Drug-Drug Interaction Profile of INVOKANA

Summary

• Caution or dose adjustments need to be considered when INVOKANA is coadministered with diuretics, angiotensin-converting enzyme inhibitors, angiotensin receptor blockers, potassium-sparing diuretics, insulin, insulin secretagogues, UDP-glucuronosyltransferase (UGT) enzyme inducers (eg, rifampin, phenytoin, phenobarbital, ritonavir), and/or digoxin. Patients should be monitored appropriately.¹
• Canagliflozin had no clinically relevant effects on the pharmacokinetics (PK) of probenecid or cyclosporine A²; individual components of an oral contraceptive (OC; containing ethinyl estradiol [EE] and levonorgestrel [LN]) or warfarin³; glyburide, simvastatin, or metformin⁴; hydrochlorothiazide (HCTZ)⁵; or teneligliptin⁶.
• Canagliflozin did not induce cytochrome P450 (CYP450) enzyme expression (3A4, 2C9, 2C19, 2B6, and 1A2) in cultured human hepatocytes. Canagliflozin did not inhibit the CYP450 isoenzymes (1A2, 2A6, 2C19, 2D6, or 2E1) and weakly inhibited CYP2B6, CYP2C8, CYP2C9, and CYP3A4 based on in vitro studies with human hepatic microsomes.¹
• Canagliflozin is a weak inhibitor of P-glycoprotein (P-gp). Canagliflozin is also a substrate of the drug transporters P-gp and multidrug resistance-associated protein-2 (MRP2).¹
• A comprehensive Food and drug administration (FDA) Adverse Event Reporting System (FAERS) analysis found that canagliflozin was strongly associated with osteomyelitis (97.66%; n=2509). A drug-drug interaction analysis between sodium-glucose cotransporter-2 inhibitors (SGLT2is) and metformin revealed a negative signal (Ω shrinkage method: Ω=-1.067, Ω025=-1.370, χ=-7.120), suggesting a potential protective role. Concurrent use of canagliflozin with metformin was associated with a reduced risk of osteomyelitis.⁷
• An analysis of FAERS data showed that coadministration of SGLT2is with nephrotoxic drugs increased acute renal failure (ARF) risk compared with single-drug therapy, with notable associations observed for dapagliflozin with sulfamethoxazole (adjusted reporting odds ratio [ROR], 6.36; 95% confidence interval [CI], 3.27-23.33), empagliflozin with spironolactone (adjusted ROR, 4.32; 95% CI, 3.14-24.53), canagliflozin with sulfamethoxazole (adjusted ROR, 7.50; 95% CI, 3.24-24.67), and canagliflozin with pregabalin (adjusted ROR, 6.13; 95% CI, 3.66-24.83).⁸
• A study performed using the real-world FAERS database found a significantly higher acute kidney injury (AKI) reporting rate with cotherapy of SGLT2is and ARNI (adjusted ROR 8.04; 95% CI, 6.20-10.42; P<0.001), with the strongest AKI signal for canagliflozin plus angiotensin receptor-neprilysin inhibitor (ARNI; adjusted ROR, 16.82; 95% CI, 3.75-75.57; P<0.001). Cotherapy with SGLT2is and ARNIs significantly increased the AKI reporting rate in patients with diabetes (adjusted ROR, 20.57; 95% CI, 11.93-35.46; P<0.001), those with hypertension (adjusted ROR, 4.30; 95% CI, 1.98-9.37; P<0.001), and those with diabetes plus hypertension (adjusted ROR, 5.44; 95% CI, 1.92-15.43; P=0.001). Combination therapy did not show a significant increase in AKI risk among patients with heart failure.⁹

CLINICAL STUDIES

Phase 1 Studies

Three independent studies in healthy subjects evaluated the effects of rifampin, probenecid, and cyclosporine A on the PK of canagliflozin, investigating potential drug interactions with UGT inducers, UGT inhibitors, and P-gp inhibitors, respectively.²

