Non-alcoholic fatty liver disease in type 2 diabetes: Emerging evidence of benefit of peroxisome proliferator-activated receptors agonists and incretin-based therapies

Abstract

Nonalcoholic fatty liver disease (NAFLD) is a global epidemic, affecting more than half of the people living with type 2 diabetes (T2D). The relationship between NAFLD and T2D is bidirectional and the presence of one perpetuates the other, which significantly increases the hepatic as well as extrahepatic complications. Until recently, there was no approved pharmacological treatment for NAFLD/ nonalcoholic steatohepatitits (NASH). However, there is evidence that drugs used for diabetes may have beneficial effects on NAFLD. Insulin sensitizers acting through peroxisome proliferator-activated receptor (PPAR) modulation act on multiple levels of NAFLD pathogenesis. Pioglitazone (PPARγ agonist) and saroglitazar (PPARα/γ agonist) are particularly beneficial and recommended by several authoritative bodies for treating NAFLD in T2D, although data on biopsy-proven NASH are lacking with the latter. Initial data on elafibanor (PPAR α/δ agonist) and Lanifibranor (pan PPAR agonist) are promising. On the other hand, incretin therapies based on glucagon-like peptide-1 (GLP-1) receptor agonists (GLP-1RA) and dual- and triple-hormone receptor co-agonists reported impressive weight loss and may have anti-inflammatory and antifibrotic properties. GLP-1 RAs have shown beneficial effects on NAFLD/NASH and more studies on potential direct effects on liver function by dual- and triple-agonists are required. Furthermore, the long-term safety of these therapies in NAFLD needs to be established. Collaborative efforts among healthcare providers such as primary care doctors, hepatologists, and endocrinologists are warranted for selecting patients for the best possible management of NAFLD in T2D.

Key Words: Non-alcoholic fatty liver disease; Type 2 diabetes; Evidence; PPAR agonists; Incretin-based therapies

Core Tip: Co-existence of non-alcoholic fatty liver disease (NAFLD) and type 2 diabetes mellitus (T2DM) synergistically act to increase the risk of adverse hepatic and extra hepatic outcomes. T2DM is an established risk factor for NAFLD progression to NASH, advanced fibrosis or cirrhosis. Timely intervention in these populations could have a significant effect on liver- related outcomes. Newer dual and pan-PPAR agonists show promising results in patients with NAFLD/NASH and T2DM. Incretin-based therapy for the treatment of NAFLD is currently being explored. With better understanding of the complex interaction between T2DM and NAFLD, PPAR agonists and incretin-based therapies are likely to provide more effective approach of NAFLD management in T2DM.

INTRODUCTION

The coexistence of non-alcoholic fatty liver disease (NAFLD) and type 2 diabetes (T2D) synergistically act to increase the risk of adverse hepatic and extrahepatic outcomes. The global prevalence of NAFLD in people with T2D has been estimated at nearly 56% in a large meta-analysis of observational studies. More than one-third of people with T2D have nonalcoholic steatohepatitis (NASH), and around 1 in 6 carries advanced fibrosis[1]. T2D is an established risk factor for NAFLD progression to NASH, advanced fibrosis, or cirrhosis. Considering the prospective clinical implications of T2D in NAFLD, timely intervention in these patients could have a significant effect on liver-related outcomes. NAFLD and T2D share underlying altered pathophysiological mechanisms, including insulin resistance (IR). Expectedly, some anti-diabetic drug classes might exert favorable effects on NAFLD, with prevention of adverse cardiovascular (CV) events as well[2]. However, an unmet need for more focused research is there on how glycemic control through insulin sensitizers or incretin-based therapies might halt or revert the progression of NAFLD in individuals with T2D. The present mini-review summarizes the evidence on strong and significant association between T2D and the risk of adverse clinical outcomes in NAFLD, the therapeutic potential of peroxisome proliferator-activated receptors (PPAR) agonists for NASH and NASH-related hepatic morbidity, and lastly the key scientific advances regarding the use of incretins for treating NAFLD.

