effects [193], being widely used in heart failure and chronic kidney disease as part of guideline-directed medical therapy in complex, multi-drug regimens. As their pharmacokinetic drug–drug interactions appear to have limited clinical significance [194], understanding their potential drug–drug interactions, including pharmacodynamic and pharmacotoxicologic interactions, is important.

Currently, four SGLT2 inhibitors are approved by EMA for use in Europe: canagliflozin, dapagliflozin, empagliflozin, and ertugliflozin [195]. In the United States, the FDA has also approved bexagliflozin and sotagliflozin, the latter being a dual SGLT2 and SGLT1 inhibitor. Several other agents (ipragliflozin, luseogliflozin, and tofogliflozin) are approved in Asian markets, predominantly in Japan [194].

3.6.1. DDIs with Pharmacokinetic Mechanisms

In vitro studies indicate that SGLT2 inhibitors undergo limited oxidative metabolism via cytochrome P450 enzymes. The predominant metabolic pathway for most compounds in this class is glucuronidation by uridine diphosphate–glucuronosyltransferases (UGTs). UGT1A9 and UGT2B4/2B7 are key enzymes involved in the metabolism of dapagliflozin, canagliflozin, and empagliflozin [196,197]. At concentrations substantially higher than those achieved at therapeutic doses, some SGLT2 inhibitors are weak inhibitors of multiple CYP isoforms, including CYP3A4, CYP2C9, and CYP2D6, making clinically relevant CYP-mediated drug–drug interactions less likely [194]. In clinical studies, co-administration of SGLT2 inhibitors with strong CYP inhibitors or inducers, as well as with UGT modulators such as rifampicin or mefenamic acid, resulted in modest changes in systemic exposure that did not necessitate dose adjustment, with the exception of canagliflozin. These observations are consistent with the minor contribution of CYP-mediated oxidation to overall clearance and the presence of multiple parallel metabolic pathways [194,198].

In vitro data show that SGLT2 inhibitors may interact with several efflux and influx transporters, most notably P-glycoprotein (P-gp or MDR1), breast cancer resistance protein (BCRP), and organic anion transporters (OAT1 and OAT3); organic anion transporting polypeptides (OATPs) are involved in a more limited manner. Some compounds within this class have demonstrated inhibitory effects on these transporters, generally observed at concentrations substantially exceeding those achieved under therapeutic dosing [194].

The high protein binding of SGLT2 inhibitors contributes to their long half-lives (roughly 12–17 h), which supports once-daily dosing [199].

Canagliflozin is metabolized predominantly via UGT1A9 and UGT2B4-mediated glucuronidation, forming the major metabolites M7 and M5 [200]. Canagliflozin weakly inhibits CYP2B6, CYP3A4, CYP2C8, and CYP2C9, with clinically irrelevant interactions with warfarin or simvastatin [194].

Canagliflozin is a substrate of the MDR1, BCRP, and MRP2 efflux transporters. It inhibits several organic cation transporters (OCTs). For OCT1 (IC₅₀ = 5.2 μM), it is the most potent inhibitor in the class of SGLT2 inhibitors; it also inhibits OCT2 (44 μM), explaining the 20% increase in metformin AUC, whose renal clearance depends on OCT2-mediated active tubular secretion [194].

Probenecid, as a UGT inhibitor, increases canagliflozin AUC by 21% [198]. Enzyme inducers (St. John’s wort, rifampicin, barbiturates, phenytoin, fosphenytoin, carbamazepine, primidone, ritonavir, efavirenz) have a higher risk of interaction with canagliflozin, being able to reduce canagliflozin exposure. Rifampicin reduces the AUC of canagliflozin by 51% and Cmax by 28%, potentially compromising efficacy [198]. With concomitant UGT inducers, glycemic control should be monitored. If a UGT inducer is used in patients tolerating canagliflozin 100 mg daily and requiring stricter glycemic control, dose increase to 300 mg daily may be considered if they have an eGFR ≥ 60 mL/min/1.73 m², or, if eGRF is lower, another antihyperglycemic strategy should be considered [198,200].

Cholestyramine may reduce exposure; administer canagliflozin ≥1 h before or 4–6 h after bile acid sequestrants [200].

Canagliflozin 300 mg daily for seven days increases digoxin AUC 20% and Cmax 36% via P-gp inhibition [201]. Patients receiving digoxin or other cardiac glycosides (digitoxin) and canagliflozin should be monitored. Canagliflozin 300 mg daily for six days increases simvastatin AUC by 12% and Cmax by 9%, and simvastatin acid AUC by 18% and Cmax by 26%; these increases are considered not clinically relevant [202]. Intestinal BCRP inhibition cannot be excluded; canagliflozin may increase exposure to BCRP substrates (certain statins like rosuvastatin, some anticancer drugs) [200].

Dapagliflozin is metabolized predominantly via UGT1A9-mediated glucuronidation, forming the inactive metabolite M15 (3-O-glucuronide). Minor pathways involve CYP1A1, CYP1A2, CYP2A6, CYP2C9, CYP2D6, CYP2E1, and CYP3A4. In vitro, dapagliflozin does not significantly inhibit CYP1A2, CYP2A6, CYP2B6, CYP2C8, CYP2C9, CYP2C19, CYP2D6, or CYP3A4 at therapeutic concentrations (IC₅₀ > 45 μM); it does not induce CYP1A2, CYP2B6, or CYP3A4 [203,204].

Clinical studies show dapagliflozin pharmacokinetics are unchanged by metformin, pioglitazone, sitagliptin, glimepiride, voglibose, hydrochlorothiazide, bumetanide, valsartan, or simvastatin. Rifampicin reduces systemic exposure by 22% without clinically significant effect on 24 h urinary glucose excretion; no dose adjustment is needed [205]. Mefenamic acid, a UGT1A9 inhibitor, increases systemic exposure by 55% and Cmax by 58%, without clinical impact on glucosuria; no dose adjustment is required [203,205].

Multikinase inhibitor sorafenib inhibits UGT1A7 and UGT1A9; an experiment in rats showed that sorafenib increased the AUC of dapagliflozin by 1.66 times and the Cmax by 1.45 times; conversely, dapagliflozin reduced the AUC of sorafenib by approximately 50.5% [206]. Patients treated with sorafenib for hepatocellular carcinoma who also receive dapagliflozin should be closely monitored.

Empagliflozin has a metabolic peculiarity: the metabolism contributes less than 10% to total clearance, empagliflozin being predominantly eliminated unchanged by the kidneys. Glucuronidation is achieved by UGT2B7, UGT1A3, UGT1A8, and UGT1A9. Inhibition of all major CYP isoenzymes is negligible.

Paradoxically, rifampicin, although an inducer of UGT, increases the AUC of empagliflozin by 35%, an effect explained by the inhibition of hepatic OATP1B1 and OATP1B3 uptake transporters. Gemfibrozil and probenecid increase exposure by 58% and 53%, respectively, by combined inhibition at the UGT and transporter level [194,207].

Empagliflozin is a substrate of OAT3, OATP1B1, OATP1B3, P-gp, and BCRP, but not of OAT1 or OCT2. Probenecid increases empagliflozin AUC by 53% by combined OAT/OATP inhibition. Both DDIs are not deemed clinically relevant [207]. Drug interactions involving CYP450 or UGT are unlikely [208].

Ertugliflozin is the most sensitive to enzyme-mediated interactions among the agents approved in Europe, because it is metabolized predominantly by UGT1A9 and UGT2B7 [196].

