Role of SGLT-2 Inhibitors in Treatment of Patients with Type 2 Diabetes and NAFLD

Combination Therapy

In a small (n=44) randomized trial of 26 week-duration, Han et al [20] showed that addition of ipragliflozin (50 mg/d) to ongoing pioglitazone therapy further decreased IHFC assessed by transient liver elastography. In a post-hoc analysis of a randomized trial (DURATION-8), Gastaldelli et al [21] compared efficacy of dapagliflozin 10 mg/d, the GLP-1 agonist exenatide (2 mg once weekly by subcutaneous injection), or their combination in reducing severity of NAFLD. After 52 weeks, they found that combination therapy was superior to each treatment alone in ameliorating serum markers of hepatic steatosis and fibrosis [21].

Histological studies

Unfortunately, data from histological studies are very limited. Akuta et al [22] examined the effects of canagliflozin 100 mg bid in 5 patients with type 2 diabetes and histological evidence of NAFLD (defined as steatosis in 5% or more of hepatocytes). After 24 weeks, repeated liver biopsies showed improvement in the NAFLD activity score, which represents the sum of steatosis, lobular inflammation, and hepatocyte ballooning scores in all 5 patients [22]. In another pilot study, Lai et al [23] evaluated empagliflozin 25 mg/d in 9 patients with type 2 diabetes and confirmed NAFLD. These authors used historical placebo group for comparison. After 24 weeks, all histological histologic outcomes (steatosis, fibrosis, ballooning), either remained unchanged or improved, except in one patient who had worsening ballooning [23]. Thus, compared with historical placebo, empagliflozin therapy was associated with significantly greater improvements in steatosis (67% vs 26%, P=0.025), ballooning (78% vs 34%, P =0.024), and fibrosis (44% vs 6%, P=0.008) [23]. These preliminary data are encouraging. Clearly, they require confirmation by randomized double-blind and placebo-controlled histological studies.

Effects of SGLT-2 inhibitors on NAFLD in patients without diabetes

In one randomized double-blind and placebo controlled, Taheri et al [24] evaluated effects of small dose empagliflozin (10 mg/d) on the degree of steatosis in 90 patients without diabetes by liver FibroScan. After 24 weeks, the percentage of patients of patients with improved steatosis was significantly greater in the empagliflozin group than the placebo group, 37.2% and 17%, respectively (P=0.035) [24]. Of note, this significant difference was observed only in the subgroup of patients (n=44) with significant steatosis at baseline [24]. Furthermore, fibrosis score was significantly improved in the empagliflozin group vs placebo (P==0.039) [24].

Effects of SGLT-2 inhibitors on transaminases

Elevation of serum aminotransaminases is considered a surrogate marker of NAFLD. However, alanine aminotransferase (ALT) and aspartate aminotransferase (AST) can be normal in many cases of NAFLD including patients with advanced fibrosis [25]. As seen in table 1, effects of SGLT-2 inhibitors on transaminases are variable, with no significant effect in most studies. A meta-analysis of randomized trials showed that treatment with SGLT-2 inhibitors significantly decreased levels of ALT (9 trials), and gammaglutamyl transferase (GGT) (6 trials), but not AST (9 trials) [3].

Predictors to favorable response to SGLT-2 inhibitors

Preliminary data from placebo-controlled trials suggest that the following 3 factors may predict a favorable response to SGLT-2 inhibitors in terms of reduction in IHFC. First, the magnitude of weight loss as shown in most studies [9, 11]. Second, the decrease in hepatic fat might be more pronounced in patients with baseline NAFLD [11, 24]. Third, response to SGLT-2 inhibitors with respect to gender was only analyzed in the study by Kahl et al [9]. In that study, there was only significant decrease in IHFC among males [-31% (95% -44 to -14; P=0.002)], but not females [-1% (95% CI -28 to +37); P=0.96] [9].

Possible mechanisms of reduction of hepatic fat by SGLT-2 inhibitors

The exact mechanisms whereby SGLT-2 inhibitors may decrease IHFC are not fully elucidated. Possible mechanisms are discussed below.