• Rifampin: A single-center, open-label, fixed-sequence study was conducted to evaluate the effects of steady-state rifampin (600 mg) on the PK of single-dose canagliflozin (300 mg) under a fasting state (N=14).
  • Coadministration of canagliflozin and rifampin decreased the maximum plasma concentration (C_max) and area under the curve (AUC) of canagliflozin by 28% and 51%, respectively, compared with canagliflozin administration alone.
  • Steady-state plasma rifampin concentration increased by 24% following coadministration with canagliflozin under the fasted state (day 10) compared with rifampin administration alone under the fed state (day 9).
• Probenecid: A single-center, open-label, fixed-sequence study was conducted to evaluate the effects of multiple-dose probenecid (500 mg twice daily) on the steady-state PK of canagliflozin (300 mg), with a standardized meal 1 hour after study drug administration (N=14).
  • Coadministration increased canagliflozin C_max during a dosing interval at steady state (C_max.ss) and AUC during a dosing interval at steady state (AUC_τ.ss) by ~13% and 21%, respectively, compared with canagliflozin administration alone. Mean canagliflozin renal clearance (CL_R) was ~34% lower for canagliflozin coadministered with probenecid compared with canagliflozin alone.
• Cyclosporine A: A single-center, open-label, fixed-sequence study was conducted to evaluate the effects of a single dose of cyclosporine A (400 mg) on the steady-state PK of canagliflozin (300 mg) under a fasted state (N=18).
  • Median time to reach maximum plasma concentration (T_max) was slightly delayed when canagliflozin was coadministered with cyclosporine A compared to canagliflozin alone (median T_max: 4 hours and 2 hours, respectively). Coadministration of a single dose of oral cyclosporine A with multiple doses of canagliflozin resulted in ~23% increase in mean canagliflozin AUC_τ.ss but did not affect mean C_max.ss.

Three independent studies in healthy subjects evaluated the effects of canagliflozin on the PK of OCs, warfarin, and digoxin, investigating potential drug interactions via CYP3A4, CYP2C9, and P-gp inhibition, respectively.³

• LN/EE: An open-label, fixed-sequence study was conducted to evaluate the coadministration of canagliflozin (200 mg) and an OC containing LN (150 mcg) and EE (30 mcg) (N=30).
  • AUCs of EE and LN were similar when administered alone or in combination with canagliflozin. Following coadministration with canagliflozin, C_max of EE and LN increased by 22%, though this value was not clinically significant.
  • Mean steady state canagliflozin plasma concentrations, T_max, C_max, and AUC were comparable when canagliflozin was administered alone and in combination with the OC.
• Warfarin: A single-center, open-label crossover study was conducted to evaluate the effects of canagliflozin (300 mg) on the PK of warfarin (30 mg) (N=14).
  • Following coadministration with canagliflozin compared to warfarin alone, there was a 6% increase in the geometric mean ratio (GMR) of AUC_∞ for (S)-warfarin, but not for (R)-warfarin. Coadministration led to an ~3% increase in the GMR of C_max with (R)-warfarin, whereas no change was observed with (S)-warfarin.
  • Pharmacodynamics (PD) of warfarin were unaffected by coadministration of canagliflozin. The international normalized ratio (INR) T_max was between 36 and 48 hours for both treatments. The GMRs for maximum International Normalized Ratio (INRmax) and INR-time curve from time 0 to the time of last measurable concentration (INR AUClast) were similar between treatments.
• Digoxin: An open-label, crossover study was conducted to evaluate the effects of canagliflozin (300 mg) on the PK of digoxin (0.5 mg on day 1, 0.75 mg on days 2 to 7) (N=18).
  • Coadministration resulted in slightly higher mean steady-state concentrations of digoxin compared to digoxin administration alone. Following coadministration, mean C_max of digoxin increased by ~36% and AUC₂₄ by ~20%. Mean trough digoxin plasma concentrations were ~18% higher when digoxin was coadministered with canagliflozin.

Three independent studies in healthy subjects evaluated the effects of canagliflozin on the PK of glyburide, metformin, and simvastatin, investigating potential drug interactions with CYP2C9, OCT2, and CYP3A4 inhibitors, respectively.⁴