At therapeutic concentrations, ertugliflozin does not inhibit any CYP or UGT isoenzymes, and neither the parent compound nor its metabolites inhibit major drug transporters at clinically relevant concentrations, thereby conferring a clean profile as a perpetrator of drug–drug interactions [194].

Rifampicin reduced ertugliflozin AUC by 39% and Cmax by 15% in clinical settings, but this decrease is not clinically relevant, and dose adjustment is not recommended [209].

The impact of UGT inhibitors on ertugliflozin pharmacokinetics has not been clinically studied. Potential exposure increases from UGT inhibition are not considered clinically relevant [210].

No longer approved in Europe, sotagliflozin received FDA approval in 2023 for reducing cardiovascular death, hospitalization for heart failure, and urgent heart failure visits in adults with T2DM, chronic kidney disease, and additional cardiovascular risk factors [195].

Sotagliflozin is metabolized predominantly via UGT1A9. It is a substrate of OAT3, OATP1B1, and OATP1B3. A clinically relevant feature is its major metabolite, M19 (sotagliflozin-3-O-glucuronide), which induces CYP3A4, inhibits CYP2D6, and modulates the activity of the OATP1B1/1B3 and MRP2 transporters [194].

This dual profile of enzyme inductor and transporter inhibitor makes sotagliflozin the agent with the highest potential for metabolite-mediated drug–drug interactions within the class [194].

Rifampicin, a UGT and CYP inducer, reduces sotagliflozin AUC by 60% and Cmax by 40%, potentially decreasing efficacy. If enzyme inducers (rifampicin, phenytoin, phenobarbital, ritonavir) are administered concomitantly, frequent glycemic monitoring is recommended [211].

Sotagliflozin 400 mg increases digoxin AUC by 27% and Cmax by 52% via P-gp inhibition [212]. Patients receiving concomitant digoxin require appropriate monitoring [211].

3.6.2. Pharmacodynamic DDIs Common to the Class of SGLT2 Inhibitors

SGLT2 inhibitors and diuretics

Prolonged combined exposure to SGLT2 inhibitors and loop diuretics (bumetanide, furosemide, torsemide) or thiazide diuretics (hydrochlorothiazide, indapamide, chlopamide) is associated with an increased risk of volume depletion-related adverse events, including hypotension, dehydration, hypokalemia, and acute kidney injury [14]. Additionally, it is important that volume depression should be corrected before initiating SGLT2 inhibitors [213].

Mechanistically, this moderate risk interaction is pharmacodynamic in nature and results from the additive natriuretic and diuretic effects of SGLT2 inhibition and loop diuretics. SGLT2 inhibitors promote osmotic diuresis and proximal tubular natriuresis by inhibiting sodium–glucose reabsorption, while loop diuretics block sodium reabsorption in the thick ascending limb of the loop of Henle. In susceptible populations, such as older patients or those with impaired renal function, this synergistic intravascular volume contraction may compromise renal perfusion and exacerbate hemodynamic instability [214].

SGLT2 inhibitors improve diuretic efficiency, and allow diuretic dose reduction in heart failure, including some patients with diuretic resistance [215].

SGLT2 inhibitors and agents that decrease blood pressure

SGLT2 inhibitors produce modest additional blood pressure reductions when combined with standard antihypertensives, without significant symptomatic hypotension in most patients with T2DM. Combining ARNI with dapagliflozin, commonly used in HF, may increase the risk of hypotension due to additive vasodilatory and natriuretic effects, especially in acutely ill or frail patients [216].

SGLT2 inhibitors and lithium

SGLT2 inhibitors may increase renal excretion of lithium, potentially reducing serum lithium levels and decreasing its effectiveness. Close monitoring of serum lithium levels is recommended when starting or adjusting SGLT2 inhibitors. However, some studies suggest SGLT2 inhibitors might help mitigate long-term lithium-associated kidney dysfunction [217].

SGLT2 inhibitors and statins

Case reports and pharmacologic reasoning suggest an interaction, even though pharmacokinetic studies in healthy subjects showed no clinically significant statin level changes during SGLT2 inhibitor co-administration. Canagliflozin is the SGLT2 inhibitor with the highest interaction potential, possibly due to pharmacokinetic or pharmacodynamic mechanisms [218]. An analysis of FAERS conducted between 2013–2021 found no increased risk of myotoxicity (myopathy or rhabdomyolysis) with concomitant use of SGLT2 inhibitors and statins. Initial signals for simvastatin and empagliflozin drug combinations disappeared after adjustment, suggesting the combination is safe in this regard [219].

SGLT2 inhibitors and valproate

SGLT2 inhibitors can increase exposure to valproate, likely through interference with hepatic glucuronidation mediated by UGT, because valproate is glucuronidated by UGT1A6, UGT1A9, UGT2B7. Increased levels of valproate may raise the risk of hepatotoxicity, thrombocytopenia, and other dose-dependent adverse effects. Monitoring of valproatemia is recommended. A case report presents an indirect interaction in which valproate toxicity is potentiated by the weight loss induced by dapagliflozin [220].

SGLT2 inhibitors and intravenous ferrum

In a post hoc exploratory analysis of a randomized clinical trial, a trend toward a greater increase in hemoglobin was observed with ferric derisomaltose in iron-deficient patients receiving an SGLT2 inhibitor at baseline. SGLT2 inhibitors may modulate intracellular iron availability when co-administered with intravenous iron, potentially enhancing cellular iron uptake. The implied mechanism is hypothesized to involve alterations in mitochondrial oxidative metabolism and hepcidin regulation [221].

Other DDIs reported in case studies or in pharmacovigilance activities

Dapagliflozin and IL-17 inhibitors (secukinumab, ixekizumab)

Case reports and pharmacovigilance data suggest a pharmacodynamic interaction whereby dapagliflozin-induced glycosuria and IL-17 inhibition potentiate each other, increasing the risk of urogenital fungal infections. Rare but severe cases, including Candida pyelonephritis, have been described [222].

Dapagliflozin and linezolid

Case series reported the occurrence of severe pancytopenia in patients treated concomitantly, sometimes with irreversible or fatal progression. The mechanism is not clearly established, but the association suggests a possible additive effect on myelosuppression and requires careful hematological monitoring [223].

Other DDIs between SGLT2 inhibitors and agents affecting glucose plasmatic levels will be addressed in Section 3.11 and Section 3.12.

3.6.3. Drug–Disease Interactions

Urinary tract infections, pyelonephritis and genital infections

SGLT-2 inhibitors are associated with an increased risk of genital mycotic infections, but no overall increase in urinary tract infection (UTI) risk, with higher doses of dapagliflozin showing an increased risk. Glucosuria makes urogenital infections more likely by promoting microbial growth. Additionally, elevated glucose concentrations in urine and genital tissues promote adhesion and proliferation of microorganisms (especially Candida), explaining the higher incidence of genital infections compared with urinary tract infections [224]. A systematic review and meta-analysis of randomized controlled trials showed that canagliflozin, dapagliflozin, and empagliflozin increased genital infection risk but only dapagliflozin raised UTI risk [225].

A pharmacovigilance analysis based on the FAERS database investigated potential drug–drug interaction signals associated with SGLT2 inhibitor therapy in relation to urinary tract infections and pyelonephritis. Disproportionality analyses identified several antidiabetic agents, including thiazolidinediones, DPP-4 inhibitors, glinides, and the alpha-glucosidase inhibitor voglibose, as being associated with an increased reporting risk of these infections when co-administered with certain SGLT2 inhibitors [226]. In addition, several statins, as well as ezetimibe, eicosapentaenoic acid ethyl, and cholestyramine, were associated with increased reporting of urinary tract infections or pyelonephritis when used concomitantly with SGLT2 inhibitors, with statins representing the most consistently implicated drug class [226].