NAFLD AND T2D- THE COMPLEX LINK

From diabetes to NAFLD/NASH

The association between T2D and NAFLD is more complex than earlier thought and the relationship appears to be bidirectional[3]. The underlying pathophysiological mechanisms for the development of NAFLD are primarily the glucose and lipid metabolism alterations, with IR, elucidating the close link between T2D and NAFLD. Increased adiposity, often found in NAFLD and T2D, is coupled with adipocyte IR and dysfunction. Dysfunctional adipose tissue (AT) exhibits resistance to the antilipolytic effect of insulin. The consequent enhanced lipolysis and release of free fatty acids (FFAs) and glycerol are responsible for the accumulation of triglyceride and lipotoxicity in the liver, and other organs. Overflow of FAs to the liver also results in increased cellular levels of toxic lipids such as diacylglycerols, ceramides, and long-chain fatty acyl-coenzyme A (CoA), which are involved in inflammatory pathways (Figure 1)[4]. Excess FFAs also induce mitochondrial dysfunction, an increase in oxidative stress, and uncoupled oxidative phosphorylation. A fibrogenic response in hepatic stellate cells may be activated that can stimulate the progression to NASH and cirrhosis, and reactive oxygen species production (Figure 1)[5]. Dysfunction of AT and IR leads to the enhanced release of pro-inflammatory cytokines and decreased release of anti-inflammatory adipokines by AT. These cytokines can damage the liver directly, or indirectly, by raising oxidative stress, liver fibrosis, and the development of tumors[6,7]. Being closely interrelated, both glucotoxicity and lipotoxicity contribute to worsening IR and impaired insulin secretion (Figure 1)[8]. Hyperglycemia induces hepatic de-novo lipogenesis (DNL). Saturated fatty acid palmitate, the primary product of lipogenesis can induce hepatic inflammation and endoplasmic reticulum stress. Persistently elevated blood glucose levels promote DNL through direct and indirect mechanisms: the direct mechanism is by enhancing the activity of TCA cycle and Acyl CoA synthesis that acts as a substrate of both DNL and gluconeogenesis, and the indirect mechanism is through the transcription factor carbohydrate response element binding protein and liver X receptor αwhich sequentially promote unfavorable gene transcriptions.

Figure 1 Relationship between lipo- and glucotoxicity, insulin resistance, and beta-cell function. DAG: Diacylglycerol; DNL: De novo lipogenesis; NAFLD: Non-alcoholic fatty liver disease; ROS: Reactive oxygen species; T2D: Type 2 diabetes.

Progression of NAFLD in T2D

T2D is one of the strongest risk factors for the progression of NAFLD to NASH, cirrhosis, or hepatocellular carcinoma (HCC). An unhealthy lifestyle, via the promotion of gut dysbiosis and T2D-associated hyperglycemia and hyperinsulinemia, influences the transition from isolated steatosis to NASH and the progression of NASH. T2D-associated hyperglycemia also promotes the progression from NASH to hepatic fibrosis and HCC (Figure 2)[8]. Swedish National Diabetes Registry showed that severe liver disease risk is higher in T2D patients than in the general population. In people with T2D, risk factors associated with severe liver disease were older age, male gender, higher body mass index, smoking, hypertension, and microalbuminuria[9]. Among persons with NAFLD, those with T2D are at greater risk of developing advanced stages of fibrosis and subsequently hepatic decompensation, HCC, and liver-related mortality than those without T2D[10,11]. Huang et al[12] report the findings of a recent large meta-analysis comprising 2016 ethnically diverse participants. Compared with individuals without diabetes, participants with T2D had an elevated risk of hepatic decompensation and HCC despite adjustment for multiple factors having a known association with the progression of NAFLD. In general, liver fibrosis progresses by one stage over 7 years for people with NASH, but data are limited regarding the time to fibrosis progression in people with T2D compared with people without T2D[13]. In a recent comprehensive, multicentric study of patients with paired liver biopsies from the Nonalcoholic Steatohepatitis Clinical Research Network, the authors found that the cumulative incidence of fibrosis progression was considerably higher in participants with T2D compared with individuals without T2D. T2D remained an important predictor of the progression of fibrosis, even after controlling for several potential confounders[14]. Altogether, these findings provide convincing clinical data to point out that T2D is associated with advanced-stage liver disease, faster disease progression, and worse hepatic outcomes in people with NAFLD. Consequently, it is compulsory to screen for the presence and severity of NAFLD in all adults with T2D with liver enzymes in a wider range than usually considered[15]. Poor glycemic control has been linked to severe long-term complications, including advanced liver disease, and adverse outcomes in NAFLD; however, an ideal HbA1c cutoff value to detect people at risk of those outcomes is yet to be identified.