• Glyburide: A single-center, open-label, fixed-sequence study was conducted to evaluate potential drug interactions between steady-state canagliflozin (200 mg) and glyburide (1.25 mg) when administered 10 minutes prior to a standardized breakfast (N=29).
  • Coadministration slightly reduced glyburide peak concentrations, after which mean glyburide concentrations were nearly similar. Plasma concentrations of M1 and M2 glyburide metabolites were also similar when glyburide was administered alone or concomitantly with canagliflozin.
• Metformin: A single-center, open-label, fixed-sequence study was conducted to evaluate potential drug interactions between canagliflozin (300 mg) and metformin immediate release (2000 mg) when administered 10 minutes prior to a standardized breakfast (N=18).
  • Canagliflozin had no effect on the C_max of metformin, but did increase metformin AUC (by ~20%) and terminal half-life (t_1/2). Mean CL/F of metformin decreased by 17% when coadministered with canagliflozin. The percentage of metformin dose excreted unchanged in urine over 72 hours, mean t_1/2, CL_R, and Vd/F were similar when coadministered, compared with administration alone.
  • Coadministration with metformin had no clinically meaningful effect on PK of canagliflozin.
• Simvastatin: A single-center, open-label, fixed-sequence study was conducted to evaluate potential drug interactions between canagliflozin (300 mg) and simvastatin (40 mg) (N=22).
  • C_max and AUC for simvastatin increased slightly following coadministration with canagliflozin. The t_1/2 was comparable between both treatment groups.
  • The active HMG-CoA reductase inhibitory activity was similar across both treatment groups.

A single-center, open-label, fixed-sequence study was conducted in healthy subjects to evaluate the effects of HCTZ (25 mg) on the PK and PD of canagliflozin (300 mg), and to assess the tolerability of canagliflozin coadministered with HCTZ (N=30).⁵

• Coadministration of canagliflozin and HCTZ increased the AUC_τ.ss and C_max.ss of canagliflozin by ~12% and ~15%, respectively.

An open-label, one-way crossover study was conducted in healthy Japanese male subjects to evaluate the effects of the glucagon-like peptide-1 receptor agonist teneligliptin (40 mg) on the PK and PD of canagliflozin (200 mg) (N=44).⁶

• Compared with use of either drug alone, coadministration showed no marked changes in any PK parameters evaluated, including changes in plasma levels and decline from plasma over time following administration.

PK CONSIDERATIONS

• Canagliflozin is extensively bound to proteins in plasma (99%), mainly to albumin. Protein binding is independent of canagliflozin plasma concentrations.¹
• O-glucuronidation is the major metabolic elimination pathway for canagliflozin, which is mainly glucuronidated by UGT1A9 and UGT2B4 to 2 inactive O-glucuronide metabolites. CYP3A4-mediated (oxidative) metabolism of canagliflozin is minimal (~7%) in humans.¹ A minor oxidative metabolite (M9) is formed predominantly by CYP3A4-mediated hydroxylation.¹⁰
• Because canagliflozin undergoes glucuronidation by 2 different UGT enzymes and glucuronidation is a high-capacity/low-affinity system, clinically relevant interactions of other drugs on INVOKANA PK via glucuronidation inhibition are unlikely to occur.¹¹
• Canagliflozin did not induce CYP450 enzyme expression (3A4, 2C9, 2C19, 2B6, and 1A2) in cultured human hepatocytes. Canagliflozin did not inhibit the CYP450 isoenzymes (1A2, 2A6, 2C19, 2D6, or 2E1) and weakly inhibited CYP2B6, CYP2C8, CYP2C9, and CYP3A4 based on in vitro studies with human hepatic microsomes. Canagliflozin is a weak inhibitor of P-gp. Canagliflozin is also a substrate of the drug transporters P-gp and MRP2.¹
• In vitro test systems were used to assess the likelihood of CYP450 inhibition by canagliflozin using basic or physiologically-based pharmacokinetic modeling (PBPK). Inhibition assays showed canagliflozin to be a weak inhibitor of CYP3A4, CYP2C9, CYP2B6, CYP2C8, P-gp, and MRP2. The predictive or PBPK approach found canagliflozin not to affect or be affected by clinically important drug interactions.¹²
• In a study evaluating inhibitory effects of sulfonylureas and nonsteroidal anti-inflammatory drugs (NSAIDs) on in vitro canagliflozin metabolism in human liver microsomes, 3 sulfonylurea derivatives were evaluated as inhibitors (chlorpropamide, glimepiride, and gliclazide) and 2 NSAIDs were used as positive control inhibitors (niflumic acid and diclofenac). Glimepiride had the most potent inhibitory effect against canagliflozin M7 metabolite formation (half maximal inhibitory concentration [IC₅₀] value of 88 μm) compared to chlorpropamide and gliclazide (IC₅₀ values of >500 μm). Diclofenac inhibited canagliflozin M5 metabolite formation more than M7, with IC₅₀ values of 32 μm for M5 and 80 μm for M7. Niflumic acid was not associated with inhibitory activity against M5 formation, but had selective inhibitory potency against M7 formation, which is catalyzed by UGT1A9 (IC₅₀ value of 1.9 μm and inhibition constant value of 0.8 μm).¹³