SGLT2 inhibitors and chronic renal impairment

Dapagliflozin shows the greatest sensitivity to renal dysfunction, with increases in AUC of approximately +45%, +100%, and +200% in mild, moderate, and severe renal impairment, respectively. Sotagliflozin follows a similar pattern, with AUC increases of about +70% and +170%. Ertugliflozin and empagliflozin exhibit more moderate exposure increases (approximately 60–70% and up to 66%), while canagliflozin shows the smallest changes (15–53%) [194].

Due to their renal mechanism of action, renal impairment may decrease their antidiabetic efficacy. Therefore, they are not recommended for use to improve glycemic control in patients with an eGFR below a certain threshold, as they are likely to be ineffective; in such cases, alternative or additional antidiabetic therapies should be considered. These thresholds are 45 mL/min/1.73 m² for dapagliflozin and ertugliflozin and 30 mL/min/1.73 m² for empagliflozin [227,228,229].

When renal impairment occurs, both the efficacy and elimination of SGLT2 inhibitors are affected in a stage-dependent manner; dose adjustment or discontinuation should be guided by the eGFR thresholds specified in the prescribing information of each individual agent (

Table 5. Regulatory recommendation for the use of SGLT2 inhibitors across stages of renal impairment [200,203,208,210,211,227,228,229,230,231].

SGLT2 Inhibitor	Regulatory Agency	Mild to Moderate (eGFR/CrCl 30–89 *)	Severe (eGFR/CrCl 15–29 *)	ESDR (eGFR/CrCl < 15 */Dialysis)
Canagliflozin	EMA (Invokana)	eGFR 60–89: 100 mg/day; may titrate to 300 mg. eGFR 30–59: 100 mg	Do not initiate. Patients already on Invokana: continue 100 mg/day (applies when albumin/creatinine ratio > 300 mg/g).	Do not initiate. Patients already on Invokana: continue 100 mg/day until dialysis or renal transplantation.
	FDA (Invokana)	eGFR ≥ 60: 100 mg once daily; may increase to 300 mg for additional glycemic control eGFR 30 to <60: maximum recommended dose 100 mg once daily	Limitation of use: not recommended for glycemic control if eGFR < 30 mL/min/1.73 m2. Initiation not recommended. Patients already on Invokana with albuminuria > 300 mg/day may continue 100 mg once daily to reduce risk of ESKD, doubling of serum creatinine, CV death, and HF hospitalization (indication: renal/CV protection).	Initiation not recommended. Patients already on Invokana with albuminuria > 300 mg/day may continue 100 mg once daily to reduce risk of ESKD, doubling of serum creatinine, CV death, and HF hospitalization (indication: renal/CV protection).
Dapagliflozin	EMA (Forxiga)	No dose adjustment based on renal function. eGFR ≥ 45: 10 mg/day (T2DM, HF, CKD). eGFR 30–44: not recommended for T2DM (reduced efficacy). HF: 10 mg; CKD: continue if eGFR ≥ 25.		Not recommended for T2DM or CKD initiation. HF: 10 mg/day; no lower eGFR cutoff specified for HF indication.
	FDA (Farxiga)	eGFR ≥ 45: 5 mg/day for T2DM glycemia (may increase to 10 mg); 10 mg/day for all other indications (HF, CKD, CV risk). eGFR 30–44: 10 mg/day (HF, CKD, CV risk; not recommended in T2DM).	eGFR 25– < 45: 10 mg once daily (all indications). eGFR < 25: do not initiate; may continue 10 mg/day if already on treatment (to reduce risk of eGFR decline, ESKD, CV death, or HF hospitalization).	ESRD/dialysis: Do not initiate for T2DM. May continue 10 mg/day for ongoing HF/CKD benefit.
Empagliflozin	EMA (Jardiance)	eGFR ≥ 45: 10 mg/day; may titrate to 25 mg. eGFR 30–44: 10 mg/day only (do not increase to 25 mg); T2DM glycemic efficacy reduced; probably absent at eGFR < 30.	Not recommended for T2DM glycaemia (efficacy probably absent at eGFR < 30). HF & CKD indications: 10 mg/day if eGFR ≥ 20.	Do not initiate if eGFR < 20. HF/CKD: not recommended if eGFR < 20.
	FDA (Jardiance)	10 mg once daily; may increase to 25 mg for T2DM. Not recommended for glycemic control if eGFR < 30.	Not recommended for T2DM glycemia (eGFR < 30). Insufficient data for T2DM or CVD with eGFR < 30 Insufficient data for HF with eGFR < 20.	Contraindicated in dialysis patients.
Ertugliflozin	EMA (Steglatro)	eGFR ≥ 45 to <60: initiate at 5 mg/day; may increase to 15 mg. eGFR < 45: do not initiate. Discontinue if eGFR falls persistently below 30.	Do not initiate. If already on treatment and eGFR falls persistently <30: discontinue.	Not recommended in severe renal impairment, ESRD, or dialysis.
	FDA (Steglatro)	Initiate at 5 mg once daily; may increase to 15 mg. Not recommended if eGFR < 45 mL/min/1.73 m2.	Not recommended.	Not recommended.
Sotagliflozin	EMA (Zynquista)	Marketing authorization no longer valid.
	FDA (Inpefa)	CV death/HF hospitalization reduction in HF or T2DM+CKD+CVD. Not approved for T2DM glycemia. 200 mg/day before main meal; may titrate to 400 mg.		Not recommended (not studied at eGFR < 25 or on dialysis).

* All renal thresholds are eGFR (mL/min/1.73 m², CKD-EPI). CKD—chronic kidney disease; CVD—cardiovascular disease; ESKD—end-stage kidney disease; HF—heart failure; T2DM—type 2 diabetes.

).

SGLT2 inhibitors in hepatic impairment

Dapagliflozin and empagliflozin show moderate exposure increases (up to 67% and 75% in severe hepatic impairment), canagliflozin exhibits only minor changes (up to 28%), ertugliflozin demonstrates modest effects (Cmax +21%), whereas sotagliflozin shows a pronounced 3- to 6-fold increase in exposure in moderate-to-severe hepatic impairment. Therefore, dapagliflozin and empagliflozin are usually not recommended in severe hepatic impairment (Child–Pugh class C), while sotagliflozin use is not recommended in Child–Pugh class B or C [194].

SGLT2 inhibitors in heart failure

In heart failure, exposure to dapagliflozin is approximately 1.2-fold higher, while for empagliflozin trough concentrations are 1.47–1.53 times higher than those observed in patients with T2DM, without clinical significance [194].

SGLT2 inhibitors and euglycemic diabetic ketoacidosis (DKA), especially in perioperative period or during prolonged fasting

Euglycemic DKA associated with SGLT2 inhibitors and ketogenic diets represents a metabolic acidosis and ketogenesis despite normal or only mildly elevated blood glucose (<250 mg/dL). SGLT2 inhibition lowers insulin and increases glucagon promoting a shift toward lipolysis and hepatic ketogenesis [187]. The risk is amplified by ketogenic or very low-carbohydrate diets, prolonged fasting, surgery, or acute illness, often with delayed recognition due to the absence of marked hyperglycemia [232,233]. SGLT2 inhibitors should be stopped at least 3 days before elective surgery (4 days for ertugliflozin [229]) to reduce the risk of euglycemic DKA.

