Abstract
Context
Metabolic dysfunction–associated fatty liver disease (MAFLD) is highly prevalent among patients with type 2 diabetes mellitus (T2DM); however, there is still no approved pharmacological treatment. Sodium-glucose cotransporter 2 (SGLT-2) inhibitors have been suggested to beneficially modify liver-related outcomes in patients with diabetes.
Objective
We aimed to investigate the effects of the SGLT-2 inhibitor canagliflozin on liver-related outcomes in patients with advanced T2DM and high cardiovascular risk.
Methods
We performed a secondary post hoc analysis of 2 large double-blind randomized controlled trials, CANVAS (NCT01032629) and CANVAS-R (NCT01989754), which included patients with T2DM and high cardiovascular risk who were randomized to receive either canagliflozin or placebo once daily. The primary endpoint was a composite of improvement of alanine aminotransferase (ALT) levels >30% or normalization of ALT levels. Secondary endpoints included change in noninvasive tests of fibrosis and weight reduction of >10%.
Results
In total, 10 131 patients were included, with a median follow-up of 2.4 years (mean age 62 years; mean duration of diabetes 13.5 years; 64.2% male). Of those patients, 8967 (88.5%) had MAFLD according to hepatic steatosis index and 2599 (25.7%) exhibited elevated liver biochemistry at baseline. The primary composite endpoint occurred in 35.2% of patients receiving canagliflozin and in 26.4% with placebo (adjusted odds ratio [aOR] 1.51; 95% CI, 1.38-1.64; P < .001). Canagliflozin led to improvements in some noninvasive tests of fibrosis (NFS, APRI, FNI). Significant weight reduction of >10% (within 6 years) was achieved in 12.7% with canagliflozin compared to 4.1% with placebo (aOR 3.45; 95% CI, 2.91-4.10; P < .001).
Conclusion
In patients with T2DM, treatment with canagliflozin vs placebo resulted in improvements in liver biochemistry and metabolism and might beneficially affect liver fibrosis.
Keywords: SGLT-2 inhibitor, nonalcoholic fatty liver disease, metabolic associated fatty liver disease, diabetes, alanine aminotransferase, noninvasive tests of fibrosis
Metabolic associated fatty liver disease (MAFLD) is highly prevalent among patients with type 2 diabetes mellitus (T2DM) (1) and it has been shown that patients with T2DM are at high risk of progression to steatohepatitis (NASH), higher stages of fibrosis, and eventually cirrhosis (2). With the global obesity pandemic, MAFLD has evolved to the leading cause of chronic liver disease at a global scale (3) and is rapidly becoming a leading cause for end-stage liver disease and consequently for liver transplantation (4). However, evidence suggests that MAFLD not only increases the risk for liver-related morbidity and mortality (5), but also for cardiovascular disease (6) and chronic kidney disease (7). Despite an increasing burden of disease, no pharmacotherapy has yet been approved (8). The current cornerstones of therapy focus on lifestyle interventions to encourage weight loss and treatment of coexisting metabolic conditions (9, 10).
Sodium-glucose cotransporter 2 (SGLT-2) inhibitors are modern antidiabetic drugs that increase urinary glucose excretion and thereby lead to improved glycemia, reduced blood pressure and body weight (11, 12) as well as improved cardiovascular and renal outcomes (13). In addition, in early preliminary experimental and clinical studies, they have been shown to reduce oxidative stress, systemic and tissue low-grade inflammation (14), and to reduce fatty liver content (15, 16). Due to these multifaceted effects, SGLT-2 inhibitors have been proposed as promising novel treatment for MAFLD; however, individual high-quality data from large randomized controlled trials are still scarce.
In this secondary analysis of 2 large randomized controlled outcome trials, we aimed at investigating the effects of the SGLT-2 inhibitor canagliflozin on liver-related outcomes in patients with advanced T2DM and high cardiovascular risk.
Methods
Study Design and Participants
This is a secondary analysis of 2 randomized, placebo-controlled, double-blind, multicenter trials: CANVAS (NCT01032629) and CANVAS-R (NCT01989754) involving patients with T2DM and high cardiovascular risk. Full details of the trial designs and results have been published previously (11). In brief, adult patients from 667 centers in 30 countries worldwide with T2DM (glycated hemoglobin level, ≥ 7.0% and ≤10.5%) and age ≥30 years with a history of symptomatic atherosclerotic cardiovascular disease, or age ≥50 years with at least 2 risk factors for cardiovascular disease, were randomized to either receive canagliflozin or placebo once daily in addition to their established antidiabetic treatment, respectively. In the CANVAS trial programs, alcohol consumption was not defined as exclusion criterion; therefore, we used the new nomenclature “metabolic dysfunction–associated fatty liver disease” (MAFLD) which includes all patients with evidence of fatty liver disease in addition to one of the following 3 features: overweight/obesity, T2DM, or metabolic dysregulation irrespective of other underlying chronic liver diseases.
All participants provided written informed consent. The conduct of the trials adhered to the Declaration of Helsinki and Good Clinical Practice Guidelines (17), and ethical committees of all participating hospitals approved the studies before patient recruitment. The studies were designed and conducted by the sponsor in collaboration with the principal investigators. The sponsor collected the data and monitored study conduct.
Procedures
Eligible patients underwent a 2-week, open-label, placebo run-in period in which background glucose-lowering therapy was unchanged. Participants in the CANVAS trial were randomly assigned to receive canagliflozin at a dose of 100 mg or 300 mg, or to receive matching placebo (1:1:1 ratio). Participants in the CANVAS-R trial were randomly assigned to receive canagliflozin at a dose of 100 mg, with an optional increase to 300 mg from week 13, or to receive matching placebo (1:1 ratio). Follow-up visits, including clinical examination and blood tests, were performed in 3 visits during the first year (week 13, 26, and 52) and at 6-month intervals thereafter.
Outcomes
The primary objectives of CANVAS and CANVAS-R were to evaluate the effects of canagliflozin treatment on cardiovascular and renal endpoints. In this secondary post hoc analysis, the primary aim was to evaluate liver-related outcomes. The primary endpoint was a composite of the proportion of patients achieving a clinically significant reduction in alanine aminotransferase (ALT) levels or achieving ALT normalization (ALT ≤30 IU/L). A reduction in ALT levels of more than 30% was deemed clinically significant (18). Secondary endpoints included clinically significant reduction of ALT levels, ALT normalization, and proportion of patients with progression to severe fibrosis or cirrhosis (F3 or F4) according to validated noninvasive fibrosis scoring systems (19): NAFLD Fibrosis Score (NFS) (20), Fibrosis-4 (Fib-4) (21), AST to Platelet Ratio Index (APRI) (22), and Fibrotic NASH Index (FNI) (23). In addition, we assessed the proportion of patients with reduction in body weight of more than 5% or 10%, respectively; as well as the proportion of patients achieving optimal glycemic control (glycated hemoglobin [HbA1c] < 7.0%), and good blood pressure control (systolic blood pressure <140 mmHg, diastolic blood pressure <90 mmHg).