Figure 2 Modifiable and non-modifiable risk factors of non-alcoholic fatty liver disease and its progression to severe liver disease. Dashed arrows indicate factors that promote or predispose to disease. Among non-modifiable risk factors, most important genetic variants of nonalcoholic fatty liver disease are being extensively studied. Type 2 diabetes-associated hyperglycemia induces progression from nonalcoholic steatohepatitis to hepatic fibrosis and HCC. CV: Cardiovascular; HCC: Hepatocellular carcinoma; NAFLD: Nonalcoholic fatty liver disease; NASH: Nonalcoholic steatohepatitis.

THERAPEUTIC ROLE OF PPAR AGONISTS

Rationale

PPARs are nuclear receptors for specific endogenous ligands and act as transcription factors involved in the regulation of lipid metabolism, glucose homeostasis, and inflammation[16]. Endogenous ligands include free fatty acids (FAs), eicosanoids, and various complex lipids. PPAR forms a heterodimer with retinoid X receptor once ligands bind to it, and the complex binds to the response element that regulates the gene expression which is important for fatty acid metabolism[17]. There are three PPAR isoforms, PPAR α is largely present in the liver involved in fatty acid oxidation, and exerts anti-inflammatory action through complex regulation of nuclear factor kappa-light-chain-enhancer of activated B cells[18]. PPAR β/δ is expressed mainly in skeletal muscle but is also present in adipose tissue and skin and has a significant role in controlling lipid metabolism[19]. PPAR γ is highly expressed in adipose tissue and plays an important role in the regulation of adipogenesis, adipocyte differentiation, and fatty acid metabolism[20]. Together, PPAR agonists have a role in insulin sensitivity, inflammation, and dyslipidemia and thus act on multiple levels in NAFLD pathogenesis. This makes it an attractive option for drug development.

PPAR single agonists

Given the differential actions of different PPAR agonists, there has been a significant effort in the development of agonists for single receptors. They have been tried in different conditions including NAFLD, the results of which are encouraging. The agents, and trials with biochemical and histological outcomes of single agonists are summarized in Table 1[21-27].

Table 1 Summary of studies of peroxisome proliferator-activated receptors single agonists in nonalcoholic fatty liver disease.

Drug class and mechanism	Agent	Trial design	Biochemical response	Histological response	Comments
PPARα (Fatty acid oxidation and Anti-inflammatory)[21,22]	Fenofibrate	16 patients, 48 wk vs placebo	Significant reduction in triglyceride and liver enzymes	Decreased ballooning, grade steatosis, inflammation/fibrosis no change	Limited efficacy
	Clofibrate	40 patients 1 yr vs UDCA	Reduced ALT	No change
PPAR β/δ (glucose homeostasis and insulin sensitivity)[23,24]	GW501516	6 patients, 2 wk vs placebo	Reduced TG and LDL	No data	Abandoned due to cancer risk in preclinical studies
	MBX-8025	181 overweight patients, 2 wk vs placebo	Favorable lipid profile and decreased liver enzymes	No data	Need more data
PPARγ (adipogenesis, insulin sensitization, fatty acid oxidation)[25-27]	Rosiglitazone	63 patients, 52 wk vs placebo	Normalized transaminase levels (38% vs 7%, P = 0.005)	Improved steatosis (47% vs 16%; P = 0.014), although only half of the patients responded, no change of other histologic parameters	Weight gain and painful swollen legs in rosiglitazone arm
	Pioglitazone	RCT, 61 patients, 12 months placebo or pioglitazone had paired biopsies	Improvement of ALT and GGT	Hepatocellular injury (P = 0.005), Mallory–Denk bodies (P = 0.004), and fibrosis (P = 0.05) were reduced in patients treated with pioglitazone	Weight gain with pioglitazone
		RCT, 259 patients pioglitazone vs vitamin E vs placebo 96 wk	Improvement of transaminases in the vitamin E and pioglitazone arm	Improvement in NASH as compared with placebo (pioglitazone P = 0.04), significant reductions in steatosis, lobular inflammation and fibrosis in pioglitazone arm	Weight gain in pioglitazone group