Concerns about lower limb amputations

The signal for lower limb amputation is most strongly documented for canagliflozin (CANVAS trial: HR 1.97), while a subsequent meta-analysis of seven RCTs did not confirm a class-level increase in risk [234]. It is presumed that the risk is either limited to canagliflozin, which could contribute through volume depletion, hemoconcentration, and reduced peripheral perfusion, or dependent on pre-existing risk factors (peripheral vascular disease, neuropathy, ulcers) [229,234].

SGLT2 inhibitors and electrolyte balance

SGLT2i therapy may affect electrolyte balances (small increases in serum potassium and magnesium concentrations) and may be associated with bone fractures. SGLT2i may increase serum phosphate, fibroblast growth factor 23, and parathyroid hormone and decrease 1,25-dihydroxyvitamin D [235].

3.7. Acarbose

Acarbose is an alpha-glucosidase inhibitor (AGI) which acts locally by competitively inhibiting intestinal alpha-glucosidases, thus delaying the digestion and absorption of complex carbohydrates and thereby reducing postprandial glucose levels [236]. Acarbose increases undigested carbohydrates in the terminal portions of the small intestine and in the proximal colon, which leads to the stimulation of GLP-1 secretion, with a supplementary hypoglycemic effect [237]. Acarbose is one of the few pharmacological interventions proven to reduce the progression to T2DM in patients with impaired glucose tolerance [238,239].

3.7.1. Pharmacokinetic Interactions Leading to Decreased Serum Levels of Co-Administered Drugs

As its systemic absorption is minimal, acarbose’s DDIs appear at gastrointestinal level. Acarbose may decrease the absorption of digoxin [240]; monitoring digoxin plasma level and its effects is warranted. Acarbose may also impair the absorption of valproic acid [236], decrease oral metronidazole exposure in diabetic and healthy subjects [241], and have occasional effects on warfarin anticoagulation; close monitoring is necessary after initiating acarbose [242].

3.7.2. Pharmacodynamic Interactions Leading to Increased Hypoglycemic Effect

Acarbose pharmacodynamic DDIs are common to other antidiabetics and will be dealt with in Section 3.11 and Section 3.12.

3.7.3. DDIs Leading to Decreased Hypoglycemic Effect

Digestive enzymes may reduce acarbose’s effect because they directly cleave the undigested oligosaccharides and starch that acarbose leaves in the gut lumen. This counteracts the core mechanism of acarbose, reducing its postprandial glucose-lowering effect. It is generally recommended to avoid or cautiously manage the combined use of pancreatin and amylase-containing products with acarbose. Intestinal absorbents (caolin, activated charcoal, colestyramine) can adsorb acarbose in the gastrointestinal tract and reduce its local availability [236].

3.7.4. Other DDIs and Drug–Disease Interactions

Because flatulence and diarrhea are common adverse effects, acarbose should be used with caution prior to bowel surgery [243]. Baicalein, quercetin, and luteolin inhibit alpha-glucosidase and exert synergistic effects when combined with it, allowing for the reduction of the acarbose dose [244].

3.8. Thiazolidinediones

Thiazolidinediones (TZDs), or glitazones, are a class of insulin sensitizers used in the treatment of T2DM. TZDs act by activating peroxisome proliferator-activated receptor gamma (PPAR-γ), a nuclear receptor involved in the regulation of genes related to glucose and lipid metabolism. Through this effect, they improve insulin sensitivity in adipose tissue, muscle, and liver [245,246]. Two glitazones, rosiglitazone and pioglitazone, are approved in the United States; however, in Europe, the marketing authorization for rosiglitazone was withdrawn in 2010, mainly due to concerns regarding an increased risk of cardiovascular events [247].

TZDs improve insulin sensitivity and provide a moderate glycemic benefit in patients with T2DM, lowering fasting glucose (about 40 mg/dL), fasting insulin and HbA1c (≈ 1%). However, as monotherapy, they are generally less effective than metformin or sulfonylureas in reducing blood glucose [248].

With regard to lipid effects, head-to-head studies have shown a more favorable profile for pioglitazone than for rosiglitazone. Pioglitazone was associated with a reduction in triglyceride levels, whereas rosiglitazone led to an increase. Both drugs increased LDL and HDL cholesterol, but rosiglitazone produced a greater rise in LDL cholesterol, while pioglitazone was associated with a more pronounced increase in HDL cholesterol [249].

TZDs are associated with several adverse effects that should be considered in clinical use. These include an increased risk of heart failure, and for rosiglitazone, a higher incidence of myocardial infarction. Other common effects include weight gain with fat redistribution, peripheral edema, and an increased risk of bone fractures, particularly in postmenopausal women. Rare adverse events include macular edema and liver injury [250].

3.8.1. DDIs Involving CYP Enzymes

Because they are metabolized through cytochrome P450 enzymes, glitazones are prone to pharmacokinetic interactions. Pioglitazone is mainly metabolized by CYP2C8 and CYP3A4 (to a lesser extent), whereas rosiglitazone is primarily metabolized by CYP2C9 and slightly by CYP2C8 [251,252].

Rifampicin, a potent inducer of cytochrome P450 enzymes, significantly reduces exposure to TZDs, with a decrease in AUC of approximately 65% for rosiglitazone and 54% for pioglitazone, along with a shortening of their elimination half-life and a possible reduction in antidiabetic efficacy. These effects are consistent with induction of CYP2C8, CYP3A4 and to a lesser extent CYP2C9, the main enzymes involved in TZD metabolism. Therefore, coadministration with rifampicin should be approached with caution [253,254].

Gemfibrozil, a strong CYP2C8 inhibitor, markedly increases exposure to TZDs by reducing their metabolism. Pioglitazone exposure increased about threefold, while rosiglitazone exposure more than doubled, with prolonged half-life in both cases. This may enhance their therapeutic effects but also increase the risk of dose-related adverse reactions, so coadministration requires caution and possible dose adjustment [255,256,257,258].

Similarly, trimethoprim, another CYP2C8 inhibitor, produced a more modest increase in rosiglitazone exposure and prolonged its half-life. This further supports the role of CYP2C8 in rosiglitazone metabolism and suggests that CYP2C8 inhibitors may increase rosiglitazone levels [259].

Rosiglitazone and pioglitazone do not appear to meaningfully alter the pharmacokinetics of co-administered drugs, suggesting that their potential to act as clinically relevant perpetrators of drug–drug interactions is low [251].

3.8.2. Pharmacodynamic DDIs

Both pioglitazone and rosiglitazone were associated with a higher frequency of edema in clinical trials. The reported incidence was around 3.0–7.5% with TZDs, compared with 1.0–2.5% in patients receiving placebo or other oral antidiabetic treatment. The highest rates were seen when TZDs were combined with insulin: edema was reported in 15.3% of patients receiving insulin plus pioglitazone and in 14.7% of those treated with insulin plus rosiglitazone, versus 7.0% and 5.4%, respectively, in the corresponding insulin-only groups. Clinical improvement was observed only after TZD therapy was discontinued [260].

Other interactions between glitazones and other antidiabetic agents will be discussed in Section 3.11 and Section 3.12.

3.8.3. Drug–Disease Interactions

Use of TZDs in patients with heart failure

TZDs are associated with an increased risk of heart failure in patients with T2DM. This risk was observed both in randomized trials and in observational studies, and heart failure typically developed after several months of treatment rather than immediately after initiation. The effect was seen at both low and high doses and was not restricted to elderly patients, indicating that TZD-related heart failure is a clinically relevant adverse effect across a broad patient population [261]. Both rosiglitazone and pioglitazone carry a boxed warning for heart failure, reflecting a class effect. They can cause or worsen fluid retention and should not be used in patients with NYHA class III–IV heart failure. Careful monitoring for weight gain, edema, and dyspnea is recommended during treatment [262,263].