Weight loss: While the use of SGLT-2 inhibitors is known to induce a weight low of approximately 2.5 kg [9, 11, 13], this minor degree of weight loss appears to be a major factor in decreasing liver fat. For example, in the study of Kahl et al [9], the difference in liver fat between the empagliflozin group and placebo group was no longer statistically significant after adjustment for change in body weight. Moreover, in the study of Cusi et al [11], significant correlation between weight loss and decrease in IHFC (correlation coefficient r =0.58, P<0.001) was observed in patients randomized to canagliflozin as well as those randomized to placebo. Conversely, Kuchay et al [15] did not find significant correlation between weight reduction and liver fat reduction (r=0.218, P=0.329). Thus, it is possible that factors other than weight loss could be also involved.

Effect on insulin resistance: Insulin resistance plays a major role in the pathogenesis of both type 2 diabetes and NAFLD [26]. Meanwhile, there is no evidence that SGLT-2 inhibitors improve insulin sensitivity either in the whole body [8] or in various tissues such as the liver, adipose tissue and muscle [9]. In addition, Cusi et al [11] did not find any effect of canagliflozin on insulin resistance in muscle and adipose tissue. Although these authors reported improvement of insulin sensitivity at the level of hepatic tissue by canagliflozin, the drug had no significant effect on IHFC [11]. Glycemic control: The design of 2 trials allowed testing the hypothesis whether amelioration of glycemic status by SGLT-2 inhibitors may have a role in decreasing IHFC. Thus, in these 2 studies, the authors maintained similar HbA1c values throughout the trials both in patients with controlled type 2 diabetes in the study of Kahl et al [9], and patients with uncontrolled type 2 diabetes in the study of Kuchay et al [15]. In both studies, SGLT-2 inhibitors decreased IHFC despite similar glycemic control between patient groups. Another finding that supports the concept that reduction in IHFC by SGLT-2 inhibitors occurs independently of glycemic control is the observation showing that empagliflozin may decrease steatosis and fibrosis in patients without diabetes [24].

Other mechanisms: Other potential mechanisms include decrease in insulin levels by SGLT-2 inhibitors leading to reduction in hepatic de novo lipid synthesis and increase in glucagon levels. The latter leads to stimulation of hepatic ß-oxidation of fatty acids [26, 27]. Further investigations are required to study the relative contribution of these mechanisms to the decrease in IHFC by SGLT-2 inhibitors.

Conclusions And Future Needs

Preliminary data suggest that SGLT-2 inhibitors may have beneficial effects for treatment of NAFLD. Most placebo-controlled studies relied on MRI-imaging modalities to assess changes of IHFC after drug intervention. This methodology suffers from important limitations. Indeed, by analysis of data from 121 patients with paired liver biopsies and MRI, Bril et al [28] concluded that quantification of liver fat with MRI techniques may be misleading as a surrogate marker of treatment response and may not accurately predict histological improvement of steatohepatitis after treatment. In fact, liver biopsy remains the gold standard method to confirm diagnosis of NAFLD and evaluate its response to different therapeutic interventions. In that respect, the study of Dapagliflozin Efficacy and Action in NASH (DEAN) (NCT03723252) is an ongoing phase 3, randomized, placebocontrolled trial that will recruit 100 biopsy-proven NASH patients with type 2 diabetes. The primary outcome of the DEAN trial is to examine the effect of dapagliflozin (10 mg/d) on hepatic histological lesions after 12 months compared with placebo. Combat T2 NASH is another randomized 48 week- trial that compares empagliflozin with the once-weekly glucagon-like peptide-1 (GLP-1) agonist semaglutide and placebo on histological liver changes in patients with type 2 diabetes. Since SGLT-2 inhibitors were used with success in patients without diabetes for cardiac protection [4], it is worth while to extend biopsy studies to patients with NAFLD without diabetes. Results of these trials should determine with more certainty the role of SGLT-2 inhibitors for treatment of NAFLD in general.