Statistical Analysis
Descriptive statistics were used to summarize differences in demographic and baseline characteristics among study groups. Baseline demographics were stratified according to presence of MAFLD as assessed by the hepatic steatosis index (24, 25). The used cutoff (hepatic steatosis index >36) has previously been validated in patients with T2DM as compared with steatosis detection via ultrasound (26). Continuous variables were given as mean ± SD, and categorical variables as numbers (percentage). For comparative analyses, data from both canagliflozin dose groups (100 and 300 mg) were pooled. For the main outcome measures, crude and multivariable adjusted estimates of the effect size and corresponding 95% CI were determined using linear or logistic regression, as appropriate. All multivariable models were adjusted for the same variables: randomized treatment, age, gender, race, body mass index (BMI), smoking status, hypertension, hyperlipidemia, heart failure, diabetic nephropathy, diabetic neuropathy, diabetic retinopathy, coronary heart disease, peripheral artery disease, cerebrovascular disease, chronic kidney disease, hepatopathy, and liver steatosis. For analyses on the secondary endpoint ALT normalization, patients with ALT ≤30 IU/L were excluded.
All tests were two-sided, P < .05 was considered significant with two-sided 95% CI. All statistical analyses were performed using STATA version 15.1 (StataCorp, College Station, TX, USA).
Results
Patient Characteristics
This study integrates data from 2 trials (CANVAS and CANVAS-R) involving a total of 10 142 participants with T2DM and high cardiovascular risk. Liver-related parameters were available for 10 131 patients (99.9% of all included patients; canagliflozin, n = 5787; placebo, n = 4344). Baseline characteristics stratified by the existence of MAFLD are shown in Table 1 and by treatment in Supplementary Table S1 (27). Participants had a mean age of 62 years, 35.8% were female, and the mean duration of diabetes was 13.5 years. The median follow-up was 2.4 years (126.1 weeks). Most patients were obese (6048; 59.7%) with a mean BMI among participants of 32 kg/m2 and a mean HbA1c of 8.2%.
Table 1.
Baseline characteristics of included patients
| Total | MAFLD unlikely | MAFLD | P value | |
|---|---|---|---|---|
| N = 10 131 | N = 1164 | N = 8967 | ||
| General characteristics | ||||
| Age, years | 62 (8.0) | 64 (9.0) | 61 (8.0) | <.001 |
| Sex | ||||
| Female | 3631 (35.8%) | 270 (23.2%) | 3361 (37.5%) | <.001 |
| Male | 6500 (64.2%) | 894 (76.8%) | 5606 (62.5%) | |
| Race/ethnicity | ||||
| White | 7933 (78.3%) | 618 (53.1%) | 7315 (81.6%) | <.001 |
| Asian | 1284 (12.7%) | 405 (34.8%) | 879 (9.8%) | |
| Black | 336 (3.3%) | 36 (3.1%) | 300 (3.3%) | |
| Other | 578 (5.7%) | 105 (9.0%) | 473 (5.3%) | |
| Body weight kg | 90.2 (20.2) | 67.5 (10.5) | 93.1 (19.3) | <.001 |
| Body mass index | ||||
| Underweight (BMI 15-19.9) | 39 (0.4%) | 39 (3.4%) | 0 (0.0%) | <.001 |
| Normal weight ( BMI 20-24.9) | 901 (8.9%) | 681 (58.5%) | 220 (2.5%) | |
| Overweight (BMI 25-29.9) | 3143 (31.0%) | 442 (38.0%) | 2701 (30.1%) | |
| Obesity class I (BMI 30-34.9) | 3332 (32.9%) | 2 (0.2%) | 3330 (37.1%) | |
| Obesity class II (BMI 35-39.9) | 1773 (17.5%) | 0 (0.0%) | 1773 (19.8%) | |
| Obesity class III (BMI ≥40) | 943 (9.3%) | 0 (0.0%) | 943 (10.5%) | |
| Systolic blood pressure, mmHg | 137 (15.8) | 134 (16.5) | 137 (15.6) | <.001 |
| Diastolic blood pressure, mmHg | 78 (9.7) | 76 (9.4) | 78 (9.7) | <.001 |
| Smoker | 1804 (17.8%) | 248 (21.3%) | 1556 (17.4%) | <.001 |
| Comorbidities (%) | ||||
| Hypertension | 9105 (89.9%) | 967 (83.1%) | 8138 (90.8%) | <.001 |
| Hyperlipidemia | 7046 (69.5%) | 712 (61.2%) | 6334 (70.6%) | <.001 |
| Heart failure | 1430 (14.1%) | 112 (9.6%) | 1318 (14.7%) | <.001 |
| Diabetic nephropathy | 1958 (19.3%) | 224 (19.2%) | 1734 (19.3%) | .94 |
| Diabetic neuropathy | 2802 (27.7%) | 279 (24.0%) | 2523 (28.1%) | .003 |
| Diabetic retinopathy | 91 (0.9%) | 5 (0.4%) | 86 (1.0%) | .07 |
| Coronary heart disease | 1221 (12.1%) | 122 (10.5%) | 1099 (12.3%) | .08 |
| Peripheral artery disease | 1457 (14.4%) | 166 (14.3%) | 1291 (14.4%) | .90 |
| Cerebrovascular disease | 1444 (14.3%) | 162 (13.9%) | 1282 (14.3%) | .73 |
| Chronic kidney disease | 288 (2.8%) | 36 (3.1%) | 252 (2.8%) | .59 |
| Hepatopathy | 28 (0.3%) | 6 (0.5%) | 22 (0.2%) | .099 |
| Steatosis | 731 (7.2%) | 45 (3.9%) | 686 (7.7%) | <.001 |
| Laboratory values | ||||
| HbA1c, % | 8.2 (0.9) | 8.2 (0.9) | 8.3 (0.9) | <.001 |
| Triglycerides, mmol/L | 2.0 (1.4) | 1.6 (1.3) | 2.1 (1.4) | <.001 |
| LDL cholesterol, mmol/L | 2.3 (0.9) | 2.3 (0.9) | 2.3 (0.9) | .74 |
| HDL cholesterol, mmol/L | 1.2 (0.3) | 1.2 (0.4) | 1.2 (0.3) | <.001 |
| eGFR, mL/min/1.73 m2 | 76 (20.0) | 76 (22.0) | 77 (20.0) | .13 |
| ALT, IU/L | 26 (14.0) | 19 (13.0) | 27 (14.0) | <.001 |
| AST, IU/L | 23 (11.0) | 22 (18.0) | 23 (10.0) | .03 |
| GGT, IU/L | 38 (44.0) | 30 (48.0) | 39 (42.9) | <.001 |
| ALP, IU/L | 77 (25.0) | 77 (26.0) | 77 (25.0) | .99 |
| Bilirubin, μmol/L | 9 (4.0) | 9 (4.0) | 9 (4.0) | .002 |
| Albumin, g/L | 41.4 (3.1) | 41.6 (3.5) | 41.3 (3.1) | .02 |
| Noninvasive scores (%) | ||||