TZDs adjustment in renal and hepatic impairment

In general, TZDs do not require dose adjustment in renal impairment, as kidney function has little influence on their pharmacokinetics. Pioglitazone can usually be used without dose modification in patients with reduced renal function (ClCr > 4 mL/min), No information is available from dialyzed patients, so pioglitazone should not be used in such patients. Similarly, rosiglitazone does not show clinically relevant pharmacokinetic changes in renal impairment [262,263,264].

The situation is different in hepatic impairment. Rosiglitazone should not be started in patients with active liver disease or ALT levels above 2.5 times the ULN, and liver enzymes should be checked before treatment and monitored thereafter [264]. Pioglitazone has been associated with rare cases of hepatic dysfunction and should not be used in patients with hepatic impairment (patients with baseline ALT levels >2.5× ULN or evidence of liver disease). Liver function tests should be performed before initiating therapy and monitored periodically thereafter. If liver enzymes rise during treatment (ALT >3× ULN) or if symptoms suggestive of hepatic injury occur, therapy should be reassessed and discontinued if abnormalities persist or if jaundice develops [262,264].

3.9. Insulin and Insulin Analogs

Insulin binds to its membranal receptors, which, after autophosphorylation, recruit intracellular proteins insulin receptor substrates 1 and 2 (IRS1/IRS2), through which two downstream pathways are activated. At the metabolic pathway level, the GLUT4 transporter is translocated to the cell membrane, allowing glucose to enter the cell, increasing glycogen synthesis, decreasing hepatic gluconeogenesis, and inhibiting lipolysis. On the mitogenic pathway, insulin regulates gene transcription, cell growth, and survival [265].

There are five pharmacokinetic classes in modern insulin therapy. Rapid-acting analogs (lispro, aspart, glulisine, and the newer ultra-rapid faster-aspart insulin) have onset within 5–15 min and cover prandial glucose excursions. Short-acting regular insulin (onset in about 30 min) is mainly used IV and in premixed products. Intermediate NPH has a pronounced peak and a duration of around 12 h. Long-acting basal analogs (glargine, detemir, and degludec) provide a near-peakless profile over 18–42 h. Fixed-ratio combinations (FRCs) associate a basal analogue with a GLP-1 RA; IDegLira (degludec + liraglutide) and iGlarLixi (glargine + lixisenatide) allow intensification of treatment with a lower risk of hypoglycemia and less weight gain [266].

3.9.1. Pharmacokinetic DDIs of Insulin

Insulin itself has relatively few classic pharmacokinetic drug–drug interactions as it is a peptide and is not metabolized by CYP enzymes.

The few clinically important interactions appear at the absorption site. Hyaluronidase co-injection depolymerizes the extracellular matrix and accelerates insulin absorption, shortening its latency [267]. Insulin absorption after subcutaneous administration can be favored by vasodilation. Propranolol, a β-blocker with peripheral vasoconstrictor properties, reduced subcutaneous insulin absorption, while the vasodilator nifedipine increased it [268]. Local heating (exercise, sauna, hot baths) or massage could accelerate insulin absorption. Careful attention should be paid to not injecting insulin in an area affected by lipohipertrophy, a common adverse effect of insulin therapy, as more fibrous tissue and a poorer blood supply can delay insulin absorption, reduce bioavailability and increase pharmacokinetic variability [269].

3.9.2. Pharmacodynamic DDIs of Insulin

Insulin has many pharmacodynamic DDIs with other antidiabetic agents, which will be discussed in Section 3.11 and Section 3.12.

3.10. Pramlintide

Pramlintide is a soluble, stable synthetic amylin analogue that decreases post-prandial glucagon release and slows gastric emptying. It is used only as an adjunct to mealtime insulin in patients with type 1 and T2DM [270].

3.10.1. Drug Interactions Due to Gastric Emptying Delaying

In a placebo-controlled, single-blind, crossover clinical study run in 24 patients with T2DM, 120 micrograms of pramlintide delays peak plasma concentrations of paracetamol by up to 1.2 h and reduces its peak levels by 14–29%, without changing total exposure (AUC) [271]. Rapid-onset oral medications (e.g., analgesics, contraceptives, and antibiotics) should be administered at least 1 h before or 2 h after pramlintide, given its gastric emptying–delaying effect [272].

Pramlintide may impair the efficacy of drugs that stimulate gastric motility [273], such as prokinetic agents (metoclopramide, itopride, prucalopride).

3.10.2. Pharmacodynamic Drug Interactions Leading to Increased Effects of Pramlintide

Added to insulin therapy, pramlintide produces additional reductions in HbA1c, plasma mean 24 h postprandial glucose level, and total daily insulin dose, concomitantly increasing the incidence of nausea, vomiting, anorexia, and hypoglycemia [274]. The additive glucose-lowering effect of pramlintide increases the risk of severe hypoglycemia when used with mealtime insulin, particularly early in therapy; therefore, the insulin dose should be reduced by approximately 50% at initiation [272].

Despite pH incompatibility, mixing pramlintide with insulin has not shown clinically significant PK/PD effects [275]; however, separate injections are recommended by the manufacturer [272].

Sharing partially overlapping mechanisms of action, the combined use of a GLP-1 RA and an amylin analogue results in significant weight loss while maintaining glycemic control [276].

Due to its central effects on satiety and body weight, pramlintide acts synergistically with other anti-obesity drugs. In a randomized, double-blind, controlled clinical trial, the association of pramlintide to metreleptin (a recombinant methyl-human leptin) demonstrated synergistic potentiation of weight loss [277]. The mechanism involves restoration of leptin sensitivity, which is reduced in obesity, through complementary actions of amylin at the brainstem level [278]. A multicenter randomized placebo-controlled trial demonstrated that adding sibutramine or phentermine to pramlintide therapy enhanced weight loss compared with pramlintide monotherapy with an acceptable safety profile [279].

3.11. Antidiabetics’ Interactions with Agents with Blood Glucose-Lowering Effects

Hypoglycemia commonly occurs as a consequence of antidiabetic therapy, in insulin overdosing, improperly timed SU use, or GLP-1 RA administration. Receiving less systematic attention is the parallel pharmacology occurring when patients on antidiabetics are co-prescribed drugs from entirely unrelated therapeutic categories that happen to lower blood glucose independently. These interactions can range from mild potentiation, easily managed by blood glucose monitoring, to life-threatening synergistic hypoglycemia requiring emergency intervention.

3.11.1. Therapeutic Useful Drug–Drug Interactions

While metformin monotherapy is still the first choice of antidiabetic treatment in patients without cardiorenal or metabolic issues, in other patients diabetes management is initiated with SGLT2 inhibitors or GLP-1 RAs. However, evidence that early combination therapy produces more durable glycemic control and better β-cell preservation accumulates [10,280,281].

Modern antidiabetics are largely used in monotherapy or in association with other hypoglycemic agents in patients with significant comorbidities. A network meta-analysis of 453 trials including monotherapies (134 trials), add-on to metformin-based therapies (296 trials), and monotherapies versus add-on to metformin therapies (23 trials) showed that insulin regimens and specific GLP-1 RAs added to metformin-based therapy produced the greatest reductions in hemoglobin A1c level [280]. Another network meta-analysis of 764 trials (n = 421,346) focused on the addition of SGLT-2 inhibitors and GLP-1 RA to existing diabetes therapy. SGLT-2 inhibitors were superior for heart failure, while GLP-1 RA appeared superior for stroke prevention [282].