ALT: Alanine aminotransferase, GGT: Gamma glutamyl transferase, LDL: Low density lipoprotein, PPAR: Peroxisome proliferator-activated receptor, RCT: Randomized controlled trial, TG: Triglyceride, UDCA: Ursodeoxycolic acid.

PPARα agonists: Fibrates are considered as PPARα agonists, albeit weak, and have limited clinical efficacy in NAFLD. In a study by Fernández-Miranda et al[21] 16 patients with NAFLD were treated with fenofibrate 200 mg/d for 48 wk and found a significant reduction in triglyceride and other liver enzymes. However, there was no significant improvement in histologic parameters. Another fibrate, clofibrate was compared with ursodeoxycholic acid in 40 patients for 1 year but no advantage was found in the clofibrate group. Newer selective PPARα-specific agonists, known as selective PPAR modulators, are in different phases of development[28,29].

PPAR β/δ agonists: Data is sparse with these PPAR isotypes. Two agents GW501516 and MBX-8025 have been tried in short trials[24,25]. They have shown some benefit in metabolic parameters (Table 1) but no histologic data is available. Further studies are required to establish any role of these agents in NAFLD.

PPARγ agonists: Thiazolidinediones are strong PPARγ agonists widely investigated in NAFLD. They are insulin-sensitizing agents and have a significant role in lipid metabolism. Rosiglitazone was compared with placebo for 1-year therapy in FLIRT trial[25] and found to have improved steatosis (47% vs 16%) and transaminitis (38% vs 7%), but lacked improvement in fibrosis and NAFLD activity score. Weight gain and painful swollen legs were significant side effects in the rosiglitazone arm. In the PIVENS trial[27], Pioglitazone was compared with vitamin E and placebo for 96 wk in patients with NASH where pioglitazone demonstrated a reduction in hepatic steatosis, lobular inflammation, and hepatic enzymes. A meta-analysis including 3 high-quality randomized controlled trials (RCTs) comparing pioglitazone vs placebo in NAFLD showed significant improvement in liver fibrosis[30]. Currently, pioglitazone is recommended in persons with T2D and biopsy-proven NASH by the American Association of Clinical Endocrinologists (AACE)[31] and recommended in both diabetic and non-diabetic adult patients with or without fibrosis by the Indian National Association for the study of the liver (INASL)[32] (Strong recommendation, high strength of evidence). However, the use of PPAR agonists can be associated with poor bone quality (in post-menopausal females) and dose-dependent pedal edema and weight gain.

Dual and pan PPAR agonists

Combination therapy using dual or pan agonists is an attractive choice as the most beneficial effects on NAFLD can be seen with the least side effects. For example, they can produce an antihyperlipidemic (PPARα) effect with insulin sensitization (PPARγ) and increase β-oxidation in the liver and skeletal muscle (α and β/δ). The efficacy can also be improved by synergistic action of different isoforms on fatty acid metabolism and insulin sensitivity. On the other hand, the side effects seen with the individual class of drugs like heart failure or weight gain with pioglitazone can be minimized. The agents and trials with biochemical and histological outcomes of dual/pan agonists are summarized in Table 2[33-40].

Table 2 Summary of studies of peroxisome proliferator-activated receptors dual and pan agonists in nonalcoholic fatty liver disease.