| NFS | ||||
| Advanced fibrosis (≥F3) | 2265 (22.4%) | 143 (12.3%) | 2122 (23.7%) | <.001 |
| Indeterminate | 6436 (63.5%) | 739 (63.5%) | 5697 (63.5%) | .98 |
| No advanced fibrosis (<F3) | 1430 (14.1%) | 282 (24.2%) | 1148 (12.8%) | <.001 |
| FIB-4 | ||||
| Advanced fibrosis (≥F3) | 273 (2.7%) | 58 (5.0%) | 215 (2.4%) | <.001 |
| Indeterminate | 3362 (33.2%) | 509 (43.7%) | 2853 (31.8%) | <.001 |
| No advanced fibrosis (<F3) | 6496 (64.1%) | 597 (51.3%) | 5899 (65.8%) | <.001 |
| APRI | ||||
| Cirrhosis (F4) | 47 (0.5%) | 9 (0.8%) | 38 (0.4%) | .099 |
| Indeterminate | 1273 (12.6%) | 109 (9.4%) | 1164 (13.0%) | <.001 |
| No cirrhosis (<F4) | 8811 (87.0%) | 1046 (89.9%) | 7765 (86.6%) | .002 |
| FNI | ||||
| Fibrosis | 4470 (44.1%) | 402 (34.5%) | 4068 (45.4%) | <.001 |
| Indeterminate | 4868 (48.1%) | 624 (53.6%) | 4244 (47.3%) | <.001 |
| No fibrosis | 793 (7.8%) | 138 (11.9%) | 655 (7.3%) | <.001 |
Data are presented as n (%) or mean (SD). MAFLD was diagnosed using hepatic steatosis index >36. Rule-in cutoffs for advanced fibrosis: NFS >0.675, FIB-4 > 2.67, FNI >0.33, and for cirrhosis APRI >1.5. Rule-out cutoffs for advanced fibrosis: NFS < −1.455, FIB-4 < 1.3, FNI <0.1, and for cirrhosis APRI <0.5.
Abbreviations: ALP, alkaline phosphatase; ALT, alanine aminotransferase; APRI, aspartate aminotransferase to platelet ratio; AST, aspartate aminotransferase; FIB-4, Fibrosis-4; FNI, Fibrotic NASH Index; GGT, gamma-glutamyl transferase; HDL, high-density lipoprotein; LDL, low-density lipoprotein; MAFLD, metabolic associated fatty liver disease; NFS, NAFLD Fibrosis Score.
The majority of patients (n = 8967; 88.5%) had MAFLD according to the hepatic steatosis index; however, at baseline merely 7.2% of patients were known to have liver steatosis and 0.3% had a previous diagnosis of unspecified hepatopathy. Among those with MAFLD, the proportions of patients from White ethnic background was significantly higher (81.6% vs 53.1%, P < .001), while patients with Asian ethnicity were significantly less represented (9.8% vs 34.8%, P < .001). While HbA1c values were almost identical (8.3% vs 8.2%), patients with MAFLD had higher blood pressure values and were more frequently diagnosed with hypertension, hyperlipidemia, and heart failure (Table 1).
Association of Baseline Characteristics With Impaired ALT Levels
In total, 2599 patients (25.7%) presented with ALT levels >30 IU/L. ALT levels were negatively correlated with age, with highest values in patients aged <50 years (29.6 IU/L, SD 15.4) and lowest among patients aged >70 years (21.2 IU/L, SD 10.0) (Supplementary Table S2 (27)). When compared with female patients, male sex was associated with higher ALT levels at baseline (between-group difference 3.5 IU/L; 95% CI, 2.95 to 4.07; P < .001). Among ethnicities, Black patients had the lowest ALT levels (21.3 IU/L, SD 11.8), while highest levels were seen in White patients (26.0 IU/L, SD 14.1) (between-group difference −4.4 IU/L; 95% CI, −5.84 to −2.87; P < .001). There was a correlation of ALT concentration and BMI, with the highest values among patients with BMI >35 kg/m2. Patients with known liver steatosis had higher ALT levels when compared to those without steatosis (between-group difference 3.9 IU/L; 95% CI, 2.90 to 4.96; P < .001). In the small fraction of patients with prior diagnosis of hepatopathy, ALT levels were at mean 12.2 IU/L higher when compared to patients without the diagnosis (95% CI, 7.27 to 17.18; P < .001).
Primary and Secondary Endpoints
In patients receiving canagliflozin, a rapid and sustained reduction in ALT and AST levels was observed, while in patients receiving placebo liver biochemistry remained unchanged (Fig. 1). The primary composite endpoint of clinically significant improvement or normalization in ALT was reached by a higher proportion of patients receiving canagliflozin (2037/5787; 35.2%) compared with placebo (1146/4344; 26.4%), resulting in an adjusted odds ratio (aOR) of 1.51 (95% CI, 1.38 to 1.64; P < .001) (Fig. 2A and Table 2). The secondary endpoint of improvement in ALT levels by more than 30% was reached by 31.7% of patients receiving canagliflozin, compared to 22.3% with placebo (aOR 1.61; 95% CI, 1.47 to 1.76; P < .001; Fig. 2B). Normalization of initially elevated ALT levels was achieved in 69.5% of patients with canagliflozin vs 57.5% of patients randomized to placebo (aOR 1.71; 95% CI, 1.44 to 2.03; P < .001; Fig. 2C). Within 6 years after randomization, a reduction in body weight by more than 5% but less than 10% was achieved by 38.8% in the canagliflozin group vs 16.3% in the placebo group (aOR 3.24; 95% CI, 2.94 to 3.58; P < .001). Likewise, a significantly larger proportion of patients receiving canagliflozin (12.7%) achieved weight loss of more than 10%, compared with 4.1% of patients receiving placebo (aOR 3.45; 95% CI, 2.91 to 4.10; P < .001) (Fig. 2D). Higher proportions of patients receiving canagliflozin vs placebo (34.0% vs 17.7%) achieved optimal glycemic control of HbA1c < 7.0% (aOR 2.51; 95% CI, 2.27 to 2.78; P < .001) as well as good blood pressure control (34.2% vs 24.6%) (aOR 1.64; 95% CI, 1.49 to 1.81; P < .001) (Table 2).