Retrospective studies

A comprehensive disproportionality analysis was conducted using the FAERS database to evaluate the risk of osteomyelitis associated with SGLT2is, both as monotherapy and in combination with metformin from 2013q2 to 2023q1 (N=2569 reports of SGLT2is and 103 reports of concomitant use of SGLT2is and metformin).⁷

• Canagliflozin was the predominant contributor to osteomyelitis reports among SGLT2is, accounting for 97.66% (n=2509) of cases.
• A drug-drug interaction analysis between SGLT2is and metformin revealed a negative signal (Ω shrinkage method: Ω=-1.067, Ω025=-1.370, χ=-7.120), suggesting a potential protective role. Concurrent use of canagliflozin with metformin was associated with a reduced risk of osteomyelitis.
• The correlation between SGLT2is-associated osteomyelitis and amputation was reported in 2150 cases, accounting for 80% of osteomyelitis-related adverse events.

Another study analyzed FAERS data (2013-2023) to evaluate ARF risk associated with SGLT2is dapagliflozin, empagliflozin, and canagliflozin coadministered with potentially nephrotoxic drugs (N=1636).⁸

• The investigation focused on identifying drug-drug interactions that elevated renal risk, aiming to provide evidence-based insights into combination therapies affecting patient safety.
• A comparative analysis showed that 23 drug combinations had higher risk of ARF compared with single-drug therapy. The combinations of dapagliflozin with sulfamethoxazole (adjusted ROR, 6.36; 95% CI, 3.27-23.33), empagliflozin with spironolactone (adjusted ROR, 4.32; 95% CI, 3.14-24.53), canagliflozin with sulfamethoxazole (adjusted ROR, 7.50; 95% CI, 3.24-24.67), and canagliflozin with pregabalin (adjusted ROR, 6.13; 95% CI, 3.66-24.83) showed a statistically significant association with ARF.

The FAERS database (the largest global database for real-world adverse event reports) was used to evaluate the impact of combined use of SGLT2is and ARNI influences the risk of AKI (N=413).⁹

• Overall, 413 reports pertaining to cotherapy with SGLT2is and ARNI were identified; of these, 99 (24.0%) were reports of documented cases of AKI.
• A disproportionality analysis demonstrated significantly higher AKI reporting rate with cotherapy of SGLT2is and ARNI (adjusted ROR, 8.04; 95% CI, 6.20-10.42; P<0.001. Additional AKI signals were detected with combination therapy for individual agents, including empagliflozin (adjusted ROR, 11.82; 95% CI, 7.54-18.55), dapagliflozin (adjusted ROR, 4.82; 95% CI, 3.34-6.95), and canagliflozin (adjusted ROR, 16.82; 95% CI, 3.75-75.57) (all P<0.001), with the strongest signal observed for canagliflozin plus ARNI.
• A subgroup analysis demonstrated that cotherapy with SGLT2is and ARNI significantly increased the AKI reporting rate in patients with diabetes (adjusted ROR, 20.57; 95% CI, 11.93-35.46; P<0.001), those with hypertension (adjusted ROR, 4.30; 95% CI, 1.98-9.37; P<0.001), and those with diabetes plus hypertension (adjusted ROR, 5.44; 95% CI, 1.92-15.43; P=0.001).
• Combination therapy did not show an increase in AKI risk among patients with heart failure.
• AKI vs non-AKI cases exhibited a higher mean age (70.3 ± 10.6 vs 65.7 ± 12.9 years; 60–80 years: 64.7% vs 29.3%), a greater proportion of males (75.8% vs 62.7%), and a markedly higher hospitalization rate (71.7% vs 33.4%) followed by life-threatening events (15.2%) and death (8.1%).

LITERATURE SEARCH

A literature search of MEDLINE^®, EMBASE^®, BIOSIS Previews^®, Derwent Drug File (and/or other resources, including internal/external databases) pertaining to this topic was conducted on 06 May 2026.