Drug class and mechanism	Agent	Trial design	Biochemical response	Histological response	Comments
PPAR α/δ agonist[33]	Elafibranor	276 patients, Elafibranor 80 mg vs 120 mg vs placebo, 52 wk, GOLDEN trial	Liver enzymes, lipids, glucose profiles, and markers of systemic inflammation were significantly reduced in the elafibranor 120-mg group	Elafibranor 120 mg was superior to the placebo, with NASH resolution without worsening of fibrosis in 19% versus 12% in the placebo group (P = 0.045), based on a post hoc analysis for the modified definition	No change in primary outcome in intention to treat analysis
		1070 patients, Elafibranor 120 mg vs placebo, 72 wk, RESOLVE IT trial	Improvement in TG, ALT, and GGT	138 (19.2%) patients in the Ela group and 52 (14.7%) patients in the placebo group achieved resolution of NASH without worsening of fibrosis (P = 0.066)	Despite the absence of safety signals, the RESOLVE-IT trial was discontinued due to the limited effect of Ela on surrogate efficacy endpoints
PPAR α/γ agonist[34-38]	Tesaglitazar, Muraglitazar, Aleglitazar		Favorable lipid profile with muraglitazar	No change	Trials terminated early due to nephrotoxicity, cardiotoxicity and gastrointestinal hemorrhage respectively
	Saroglitazar	106 patients, 16 wk vs placebo	Improvement in ALT and lipid profile	No data	DCGI approved for NAFLD in India
		85 patients, 12 wk vs placebo	Improvement in ALT and TG	Significant reduction in liver fibrosis (fibroscan)
Pan PPAR agonists[39,40]	Bezafibrate		Improve HbA1c and atherogenic dyslipidemia in mice	No data	Mostly animal data. Human studies are ongoing
	Lanifibranor		Improve insulin sensitivity in mice	Improve steatosis and fibrosis in liver tissue in mice

ALT: Alanine aminotransferase; DCGI: Drug Controller General of India; GGT: Gamma glutamyl transferase; PPAR: Peroxisome proliferator-activated receptor; TG: Triglyceride.

PPAR α/δ agonist: Elafibranor, an agonist of PPARα and PPARδ, works on insulin sensitivity, glucose homeostasis, and lipid metabolism. It has been studied in an RCT by Ratziu et al[33] (Golden trial) where two doses (80 mg and 120 mg) of Elafibranor were compared with a placebo. In the intention-to-treat analysis, there was no improvement in the resolution of NASH without worsening fibrosis (primary outcome). However, Elafibranor 120 mg became superior to placebo in 19% vs 12% in the placebo group (P = 0.045), based on a post hoc analysis for the modified definition of NASH resolution after excluding patients with mild steatohepatitis from the analysis. Another study RESOLVE IT (NCT02704403) was terminated early due to the limited effect of Elafibranor on surrogate efficacy endpoints. No safety signal was found in the studies.

PPAR α/γ agonists: Glitazars are PPAR α/γ agonists and improve dyslipidemia through action and insulin sensitivity through g action thus addressing two important issues of NAFLD. Initial glitazars, Tesaglitazar, Muraglitazar, and Aleglitazar, although showed significant improvement in dyslipidemia, trials for NFALD were terminated early due to nephrotoxicity, cardiotoxicity, and gastrointestinal hemorrhage respectively[34-36]. Saroglitazar is the new molecule in this class containing a unique pyrol moiety and lacks the glitazone ring and unlike its congeners has predominant PPARα action and moderate PPARγ action. Thus, it lacks the typical glitazone side effects. In mice models, saroglitazar has been shown to reduce histological NASH as compared with pioglitazone and fenofibrate[41]. Serum transaminase levels improved in patients with T2D treated with saroglitazar as compared with placebo[38,42]. A similar trial is ongoing at the United Nations with saroglitazar (NCT03061721) which will evaluate the improvement of serum transaminases and hepatic steatosis on MR imaging-estimated proton density fat fraction (MR-PDFF). Additionally, a recent phase III study (CTRI/2015/10/006236) in India is currently evaluating the histological efficacy of saroglitazar in comparison with a placebo in patients with biopsy-proven NASH.