Figure 1.
Effects of canagliflozin on liver function tests over time. Temporal dynamics of liver biochemistry: alanine aminotransferase, ALT (A) and aspartate aminotransferase, AST (B) in patients randomized to canagliflozin or placebo.
Figure 2.
Primary and secondary outcomes. Endpoints up to 312 weeks after randomization are shown. The results of the combined primary endpoint show a significant superiority of canagliflozin compared with placebo (Panel A). Results on secondary endpoints show superiority of canagliflozin on improvement of ALT >30% (Panel B) and normalization of ALT (Panel C). Panel D shows weight loss achieved with canagliflozin compared with placebo at 1 year after randomization.
Table 2.
Effects of canagliflozin on primary and secondary outcomes
| Placebo (n = 4344) | Canagliflozin (n = 5787) | Crude odds ratio (95% CI) | P value | Multivariable adjusted odds ratio (95% CI) | P value | |
|---|---|---|---|---|---|---|
| Primary endpoint | ||||||
| Clinically significant improvement of ALT or ALT normalization | 1146 (26.4) | 2037 (35.2) | 1.52 (1.40−1.65) | <.001 | 1.51 (1.38−1.64) | <.001 |
| Secondary endpoints | ||||||
| Clinically significant improvement of ALT | 968 (22.3) | 1834 (31.7) | 1.62 (1.48−1.77) | <.001 | 1.61 (1.47−1.76) | <.001 |
| ALT normalization | 584 (57.5) | 1002 (69.5) | 1.69 (1.43−2.00) | <.001 | 1.71 (1.44−2.03) | <.001 |
| Weight reduction >5% | 710 (16.3) | 2243 (38.8) | 3.24 (2.94−3.57) | <.001 | 3.24 (2.94−3.58) | <.001 |
| Weight reduction >10% | 176 (4.1) | 735 (12.7) | 3.45 (2.91−4.08) | <.001 | 3.45 (2.91−4.10) | <.001 |
| Good blood pressure control | 1068 (24.6) | 1980 (34.2) | 1.60 (1.46−1.74) | <.001 | 1.64 (1.49−1.81) | <.001 |
| Good glycemic control | 732 (17.7) | 1869 (34.0) | 2.40 (2.18−2.64) | <.001 | 2.51 (2.27−2.78) | <.001 |
Clinically significant improvement of ALT was defined as a reduction from baseline by more than 30%. Normalization of ALT was defined as a reduction from baseline to under 30 IU/L (patients with ALT ≤ 30 IU/L at baseline were excluded). Good blood pressure control was defined as blood pressure control below 140/90 mmHg. Good glycemic control was defined as reduction of HbA1c below 7.0% target. Abbreviation: ALT, alanine aminotransferase.
Noninvasive Tests of Fibrosis
At baseline, advanced fibrosis (stage ≥F3) could be ruled out in 14.1% of patients according to NFS. In contrast, according to FIB-4, advanced fibrosis could be ruled out in the majority of patients (64.1%) (Table 1). Treatment with canagliflozin vs placebo led to significant reductions in absolute NFS values both at 1 year (between-group difference −0.049; 95% CI, −.077 to −.022; P < .001), and at 3 years after randomization (between-group difference −0.062; 95% CI, −.116 to −.008; P = .002) (Table 3). However, this did not result in a reduction of the proportion of patients categorized as having advanced fibrosis (Supplementary Fig. S1 (27)). Absolute values of FIB-4 scores were mostly unchanged over time in both groups without differences between the treatment groups (Table 3 and Supplementary Fig. S1 (27)). According to baseline APRI scores, 0.5% of patients were categorized with suspected cirrhosis. Treatment with canagliflozin vs placebo led to a significant reduction in absolute APRI values at 1 year (between-group difference −0.016; 95% CI, −.025 to −.007; P = .001), but not at 3 years after randomization (Table 3). According to the novel noninvasive score FNI, 44.1% of patients were ruled in to have fibrosis and 7.8% were ruled out (Table 1). Canagliflozin led to a significant reduction in absolute FNI scores at 1 year (between-group difference −0.082; 95% CI, −.089 to −.075; P < .001) and 3 years (between-group difference −0.054; 95% CI, −.067 to −.041; P < .001) (Table 3).
Table 3.
Effects of canagliflozin on noninvasive tests of fibrosis
| Baseline | 1 year | Change from baseline to 1 year (95% CI) | P value | 3 years | Change from baseline to 3 years (95% CI) | P value | |
|---|---|---|---|---|---|---|---|
| NFS | |||||||
| Placebo | −0.187 (1.17) | −0.151 (1.18) | .052 (−.967, 1.109) | −0.267 (1.21) | .097 (−1.074, 1.141) | ||
| Canagliflozin | −0.262 (1.18) | −0.257 (1.18) | .012 (−1.034, 1.025) | −0.371 (1.17) | .047 (−1.064, 1.192) | ||
| Between-group difference | − | − | −.049 (−.077, −.022) | <.001 | − | −.062 (−.116, −.008) | .002 |
| FIB-4 | |||||||
| Placebo | 1.262 (0.63) | 1.262 (0.70) | .008 (−.536, .569) | 1.249 (0.61) | .040 (−.562, .559) | ||
| Canagliflozin | 1.243 (0.78) | 1.247 (0.59) | .009 (−.537, .551) | 1.231 (0.62) | .060 (−.500, .599) | ||
| Between-group difference | − | − | −.004 (−.024, 0.017) | .74 | − | .009 (−.025, .044) | .60 |
| APRI | |||||||
| Placebo | 0.339 (0.22) | 0.336 (0.27) | −.002 (−.202, 0.202) | 0.352 (0.28) | .012 (−.206, .250) | ||
| Canagliflozin | 0.343 (0.49) | 0.322 (0.19) | −.022 (−.240, .167) | 0.333 (0.22) | .004 (−.207, .207) | ||
| Between-group difference | − | − | −.016 (−.025, −.007) | .001 | − | −.011 (−.027, .005) | .17 |
| FNI | |||||||
| Placebo | 0.346 (0.21) | 0.325 (0.22) | −.019 (−.319, .291) | 0.325 (0.22) | −.025 (−.348, .334) | ||
| Canagliflozin | 0.351 (0.21) | 0.248 (0.18) | −.103 (−.421, .167) | 0.272 (0.19) | −.080 (−.404, .211) | ||
| Between-group difference | − | − | −.082 (−.089, −.075) | <.001 | − | −.054 (−.067, −.041) | <.001 |
Abbreviations: APRI, AST to Platelet Ratio Index; FIB-4, Fibrosis-4; FNI, Fibrotic NASH Index; NFS, NAFLD Fibrosis Score.