Combining agents with complementary hypoglycemic mechanisms leads to additive or supra-additive decreases of plasma glucose concentrations, often with additional cardiorenometabolic benefits, while minimizing the risk of hypoglycemia. Examples of such synergistic mechanisms are the simultaneous use of metformin and SGLT2 inhibitors, metformin and GLP-1 RA, or SGLT2 inhibitors and DPP4 inhibitors. The efficacy of SGLT2 inhibitors + GLP-1 RA dual therapy (without metformin) was recently confirmed in a meta-analysis of more than 1.16 million patients showing an RR of 0.56 for MACE vs. either monotherapy. In order to improve medication adherence, fixed drug combinations are frequently used [10].

Combining basal insulin with a GLP-1RA may improve glycemic control and reduce weight in T2DM patients [283]. SGLT2 inhibitors combined with insulin decrease the blood pressure, adverse cardiovascular outcomes, and visceral adipose tissue volume [284]. Nevertheless, these types of combinations pose a high risk of hypoglycemia.

Concomitant administration of SGLT2 inhibitors and insulin or insulin secretagogues (SUs) results in a pharmacodynamic interaction due to additive glucose-lowering effects. While this combination improves glycemic control and may allow insulin dose reduction, it increases the risk of hypoglycemia and, in the setting of relative insulin deficiency, may contribute to euglycemyc DKA [285]. If such associations are necessary, it is recommended to reduce the insulin dose by 10–20%, and the SUs dose by 50% [286].

The risk of lactic acidosis after metformin is higher in patients with severe dehydration or acute renal failure precipitated by SGLT2 inhibitors. In patients at renal risk, monitoring of renal function is necessary [287].

A meta-analysis that evaluated the drug interaction profiles of alpha-glucosidase inhibitors (acarbose, miglitol, voglibose) concluded that they do not alter the pharmacodynamic profiles of other antidiabetics such as metformin, vildagliptin, or SGLT2 inhibitors [288]. Enhanced glucose-lowering effects occur when acarbose is used concomitantly with insulin or SUs [242].

Insulin and glitazones have complementary pharmacodynamic effects: insulin increases insulin availability, whereas glitazones improve insulin sensitivity. This combination may enhance glycemic control, but it also clearly increases the risk of fluid retention, edema, and heart failure [289].

3.11.2. Drug–Drug Interactions Causing Potentially Dangerous Hypoglycemia

DDIs inside the class of antidiabetic agents

Co-prescribed antidiabetic agents can amplify adverse drug reactions, most prominently hypoglycemia. The association of SUs with alpha-glucosidase inhibitors, DPP4 inhibitors, GLP-1 RAs, SGLT2 inhibitors, or TZDs bears a high risk of hypoglycemia, warranting careful monitoring or even therapy modification.

GLP-1 RAs may increase hypoglycemic effects of meglitinides. Meglitinide doses should be reduced when used in combination with GLP-1 agonists, particularly when also used with basal insulin.

DDIs of antidiabetic agents with other agents with blood glucose-lowering effects

The following DDIs possess a high risk of hypoglycemia, warranting active monitoring and potential dose adjustment:

• Co-trimoxazole with sulfonylureas. Co-trimoxazole, or the combination trimethoprim–sulfamethoxazole (TMP-SMX) acts synergistically with SUs to reduce glycemia by a double mechanism. Sulfamethoxazole, a sulfonamide antibacterial, acts as a direct blocker of ATP-sensitive potassium (KATP) channels on pancreatic β-cells, triggering insulin releases, while trimethoprim inhibits CYP2C9, the principal enzyme responsible for metabolic clearance of many SUs (glibenclamide, glipizide, glimepiride). A fourfold increase in emergency room visits for hypoglycemia was reported in patients combining sulfonylureas with co-trimoxazole compared with those who used amoxicillin [290]. Reducing sulfonylurea dose or choosing an alternative antibiotic should be envisaged.
• Fluoroquinolones with secretagogues. Fluoroquinolones block the KATP channels in pancreatic β-cells, mimicking the pharmacology of SUs. Older patients are more susceptible to severe hypoglycemia when using fluoroquinolones, especially levofloxacin. Monitor blood glucose and consider temporary insulin substitution in high-risk patients. Other antimicrobials are also associated with hypoglycemia: tigecycline, ertapenem, and clarithromycin [291].

The following DDIs indicate a moderate risk of hypoglycemia, calling for monitoring and occasional dose adjustment:

• Tramadol and methadone in patients on insulin or SUs. By activating μ-opioid receptors, tramadol and methadone reduce hepatic gluconeogenesis; complementarily, tramadol inhibits the reuptake of norepinephrine and serotonin, modulating sympathoadrenal glucose counter-regulation. It is necessary to increase the frequency of self-monitoring when starting or increasing the dosage of antidiabetic medications [292].
• SSRIs and antidiabetic agents. A recent review has reported that SSRIs produce alterations in glucose homeostasis, with hypoglycemic episodes and loss of consciousness [293].
• Sunitinib and other tyrosine kinase inhibitors (TKIs) in diabetics. Sunitinib inhibits PDGFR, c-KIT, and other kinase pathways, improving insulin sensitivity and reducing gluconeogenesis. Dasatinib and imatinib exhibit a similar, albeit less pronounced, pattern. Standard practical approaches consist in blood sugar monitor during active treatment cycles and antidiabetic dose reduction [294].
• Testosterone replacement in hypogonadal diabetics. Androgens promote lean body mass and muscle glucose uptake, testosterone deficiency being associated with insulin resistance and elevated HbA1c in men. Testosterone replacement therapy improves glycemic control, with potential hypoglycemia in patients already well-managed on insulin or SUs. Glucose levels should be monitored within 4–8 weeks after testosterone initiation and reducing antidiabetic dose is warranted [295]
• Mifepristone and antidiabetics. Mifepristone is a glucocorticoid receptor antagonist used in hyperglycemic patients with Cushing’s syndrome. If the doses of insulin or SUs concomitantly used are not reduced once glucocorticoid-driven hyperglycemia improves due to the administration of mifepristone, there is a risk of hypoglycemia [296].

Monitoring is recommended due to the lower but non-negligible risk of hypoglycemia associated with the following DDIs:

• Somatostatin analogues. Somatostatin inhibits secretion of both insulin and glucagon from the pancreatic islet. First-generation somatostatin (SST) receptor ligands (octreotide, lanreotide) and the newer selective SST2 agonist paltusotine appear to suppress glucagon and growth hormone more than insulin, impairing glucagon counter-regulation to hypoglycemia. Antidiabetic agent doses should be revisited after initiation somatostatin analogs [297].
• MAOIs with insulin. MAOIs reduce norepinephrine and epinephrine turnover, and thus the adrenergic counter-regulation that mediates the recovery from hypoglycemia is impaired. Insulin doses should be reassessed, while patients should be counselled on blunted adrenergic symptoms of hypoglycemia (tachycardia, tremor). Moclobemide, a reversible MAO-A inhibitor, appears to carry a lower hypoglycemic risk than classic MAOI (phenelzine, tranylcypromine, isocarboxazid), but vigilance is still warranted [298].
• Bitter melon, fenugreek, gymnema, ginseng, aloe, neem, and other “antidiabetic” botanicals enhance the glucose-lowering effect [299].