Pan PPAR agonist: Pan PPAR agonists are still in various phases of development and many animal data are published. Bezafibrate and Lanifibranor have shown anti-lipid, anti-inflammatory, and antifibrotic properties in rats. An ongoing study is evaluating the efficacy of lanifibranor in patients with diabetes and NAFLD (NCT03459079). These agents may have a promising role in the management of NAFLD.

To conclude, given the multiple actions of the PPAR on lipid metabolism, oxidation of FAs, glucose homeostasis, and inflammation, PPAR agonists are attractive targets for the treatment of patients with NAFLD. Of all the PPAR agonists, the PPARα agonist pioglitazone is the most extensively evaluated and barring a few side effects, is most useful in patients with NAFLD and recommended by AACE and INASL guidelines. Emerging data of dual PPAR agonists and pan-PPAR agonists appear encouraging and may hold promise for patients with NAFLD.

THERAPEUTIC ROLE OF INCRETIN-BASED THERAPIES

In response to the intake of food, incretins play an important role in the regulation of blood glucose levels by enhancing insulin secretion from β-cells of the pancreas after meals. There are two main incretin hormones which are glucagon-like peptide-1 (GLP-1) and glucose-dependent insulinotropic polypeptide (GIP). GLP-1 is produced by L cells in the ileum and colon whereas GIP is produced from K cells in the duodenum and jejunum. The release of both is triggered as a response to the presence of nutrients (carbohydrates for both GLP1 and GIP, fat for GIP) in the digestive system. GLP-1 and GIP enhance glucose-dependent insulin release and inhibit glucagon secretion (glucagon stimulates the liver to release glucose) into the bloodstream. In addition, GLP-1 delays gastric emptying, regulates nutrient absorption, and promotes satiety whereas GIP promotes fat storage in adipose tissue[43].

Incretin-based therapies were originally developed for T2D which include GLP-1 receptor agonists (GLP-1 RAs) and dipeptidyl peptidase-4 (DPP-4) inhibitors. GLP-1 RAs (e.g., exenatide, liraglutide, semaglutide) bind to the GLP-1R and increase insulin secretion, inhibit glucagon release, slow gastric emptying, and promote satiety. DPP-4 inhibitors (sitagliptin, saxagliptin, linagliptin) increase the endogenous concentration of GLP-1 by inhibiting the DDP-4 enzyme which degrades GLP-1[44].

NASH is a multi-system disease with underlying metabolic stress driving its manifestations in several organs. The classical progression of NAFLD to NASH to cirrhosis and hepatocellular carcinoma is accelerated in T2D (Figure 3)[45]. Hence agents directed towards underlying metabolic dysfunction like incretin-based therapies could be helpful. These agents have pleiotropic effects apart from regulation of glucose metabolism such as anti-inflammatory and anti-fibrotic effects[46]. Several studies have assessed the clinical efficacy of incretin-based therapies in NASH.

Figure 3 Natural history of nonalcoholic fatty liver disease and nonalcoholic steatohepatitis. HCC: Hepatocellular carcinoma; NASH: Nonalcoholic steatohepatitis; NAFLD: Nonalcoholic fatty liver disease.

Mechanisms of action of incretin-based therapies in NASH

Insulin sensitivity improvement: GLP-1RAs act through augmenting incretins actions, namely GLP-1. GLP-1 is produced from the gut and has an impact on the pancreas to produce insulin and thus improves IR[47,48].

Effect on central nervous system: GLP-1 also has effects on central nervous system which reduce appetite and can lead to weight loss. GLP-1-based therapies are the mainstay of drug-induced weight loss. Within 6-12 months, semaglutide (a GLP-1 RA) induces 10%-15% weight loss comparable to bariatric procedures[49].

Anti-inflammatory effects and reduction of liver fat content: GLP-1 receptors are present in pancreas, gastrointestinal tract, CV system, kidney and many others organs. It promotes glucose uptake, and storage of glycogen in the liver and prevents gluconeogenesis[43]. Stimulation of FXR and LXR in the liver using GLP-1 RA may reduce liver fat as shown in animal models/human studies and inflammation.

Anti-Fibrotic Properties: GLP-1RAs like semaglutide have been shown to have dose dose-dependent trend for fibrosis reduction in NASH. With higher doses of GLP-1 RAs, significantly lower fibrosis worsening has been noted highlighting its potential for slowing NASH progression[50].