Differences in Effects Among Subgroups
All observed effects were broadly consistent across a wide range of prespecified subgroups (Supplementary Fig. S2 (27)). There was a significant difference in efficacy on the primary composite endpoint according to race, with less pronounced effects among patients from White ethnic background (Pfor heterogeneity = .001). Furthermore, treatment effects with canagliflozin on the primary composite endpoint were more pronounced among patients without peripheral artery disease when compared to patients with that diagnosis. Secondary analyses according to improvement in HbA1c of 0.5% or more vs less than 0.5% at 1 year using an interaction term did not reveal any evidence of significant between-group difference (Pfor interaction > .05). Likewise, subgroup analyses on body weight change (more than 5% vs less than 5% weight change at 1 year) did not reveal a subgroup difference (Pfor interaction > .05).
Discussion
This is a secondary analysis of 2 large randomized, double-blind, placebo-controlled trials including more than ten thousand patients with long-standing diabetes and high cardiovascular risk. In this metabolically ill population, most patients had concomitant MAFLD and up to one-fifth were suspected to have advanced fibrosis. Treatment with the SGLT-2 inhibitor canagliflozin vs placebo in addition to the established antidiabetic regimen led to (i) significant weight reduction with improved blood pressure and glycemic control; (ii) a rapid and sustained improvement of liver biochemistry, which seemed to be independent from improvements in glycemic control or body weight; and (iii) reduced absolute scores in noninvasive tests of liver fibrosis.
This is the largest clinical trial, so far, investigating the effects of an SGLT-2 inhibitor on MAFLD in individuals with T2DM. Patients with long-standing T2DM and multiple cardiovascular risk factors are known to be at highest risk for the development of NASH (28, 29), of advanced stages of fibrosis (30), and to encounter adverse liver-related outcomes (31). These patients are characterized by high insulin resistance and chronic low-grade systemic inflammation, both of which are central pathophysiological mechanisms in the development of MAFLD-related adverse outcomes (32). Observational data suggest that, in patients with T2DM, the prevalence of MAFLD is approximately 67% (33). In our study population with a mean duration of diabetes of more than 10 years, the prevalence of MAFLD—estimated by hepatic steatosis index—was even higher at almost 90%, comparable with high prevalence previously described in morbidly obese patients (34). In contrast, elevated liver transaminases were found in around one-quarter of patients, underscoring the notion that altered liver biochemistry tests alone do not qualify as diagnostic screening test for MAFLD. Of note, in CANVAS and CANVAS-R < 1% of patients had a prior diagnosis of liver disease at baseline. Hence, our data bring to light the huge diagnostic gap in patients with diabetes, who still do not routinely undergo screening for hepatic steatosis or stiffness in most health care systems (35), often resulting in late diagnosis of MAFLD at higher fibrosis stages or eventually at the stage of cirrhosis, when treatment strategies become more complex and longer lasting (36).
There is increasing evidence in favor of therapeutic value of SGLT-2 inhibitors as well as glucagon-like peptide 1 (GLP-1) agonists in MAFLD (37); however, most studies have the limitation of rather small sample sizes and/or short study durations (38). In line with previous small observational studies (39), we observed that treatment with canagliflozin vs placebo led to a rapid and sustained reduction of liver transaminases—even in patients with “normal” ALT levels at baseline. Serum transaminases have been shown to correlate with the risk of fibrosis progression, therefore sustained reduction or normalization can be considered a clinically meaningful endpoint (40). Indeed, previous studies with paired biopsies observed that, besides improvements in HbA1c, ALT normalization was one of the strongest predictors of fibrosis improvement compared with other biomarkers (41).
In our study, treatment with canagliflozin furthermore was associated with significantly improved body weight and glycemic and blood pressure control. In total, 23.5% of patients achieved a significant weight reduction of more than 5% total body weight and 4.3% even achieved more than 10% body weight reduction within 1 year of treatment.
In our study, the diagnostic performance of established noninvasive tests of fibrosis was highly heterogeneous. In fact, according to NFS and FNI, the majority of patients would have to undergo further assessment of MAFLD severity, since these tests ruled out fibrosis only in a small proportion of patients. However, this was not the case with the FIB-4 score. This heterogeneity among noninvasive tests has been shown in previous studies involving patients with T2DM, in whom such tests seem to have poor diagnostic discrimination and often lead to unspecific and heterogenous results even with age-adjusted cutoffs (42). Recently, the use of sequential assessments using noninvasive tests has gained clinical relevance, since the large REGENERATE study has demonstrated that changes in noninvasive tests over time strongly correlated with histological changes of liver fibrosis (18). We observed statistically significant but rather small changes in NFS, APRI, and FNI scores at 1 and 3 years of treatment with canagliflozin when compared with placebo. In contrast to data also indicating an effect on FIB-4 scores (37), in our study FIB-4 remained largely stable.