Another important DDI is that of antidiabetic agents with beta-blockers. Besides their hypoglycemic effect, they can blunt the adrenergic symptoms of hypoglycemia (tachycardia, agitation). Non-selective beta-blockers, like propranolol, lead to greater hyperglycemia, because they inhibit insulin secretion more strongly. Beta-blockers with intrinsic sympathomimetic activity, such as pindolol, have a lesser impact on insulin sensitivity. Similarly, those with alpha-blocking effects, such as carvedilol, may promote insulin-sensitizing effects through vasodilation, having neutral or mildly positive effects on glycemic control [300].

3.12. Drug Interactions That Can Attenuate Antidiabetic Efficacy

Drugs that lower insulin secretion, increase insulin resistance, or stimulate glucose production can interfere with concomitant antidiabetic agents’ efficacy. Especially in the context of a complex medication regimen, using drugs with hyperglycemic effects in a patient already receiving antidiabetics necessitates consideration, as clinical and metabolic decompensation due to cumulative changes could appear.

3.12.1. Glucocorticoids and ACTH

Glucocorticoids bind the intracellular glucocorticoid receptor (GR), increasing hepatic glucose output. They suppress IRS-1 and GLUT4 translocation in skeletal muscle, thus decreasing glucose peripheral uptake. They also impair Ca²⁺-dependent insulin exocytosis in beta cells. The resulting hyperglycemia is post-prandial and afternoon-dominant, tracking the steroid’s pharmacokinetic profile; fasting glucose monitoring alone is insufficient. NPH insulin co-administered with the morning steroid dose best covers the post-prandial excursion. SUs risk nocturnal hypoglycemia as steroid levels wane; SGLT-2 inhibitors are useful adjuncts in outpatient chronic use [301].

3.12.2. Thiazide and Loop Diuretics

Thiazides (hydrochlorothiazide, indapamide, chlorthalidone) inhibit the renal NaCl cotransporter, causing urinary K⁺ loss. Hypokalemia constitutively opens KATP channels on beta cells, impairing glucose-stimulated insulin secretion. Loop diuretics share the same kaliuretic mechanism via Na⁺/K⁺/2Cl⁻ reuptake inhibition. Volume contraction additionally raises catecholamines, stimulating hepatic glycogenolysis (β₂) and suppressing insulin secretion (α₂). Because the deficit is primarily secretory, SUs or glinides are more mechanistically targeted to correct this hyperglycemia than insulin sensitizers. Correcting hypokalemia often partly reverses the hyperglycemia [302].

3.12.3. Second-Generation (Atypical) Antipsychotics

Second-generation antipsychotics raise blood glucose through three parallel pathways. The H1 receptor blockade in the hypothalamus activates AMPK, driving hyperphagia and visceral adiposity that secondarily causes insulin resistance. They also block M3 muscarinic receptor on beta cells and blunt cholinergic potentiation of glucose-stimulated insulin secretion. D2 receptor blockade in islets may disinhibit glucagon secretion [303]. Clozapine and olanzapine carry the highest hyperglycemic risk, followed by quetiapine, with aripiprazole showing the lowest risk among atypical antipsychotics such as aripiprazole [304]. GLP-1 RAs address both the secretory deficit and H1-mediated hyperphagia; metformin targets insulin resistance. Mandatory metabolic monitoring (weight, fasting glucose, HbA1c, lipids) at baseline and every three months is required for clozapine and olanzapine.

3.12.4. PI3K/AKT/mTOR Pathway Inhibitors

PI3Kα (p110α) inhibition (alpelisib, copanlisib) blocks PIP₂ toPIP₃ conversion, preventing AKT activation, GLUT4 translocation, and glycogen synthesis, thereby unleashing hepatic gluconeogenesis [305]. This is an on-target toxicity inseparable from anti-tumor efficacy. mTORC1 inhibitors (everolimus, sirolimus) produce insulin resistance by impaired GLUT4 trafficking. Metformin could be used to manage the resulting hyperglycemia. Insulin may be required at higher-than-expected doses due to reduced PI3K-dependent signaling [306].

3.12.5. HIV Protease Inhibitors

Indinavir directly and reversibly blocks GLUT4-mediated glucose uptake independently of insulin receptor signaling. Ritonavir impairs IRS-1 phosphorylation and AKT activation upstream of GLUT4. Lopinavir activates the unfolded protein response in beta-cell endoplasmic reticulum, impairing insulin biosynthesis. Concurrent lipodystrophy (visceral fat redistribution) compounds chronic insulin resistance [307].

3.12.6. GnRH Agonists and Androgen Deprivation Therapy

After testosterone suppression, muscle loss, visceral adiposity, reduced adiponectin, and lowered IRS-1 amplitude develop and confer a high risk of incident diabetes. HbA1c should be measured at baseline and at three and six months after androgen deprivation therapy initiation. Resistance exercise disproportionately offsets the insulin resistance. Metformin is a first-line treatment; GLP-1 receptor agonists address the visceral adiposity component [308].

3.12.7. Catecholamines and Sympathomimetics

Epinephrine raises blood glucose rapidly. β₂-receptor activation leads to glycogenolysis via cAMP/PKA. Gluconeogenesis appears after α₁ stimulation; glucose-stimulated insulin secretion is suppressed and glucose peripheral uptake is decreased. The acute hyperglycemia produced by intravenous epinephrine is manageable only with IV insulin [309].

3.12.8. Calcineurin Inhibitors and mTOR Inhibitors (Post-Transplant Diabetes)

Tacrolimus inhibits calcineurin and suppresses insulin gene transcription. Sirolimus inhibits mTORC1, impairing translation of secretory proteins and beta-cell mass maintenance, while simultaneously generating peripheral insulin resistance [310].

3.12.9. Miscellaneous Agents

Diazoxide opens the SUR1/KATP channel on beta cells, completely antagonizing secretagogue therapy. It is used to treat hypoglycemia in hyperinsulinemic situations [311].

Glucagon and dasiglucagon stimulate hepatic glycogenolysis and gluconeogenesis acutely [312].

Statins can modestly worsen hyperglycemia by impairing insulin sensitivity [313].

Moderate risk interactions could manifest with the decrease in therapeutic effects if SGLT2 inhibitors are used concomitantly with alpha-lipoic acid (possibly through its antioxidant effects and protein kinase activating properties [314]).

3.13. Comparative Cross-Class Synthesis

The present analysis of antidiabetic medications’ DDIs shows substantial heterogeneity, both in the mechanism of the interactions and in their clinical magnitude. To facilitate the cross-class comparison central to clinical decision-making,

Table 6. Comparative interaction risk profile across antidiabetic drug classes.