Improvement in the extra-hepatic milieu in NASH: GLP-1 has many biological effects other than those mentioned above. It has a cardio-protective effect and promotes diuresis/natriuresis in kidneys improving blood pressure (Figure 4). NASH is a multi-system disease with metabolic stress and systemic inflammation[46]. The prerequisites of NASH therapy are that it should improve liver disease and at least be neutral or beneficial with other comorbidities. Incretin-based therapies fulfill these criteria.

Figure 4 Pleotropic effects of glucagon-like peptide 1 and its effect of several organ systems in systematic inflammation characteristic of nonalcoholic steatohepatitis. GLP-1: Glucagon-like peptide 1.

Clinical evidence

Several clinical trials have explored the efficacy of incretin-based therapies in NASH (Table 3)[50-64]. The earlier studies were with liraglutide. The landmark LEAN study in 2016 showed that subcutaneous liraglutide 1.8 mg/d led to the resolution of biopsy-proven NASH in 39% of cases compared to 9% in placebo at 48 wk. Moreover, progression of fibrosis was seen in only 9% compared to 36% in placebo. GLP-1 RA also improved body weight, liver enzymes, and liver fat content[55].

Table 3 Summary of trials exploring the efficacy of incretin-based therapies in nonalcoholic steatohepatitis.

Ref.	Agent	Comparator	Participants	Duration	Outcomes
Newsome et al[50], 2021	Semaglutide 0.1, 0.2 or 0.4 mg s/c once daily	Placebo	NASH with F1, F2, or F3 fibrosis stage	72 wk	NASH resolution without worsening of fibrosis: 40% (0.1 mg), 36% (0.2 mg), 59% (0.4 mg), 33% (placebo); Improvement in fibrosis (P < 0.001 for 0.4 mg vs placebo); 43% (0.4 mg), 33% (placebo) (P = 0.48); Weight loss: 13% (0.4 mg), 1% placebo
Loomba et al[51], 2023	Semaglutide 2.4 mg s/c	Placebo (1: 2)	NASH with compensated cirrhosis	48 wk	Improvement in liver fibrosis by ≥ 1 stage without worsening NASH: 11% Semaglutide, 29% placebo (P = 0.087); Resolution of NASH at 48 wk: 34% semaglutide, 21% placebo (P = 0.29); No new safety concerns
Romero-Gomez et al[52], 2023	Efinopegdutide 10 mg s/c once weekly	Semaglutide 1 mg s/c once weekly	NAFLD with liver fat content (LFC) ≥ 10%	24 wk	Relative reduction in LFC from baseline higher with efinopegdutide (72.7%) than semaglutide (42.3%); Similar reduction in body weight
Alkhouri et al[53], 2022	Semaglutide 2.4 mg s/c + cilofexor once daily (30 or 100 mg)+ fircosostat (20 mg)	Semaglutide alone Semaglutide+cilofexor/fircosostat	NASH (MRI-PDFF > 10%, elastography ≥ 7 kPa	24 wk	Combination well tolerated; Greater reduction in liver fat with combination groups vs semaglutide (10%-11% vs 8.6%); Similar weight loss in all the groups
Flint et al[54], 2021	Semaglutide 0.4 mg s/c once daily	Placebo	NAFLD, steatosis ≥ 10% (MRI-PDFF), MRE: 2.5-4.63 kPa	48 wk	≥ 30% reduction in liver steatosis at 24, 48 and 72 wk (64.7%, 76.5%, 73.5%) significantly higher than placebo; Change in liver stiffness in NAFLD not significantly different compared to placebo; Improvement in liver enzymes, body weight and HbA1c
Armstrong et al[55], 2016	Liraglutide 1.8 mg/d s/c	Placebo	Biopsy proven NASH	48 wk	39% had resolution of NASH (vs 9% placebo); Progression of fibrosis: 9% Liraglutide, 36% placebo
Khoo et al[56], 2017	Liraglutide 3.0 mg/d	Diet and moderate exercise	Obesity and NAFLD (Liver fat content > 5% on MRI)	26 wk	Similar reduction in weight, liver fat, liver enzymes in both the groups without any significant difference between the two
Yan et al[57], 2019	Liraglutide 1.8 mg/d	Insulin glargine or sitagliptin	T2D and NAFLD	26 wk	Liraglutide and sitagliptin along with metformin reduced body weight, liver fat content, visceral adipose tissue in addition to glycemic control in contrast to Insulin, subcutaneous fat also decreased in liraglutide arm
Bizino et al[58], 2020	Liraglutide 1.8 mg/d	Placebo	T2D and NAFLD	26 wk	Liraglutide reduced significantly more body weight and subcutaneous fat but not visceral fat
Guo et al[59], 2020	Liraglutide 1.8 mg/d	Insulin glargine or placebo	T2D and NAFLD	26 wk	Liraglutide plus 2 gm metformin for 26 wk significantly reduced liver, subcutaneous and visceral fat
Zhang et al[60], 2020	Liraglutide 1.8 mg/d	Pioglitazone	T2D and NAFLD	24 wk	Liraglutide reduced liver fat significantly compare to pioglitazone which could be attributed to weight loss
Liu et al[61], 2020	Exenatide 1.8 mg/d	Insulin glargine	T2D and NAFLD	24 wk	Both reduced liver fat but exenatide led to higher reduction in body weight, visceral fat, liver enzymes
Miyake et al[62], 2022 (Trial Registration: jRCTs061210009)	Semaglutide 0.5 mg/wk + Luseogliflozin 2.5 mg/d	Semaglutide 0.5 mg/wk	T2D and NASH	52 wk	Ongoing study
Gastaldelli et al[63], 2022	Dual GIP/GLP-1R agonist: tirzepatide10 mg or 15 mg/wk	Insulin degludec	T2D and NAFLD	52 wk	Significant decrease in liver fat content, visceral and subcutaneous adipose tissue compared to insulin degludec
Kuchay et al[64], 2020	Glucagon-like peptide-1 receptor (GLP-1r) agonist (GLP-1 RA)	Usual care	T2D and NAFLD	24 wk	Dulaglutide significantly reduces liver fat comma gamma glutamyl transpeptidase