These observations in noninvasive scores are in line with recent studies examining effects of SGLT-2 inhibitors on liver histology. In a small observational study investigating the effects of a 1-year treatment with canagliflozin in patients with T2DM, some histopathological improvements in markers of inflammation and fibrosis were observed in 6 out of 7 patients (43). A larger, 48-week, randomized, open-label, parallel-group trial involving patients with T2DM and biopsy-proven NAFLD (n = 40) confirmed significant improvements in both inflammation and fibrosis upon treatment with tofogliflozin. Most interestingly, gene expression profiling of liver tissue was performed, revealing that tofogliflozin effectively altered hepatic expression of genes involved in energy metabolism, inflammation, and fibrosis. In this study tofogliflozin was compared to glimepiride, which led to similar improvements of blood sugar control but without most of the effects on liver histology or gene expression, indicating that there may be direct effects on hepatocytes that go beyond glycemic control (44). Indeed, our results suggest that improvements in liver biochemistry were independent from glycemic control or body weight reduction, strengthening the hypothesis of possible direct effects of SGLT-2 inhibitors on the liver that need to be further explored in mechanistic studies.
SGLT-2 inhibitors have been shown to promote fasting-like metabolic changes, increasing fatty acid oxidation and ketone body formation via FGF21-dependent and -independent mechanisms (45). Multisystem effects of SGLT-2 inhibitors with their beneficial effects in heart failure and chronic kidney disease make them highly interesting since MAFLD is more and more seen as a multisystem disorder. NASH is increasingly considered an independent cardiovascular risk factor (46) and cardiovascular disease remains the most common cause of death in these patients (47). Furthermore, higher rates of chronic kidney disease are increasingly diagnosed in patients with MAFLD (7, 48). Given the paucity of large intervention studies assessing the effects of these novel glucose-lowering drugs on liver-related outcomes, our results should pave the ground for more studies in patients with NASH and/or MAFLD. This is especially relevant since it has been shown that the type of antidiabetic treatment may affect MAFLD-associated adverse events, such as the risk for the development of hepatocellular cancer (49).
Our findings should be interpreted in light of some limitations. The CANVAS and CANVAS-R trials were originally designed to investigate cardiovascular and renal outcomes in patients with T2DM and high cardiovascular risk (11). Therefore, the results should be seen as exploratory efficacy analyses. Second, although in previous studies improvements in noninvasive parameters have been associated with histologic response (18), future studies with paired liver biopsy are warranted to confirm and extend our results. Third, since only a small fraction of patients mirrored advanced fibrosis, possible antifibrotic effects were strongly underpowered and may be underestimated. Fourth, the CANVAS trials included diabetics only, therefore it remains unclear how MAFLD patients without diabetes would respond to canagliflozin. Finally, differences in the prevalence of MAFLD may be attributed to varying diagnostic accuracy of hepatic steatosis index across ethnicities (50). Strengths of the study include the randomized, placebo-controlled design, the well-characterized, large patient population with patients at high risk for MAFLD. Moreover, the median follow-up of more than 2 years is likely sufficient to capture meaningful changes in metabolism and fibrosis.
Conclusion
Our data demonstrate that SGLT-2 inhibitors are not only highly beneficial in the treatment of diabetes, heart failure, and chronic kidney disease, but also highly promising in the treatment of MAFLD. In addition to the known metabolic effects on glycemia and body weight, treatment with canagliflozin vs placebo significantly improved liver biochemistry and reduced scores of noninvasive tests of fibrosis independent from improvements in glycemic control or body weight. Given the established cardiovascular and renal benefits in patients with T2DM, SGLT-2 inhibitors could prove to be suitable and cost-effective pharmaceutical agents in the treatment of patients with MAFLD. Future studies are needed to investigate the evolution of liver histology upon SGLT-2 inhibitor treatment in both diabetic and nondiabetic patients.
Acknowledgments
This study, carried out under YODA Project # 2020-4409, used data obtained from the Yale University Open Data Access Project, which has an agreement with JANSSEN RESEARCH & DEVELOPMENT, L.L.C. The interpretation and reporting of research using this data are solely the responsibility of the authors and does not necessarily represent the official views of the Yale University Open Data Access Project or JANSSEN RESEARCH & DEVELOPMENT, L.L.C.
Abbreviations
- ALT
alanine aminotransferase
- AST
aspartate aminotransferase
- APRI
AST to Platelet Ratio Index
- BMI
body mass index
- FIB-4
Fibrosis-4
- FNI
Fibrotic NASH Index
- HbA1c
glycated hemoglobin
- MAFLD
metabolic associated fatty liver disease
- NASH
nonalcoholic steatohepatitis
- NFS
NAFLD Fibrosis Score
- OR
odds ratio
- SGLT-2
sodium-glucose cotransporter 2
- T2DM
type 2 diabetes mellitus
Contributor Information
Angel N Borisov, Department of Gastroenterology and Hepatology, University Center for Gastrointestinal and Liver Diseases, CH-4031 Basel, Switzerland; Division of Endocrinology, Diabetes, and Metabolism, University Hospital Basel, 4031 Basel, Switzerland.
Alexander Kutz, Division of Endocrinology, Diabetes, and Metabolism, University Department of Medicine, Kantonsspital Aarau, 5001 Aarau, Switzerland; Division of General and Emergency Medicine, University Department of Medicine, Kantonsspital Aarau, 5001 Aarau, Switzerland; Division of Pharmacoepidemiology and Pharmacoeconomics, Department of Medicine, Brigham and Women's Hospital and Harvard Medical School, Boston, Massachusetts, USA.
Emanuel R Christ, Division of Endocrinology, Diabetes, and Metabolism, University Hospital Basel, 4031 Basel, Switzerland.
Markus H Heim, Department of Gastroenterology and Hepatology, University Center for Gastrointestinal and Liver Diseases, CH-4031 Basel, Switzerland.
Fahim Ebrahimi, Department of Gastroenterology and Hepatology, University Center for Gastrointestinal and Liver Diseases, CH-4031 Basel, Switzerland; Division of Endocrinology, Diabetes, and Metabolism, University Hospital Basel, 4031 Basel, Switzerland; Department of Medical Epidemiology and Biostatistics, Karolinska Institutet, 17165 Stockholm, Sweden.
Funding
F.E. was supported by the Swiss National Science Foundation (P500PM_210866). A.K. was supported by the Swiss National Science Foundation (P400PM 194479/1). The funding sources had no role in study design, data collection, data analysis, data interpretation, or writing of the report.
Author Contributions
A.B., E.R.C., and F.E. designed the study. A.B. and F.E. analyzed the data and were responsible for the decision to submit the manuscript. All authors provided substantial comments on drafts and approved the final report.
Disclosures
A.N.B., E.R.C., and M.H.H. have nothing to declare. F.E. has previously consulted for Boehringer Ingelheim. A.K. received an educational grant from Novo Nordisk.
Data Availability
Restrictions apply to the availability of some or all data generated or analyzed during this study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.