Drug/Drug Class	Predominant PK Interaction Mechanism	Predominant PD Interaction Mechanism	Principal Disease-Related Risk	Indicative Interaction Burden *	Highest-Risk Populations
Metformin	OCT2/MATE substrate (cimetidine, ranolazine, dolutegravir, trimethoprim)	Limited; lactate-metabolism modifiers	Renal impairment; hepatic dysfunction; hypoperfusion	Low-moderate	Severe CKD, decompensated HF, contrast-media exposure, intercurrent acute illness
Sulfonylureas	CYP2C9 substrate (fluconazole, amiodarone, statins); OATP1B1/1B3	Hypoglycemia (broad; antibiotics, ACEi/ARB, β-blockers, SSRIs, alcohol, tramadol)	Hepatic and renal impairment	High	Elderly, CKD, hepatic impairment, polypharmacy, alcohol use
Meglitinides	CYP3A4/CYP2C8 substrate (gemfibrozil + repaglinide is clinically hazardous; clopidogrel; cyclosporine)	Hypoglycemia (less than sulfonylureas)	Hepatic impairment	Moderate	Hepatic impairment, gemfibrozil/clopidogrel co-therapy
DPP-4 inhibitors	Generally minimal; saxagliptin via CYP3A4/5; sitagliptin via OAT3/P-gp	Low; angioedema risk with ACEi	Renal dose adjustment for most agents (linagliptin excepted)	Low	Severe renal impairment without dose adjustment
GLP-1 RAs	Delayed gastric emptying—variable effect on concomitant oral drug absorption (most notably oral semaglutide/levothyroxine, alectinib)	Low	Pre-existing gastrointestinal disease; perioperative aspiration risk	Moderate	Patients on narrow-therapeutic-index oral drugs; perioperative period
SGLT2 inhibitors	Minimal (UGT-mediated metabolism; no major CYP interactions)	Volume depletion with diuretics; ketogenesis with caloric restriction	Renal impairment; urogenital infection; volume depletion; rare DKA	Moderate	Concomitant loop/thiazide diuretics, dehydration-prone patients, ketogenic diet, recurrent UTI/genital mycotic infection
Acarbose	Reduced absorption of digoxin, paracetamol, metronidazole	Additive GI effects	GI disease	Low	Inflammatory bowel disease, prior bowel surgery
Thiazolidindiones	CYP2C8 substrate (gemfibrozil, trimethoprim, rifampicin)	Fluid retention exacerbated by NSAIDs, insulin	Heart failure (contraindication)	Moderate	Patients with HF, edema, or osteoporosis
Insulin	Negligible	Hypoglycemia (broad—overlaps with sulfonylurea interaction set).	Renal, hepatic or cardiac impairment	High (PD)	Polypharmacy, hepatic and renal impairment, perioperative period
Pramlindite	Delayed gastric emptying	Hypoglycemia when combined with insulin	GI disease	Moderate	Concomitant rapid-acting insulin

* The ‘Indicative interaction burden’ column synthesizes the class-by-class evidence presented in Section 3.1, Section 3.2, Section 3.3, Section 3.4, Section 3.5, Section 3.6, Section 3.7, Section 3.8, Section 3.9, Section 3.10, Section 3.11, Section 3.12 and Section 3.13—the range of pharmacokinetic interactions (CYP/transporter substrates), the range of hypoglycemia-enhancing pharmacodynamic interactions, and the presence of disease-related risks. The rating represents an authors’ integrative synthesis intended as a clinical orientation tool, not an externally validated scoring system. ACEi—angiotensin-converting enzyme inhibitor; ARB—angiotensin receptor blocker; CKD—chronic kidney disease; CYP—cytochrome P450; DKA—diabetic ketoacidosis; GI—gastrointestinal; HF—heart failure; MATE—multidrug and toxin extrusion protein; NSAID—non-steroidal anti-inflammatory drug; OAT—organic anion transporter; OATP—organic anion transporting polypeptide; OCT—organic cation transporter; P-gp—P-glycoprotein; PD—pharmacodynamic; PK—pharmacokinetic; SSRI—selective serotonin reuptake inhibitor; UGT—UDP-glucuronosyltransferase; UTI—urinary tract infection.

summarizes, for each class, the predominant pharmacokinetic and/or pharmacodynamic interaction mechanism, the principal disease-related interaction risk, the overall interaction burden in polypharmacy context, and the patient populations at the greatest risk.

Sulfonylureas and insulin carry the highest overall interaction burden, dominated by hypoglycemia-enhancing pharmacodynamic interactions across a broad spectrum of co-medications (antibiotics—particularly co-trimoxazole and fluoroquinolones; ACE inhibitors and angiotensin receptor blockers; beta-blockers; selective serotonin reuptake inhibitors; alcohol; tramadol). For sulfonylureas, this is compounded by clinically significant CYP2C9-mediated pharmacokinetic interactions and by the loss of safety provided by the glucose dependence that characterizes the newer secretagogue-independent classes.

DPP-4 inhibitors, metformin (in eligible patients), and GLP-1 RAs carry the lowest overall interaction burden and are the preferred candidates in patients with extensive polypharmacy. Renal dose adjustment are warranted for several DPP-4 inhibitors, while GLP-1 RAs could lead to gastric-emptying-mediated absorption issues.

SGLT2 inhibitors occupy an intermediate but clinically important position: their pharmacokinetic interaction profile is minimal, but they carry distinctive disease-related interaction risks (volume depletion in patients on diuretics, euglycemic ketoacidosis under conditions of caloric restriction or acute illness, and urogenital infection in predisposed patients) that require active anticipation rather than passive monitoring.

3.14. Limitations

A key limitation of this review is the significant lack of available data, as most drug–drug interaction studies have been conducted in healthy volunteers over short durations. Consequently, evidence regarding clinical relevance in patients with polypharmacy and multiple comorbidities remains limited.

4. Conclusions

The pharmacotherapy of T2DM has become increasingly complex in the context of expanding therapeutic options, multimorbidity, and polypharmacy. As highlighted in this review, we have organized and classified DDIs and drug disease interactions so that they can be more easily identified, avoided, monitored or taken into account, to increase the treatment efficacy and safety. The cross-class analysis presented in this review supports three priority actions for the prescribing clinician.

First, systematic medication reconciliation should be performed at every patient encounter, with focused attention on three categories of co-medication that account for the majority of clinically meaningful interactions identified above: CYP2C9 and CYP3A4/CYP2C8 modulators (relevant for SUs, meglinides, saxagliptin, and TZDs); OCT/MATE substrates and inhibitors (relevant for metformin); and drugs predisposing to volume depletion or to the alteration of lactate, glucose, or ketone homeostasis (relevant for SGLT2 inhibitors, metformin, and the secretagogues).

Second, monitoring strategies must be class-specific rather than generic. Renal function and lactate-relevant co-medication require active surveillance for metformin. Glycemic vigilance and the avoidance of hypoglycemia-enhancing co-medications are central for SUs, meglinides, and insulin. Volume status, urinary symptoms, and ketone surveillance, particularly under conditions of caloric restriction, acute illness, or perioperative fasting, are essential for SGLT2 inhibitors. Gastrointestinal tolerance and the timing of concomitant oral drug absorption (most notably for narrow-therapeutic-index agents and oral contraceptives) require attention with GLP-1 RAs and pramlintide. Cardiovascular and hepatic monitoring remain priorities for TZDs.

Third, agent selection in patients with extensive polypharmacy or significant multimorbidity should preferentially favor drugs with the lowest overall interaction burden (DPP-4 inhibitors, metformin in eligible patients, and GLP-1 RAs) provided that glycemic target and cardiovascular benefits are achievable with these agents. SUs and insulin, although still effective and inexpensive, should be deployed with the understanding that their interaction profile is the broadest of any antidiabetic class and that hypoglycemia risk increases substantially in patients receiving multiple potentially interacting co-medications or with comorbidities affecting drug clearance or counter-regulatory mechanisms.

Therefore, it is important that the treatment of patients with T2DM must be monitored and periodically evaluated by a multidisciplinary team which includes a diabetologist, specialists managing the patient’s other chronic conditions, a clinical pharmacist who can identify potential drug interactions and optimize medication choices, but also a nutritionist who can adapt the patient’s diet according to their individual needs. Pharmacist-led medication review is one direct mechanism by which the interactions discussed in this review can be identified and prevented in routine clinical practice; the integration of pharmacology expertise into the diabetes care team and the routine use of regulatory product information and validated DDI databases at the point of prescribing should be considered standard practice in the management of patients with T2DM and concurrent multimorbidity.