Clinical Trial Information
ClinicalTrials.gov identification nos. NCT01032629 (CANVAS) and NCT01989754 (CANVAS-R)
References
- 1. Eslam M, Sanyal AJ, George J, et al. MAFLD: A consensus-driven proposed nomenclature for metabolic associated fatty liver disease. Gastroenterology. 2020;158(7):1999‐2014.e1. [DOI] [PubMed] [Google Scholar]
- 2. Lomonaco R, Bril F, Portillo-Sanchez P, et al. Metabolic impact of nonalcoholic steatohepatitis in obese patients with type 2 diabetes. Diabetes Care. 2016;39(4):632‐638. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Riazi K, Azhari H, Charette JH, et al. The prevalence and incidence of NAFLD worldwide: a systematic review and meta-analysis. Lancet Gastroenterol Hepatol. 2022;7(9):851‐861. [DOI] [PubMed] [Google Scholar]
- 4. Golabi P, Rhea L, Henry L, Younossi ZM. Hepatocellular carcinoma and non-alcoholic fatty liver disease. Hepatol Int. 2019;13(6):688‐694. [DOI] [PubMed] [Google Scholar]
- 5. Sanyal AJ, Van Natta ML, Clark J, et al. Prospective study of outcomes in adults with nonalcoholic fatty liver disease. N Engl J Med. 2021;385(17):1559‐1569. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Simon TG, Roelstraete B, Khalili H, Hagström H, Ludvigsson JF. Mortality in biopsy-confirmed nonalcoholic fatty liver disease: results from a nationwide cohort. Gut. 2021;70(7):1375‐1382. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Park H, Dawwas GK, Liu X, Nguyen MH. Nonalcoholic fatty liver disease increases risk of incident advanced chronic kidney disease: a propensity-matched cohort study. J Intern Med. 2019;286(6):711‐722. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Vuppalanchi R, Noureddin M, Alkhouri N, Sanyal AJ. Therapeutic pipeline in nonalcoholic steatohepatitis. Nat Rev Gastroenterol Hepatol. 2021;18(6):373‐392. [DOI] [PubMed] [Google Scholar]
- 9. Sumida Y, Yoneda M. Current and future pharmacological therapies for NAFLD/NASH. J Gastroenterol. 2018;53(3):362‐376. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Petroni ML, Brodosi L, Bugianesi E, Marchesini G. Management of non-alcoholic fatty liver disease. BMJ. 2021;372:m4747. [DOI] [PubMed] [Google Scholar]
- 11. Neal B, Perkovic V, Mahaffey KW, et al. Canagliflozin and cardiovascular and renal events in type 2 diabetes. N Engl J Med. 2017;377(7):644‐657. [DOI] [PubMed] [Google Scholar]
- 12. Wiviott SD, Raz I, Bonaca MP, et al. Dapagliflozin and cardiovascular outcomes in type 2 diabetes. N Engl J Med. 2019;380(4):347‐357. [DOI] [PubMed] [Google Scholar]
- 13. Marx N, Davies MJ, Grant PJ, et al. Guideline recommendations and the positioning of newer drugs in type 2 diabetes care. Lancet Diabetes Endocrinol. 2021;9(1):46‐52. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Bonnet F, Scheen AJ. Effects of SGLT2 inhibitors on systemic and tissue low-grade inflammation: the potential contribution to diabetes complications and cardiovascular disease. Diabetes Metab. 2018;44(6):457‐464. [DOI] [PubMed] [Google Scholar]
- 15. Akuta N, Kawamura Y, Watanabe C, et al. Impact of sodium glucose cotransporter 2 inhibitor on histological features and glucose metabolism of non-alcoholic fatty liver disease complicated by diabetes mellitus. Hepatology Res. 2019;49(5):531‐539. [DOI] [PubMed] [Google Scholar]
- 16. Raj H, Durgia H, Palui R, et al. SGLT-2 inhibitors in non-alcoholic fatty liver disease patients with type 2 diabetes mellitus: A systematic review. World J Diabetes. 2019;10(2):114‐132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. World Medical Association . Declaration of Helsinki. Ethical principles for medical research involving human subjects. vol. 107. 2009. 10.1515/9783110208856.233 [DOI] [Google Scholar]
- 18. Rinella ME, Dufour JF, Anstee QM, et al. Non-invasive evaluation of response to obeticholic acid in patients with NASH: results from the REGENERATE study. J Hepatol. 2022;76(3):536‐548. [DOI] [PubMed] [Google Scholar]
- 19. Mózes FE, Lee JA, Selvaraj EA, et al. Diagnostic accuracy of non-invasive tests for advanced fibrosis in patients with NAFLD: an individual patient data meta-analysis. Gut. 2021;71(5):1006‐1019. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Angulo P, Hui JM, Marchesini G, et al. The NAFLD fibrosis score: A noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology. 2007;45(4):846‐854. [DOI] [PubMed] [Google Scholar]
- 21. Sterling RK, Lissen E, Clumeck N, et al. Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology. 2006;43(6):1317‐1325. [DOI] [PubMed] [Google Scholar]
- 22. Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology. 2003;38(2):518‐526. [DOI] [PubMed] [Google Scholar]
- 23. Tavaglione F, Jamialahmadi O, De Vincentis A, et al. Development and validation of a score for fibrotic nonalcoholic steatohepatitis. Clin Gastroenterol Hepatol. 2022:S1542-3565(22)00385-8. [DOI] [PubMed] [Google Scholar]
- 24. Lee JH, Kim D, Kim HJ, et al. Hepatic steatosis index: a simple screening tool reflecting nonalcoholic fatty liver disease. Dig Liver Dis. 2010;42(7):503‐508. [DOI] [PubMed] [Google Scholar]
- 25. Sviklāne L, Olmane E, Dzērve Z, Kupčs K, Pīrāgs V, Sokolovska J. Fatty liver index and hepatic steatosis index for prediction of non-alcoholic fatty liver disease in type 1 diabetes. J Gastroenterol Hepatol. 2018;33(1):270‐276. [DOI] [PubMed] [Google Scholar]
- 26. Morieri ML, Vitturi N, Avogaro A, et al. Prevalence of hepatic steatosis in patients with type 2 diabetes and response to glucose-lowering treatments. A multicenter retrospective study in Italian specialist care. J Endocrinol Invest. 2021;44(9):1879‐1889. [DOI] [PubMed] [Google Scholar]
- 27. Borisov A, Kutz A, Christ E, Heim M, Ebrahimi F. Supplementary appendix: canagliflozin and metabolic associated fatty liver disease in patients with diabetes mellitus: new insights from CANVAS. Mendeley Data. 2023. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Anstee QM, Targher G, Day CP. Progression of NAFLD to diabetes mellitus, cardiovascular disease or cirrhosis. Nat Rev Gastroenterol Hepatol. 2013;10(6):330‐344. [DOI] [PubMed] [Google Scholar]
- 29. Yki-Järvinen H. Non-alcoholic fatty liver disease as a cause and a consequence of metabolic syndrome. Lancet Diabet Endocrinol. 2014;2(11):901‐910. [DOI] [PubMed] [Google Scholar]
- 30. Kwok R, Choi KC, Wong GLH, et al. Screening diabetic patients for non-alcoholic fatty liver disease with controlled attenuation parameter and liver stiffness measurements: A prospective cohort study. Gut. 2016;65(8):1359‐1368. [DOI] [PubMed] [Google Scholar]
- 31. Younossi ZM, Otgonsuren M, Venkatesan C, Mishra A. In patients with non-alcoholic fatty liver disease, metabolically abnormal individuals are at a higher risk for mortality while metabolically normal individuals are not. Metab Clin Exp. 2013;62(3):352‐360. [DOI] [PubMed] [Google Scholar]
- 32. Chitturi S, Abeygunasekera S, Farrell GC, et al. NASH And insulin resistance: insulin hypersecretion and specific association with the insulin resistance syndrome. Hepatology. 2002;35(2):373‐379. [DOI] [PubMed] [Google Scholar]
- 33. Fan N, Zhang L, Xia Z, Peng L, Wang Y, Peng Y. Sex-specific association between serum uric acid and nonalcoholic fatty liver disease in type 2 diabetic patients. J Diabetes Res. 2016;2016:3805372. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Lassailly G, Caiazzo R, Buob D, et al. Bariatric surgery reduces features of nonalcoholic steatohepatitis in morbidly obese patients. Gastroenterology. 2015;149(2):379‐388. [DOI] [PubMed] [Google Scholar]
- 35. Blais P, Husain N, Kramer JR, Kowalkowski M, El-Serag H, Kanwal F. Nonalcoholic fatty liver disease is underrecognized in the primary care setting. Am J Gastroenterol. 2015;110(1):10‐14. [DOI] [PubMed] [Google Scholar]
- 36. Budd J, Cusi K. Nonalcoholic fatty liver disease: what does the primary care physician need to know? Am J Med. 2020;133(5):536‐543. [DOI] [PubMed] [Google Scholar]
- 37. Zafar Y, Rashid AM, Siddiqi AK, et al. Effect of novel glucose lowering agents on non-alcoholic fatty liver disease: a systematic review and meta-analysis. Clin Res Hepatol Gastroenterol. 2022;46(7):101970. [DOI] [PubMed] [Google Scholar]
- 38. Yoneda M, Honda Y, Ogawa Y, et al. Comparing the effects of tofogliflozin and pioglitazone in non-alcoholic fatty liver disease patients with type 2 diabetes mellitus (ToPiND study): a randomized prospective open-label controlled trial. BMJ Open Diabetes Res Care. 2021;9(1):e001990. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Cusi K, Bril F, Barb D, et al. Effect of canagliflozin treatment on hepatic triglyceride content and glucose metabolism in patients with type 2 diabetes. Diabetes Obes Metab. 2019;21(4):812‐821. [DOI] [PubMed] [Google Scholar]
- 40. Ratziu V. A critical review of endpoints for non-cirrhotic NASH therapeutic trials. J Hepatol. 2018;68(2):353‐361. [DOI] [PubMed] [Google Scholar]
- 41. Vilar-Gomez E, Calzadilla-Bertot L, Friedman SL, et al. Serum biomarkers can predict a change in liver fibrosis 1 year after lifestyle intervention for biopsy-proven NASH. Liver Int. 2017;37(12):1887‐1896. [DOI] [PubMed] [Google Scholar]
- 42. Blank V, Petroff D, Beer S, et al. Current NAFLD guidelines for risk stratification in diabetic patients have poor diagnostic discrimination. Sci Rep. 2020;10(1):18345. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Akuta N, Kawamura Y, Fujiyama S, et al. SGLT2 Inhibitor treatment outcome in nonalcoholic fatty liver disease complicated with diabetes mellitus: the long-term effects on clinical features and liver histopathology. Internal Medicine. 2020;59(16):1931‐1937. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44. Takeshita Y, Honda M, Harada K, et al. Comparison of tofogliflozin and glimepiride effects on nonalcoholic fatty liver disease in participants with type 2 diabetes: a randomized, 48-week, open-label, active-controlled trial. 2064 Diabetes Care. 2022;45(9):2064‐2075. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 45. Osataphan S, Macchi C, Singhal G, et al. SGLT2 Inhibition reprograms systemic metabolism via FGF21-dependent and -independent mechanisms. JCI Insight. 2019;4(5):e123130. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 46. Simon TG, Roelstraete B, Hagström H, Sundström J, Ludvigsson JF. Non-alcoholic fatty liver disease and incident major adverse cardiovascular events: results from a nationwide histology cohort. Gut. 2022;71(9):1867‐1875. [DOI] [PubMed] [Google Scholar]
- 47. Sanyal AJ, Harrison SA, Ratziu V, et al. The natural history of advanced fibrosis due to nonalcoholic steatohepatitis: data from the simtuzumab trials. Hepatology. 2019;70(6):1913‐1927. [DOI] [PubMed] [Google Scholar]
- 48. Sinn DH, Kang D, Jang HR, et al. Development of chronic kidney disease in patients with non-alcoholic fatty liver disease: a cohort study. J Hepatol. 2017;67(6):1274‐1280. [DOI] [PubMed] [Google Scholar]
- 49. Kramer JR, Natarajan Y, Dai J, et al. Effect of diabetes medications and glycemic control on risk of hepatocellular cancer in patients with nonalcoholic fatty liver disease. Hepatology. 2022;75(6):1420‐1428. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 50. Xia MF, Yki-Järvinen H, Bian H, et al. Influence of ethnicity on the accuracy of non-invasive scores predicting non-alcoholic fatty liver disease. PLoS One. 2016;11(8):e0160526. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
Restrictions apply to the availability of some or all data generated or analyzed during this study to preserve patient confidentiality or because they were used under license. The corresponding author will on request detail the restrictions and any conditions under which access to some data may be provided.