GIP: Glucose-dependent insulinotropic polypeptide; MRE: Magnetic resonance elastography; NAFLD: Nonalcoholic fatty liver disease; NASH: Nonalcoholic steatohepatitis; PDFF: Proton density fat fraction.

GLP-1 RAs also reduce visceral and subcutaneous fat which is not seen with insulin which only affects glycemic control. The reduction in liver fat content with liraglutide was higher than that of pioglitazone[56-60]. Similarly, exenatide 1.8 mg/d reduced liver fat with a higher reduction in body weight, visceral fat, and liver enzymes as compared to insulin glargine[61].

Semaglutide is longer acting GLP-1 RA (half-life 165 h as compared to 13-15 h with liraglutide) which is used as a once-weekly subcutaneous injection (2.4 mg) and is as effective as once-daily liraglutide (3 mg) for weight loss[51,52,65]. However, daily semaglutide with lower doses (up to 0.4 mg) has been used in NASH trials which have shown significant dose-dependent NASH resolution (most pronounced with 0.4 mg) without worsening fibrosis [50,51].

Future directions and implications for NASH management

The tremendous redundancy of the downstream pathways leading to NASH progression warrants combination therapies rather than using single agents. Blocking a single pathway was not shown to be very effective. The best effects are seen with agents that target the underlying metabolic dysregulation with pleiotropic effects such as incretins.

One such approach is combining semaglutide (GLP-1 RA) with small molecules such as firsocostat (acetyl-CoA carboxylase- rate-controlling enzyme for de novo lipogenesis) and cilofexor (second generation farnesoid X receptor agonist). The combination was safe and led to a greater reduction in liver fat compared to semaglutide or a combination of two agents.

GIP reduces ectopic fat accumulation and has anti-inflammatory effects (reduced macrophage polarization and cytokine production). More importantly, GIP ameliorates nausea induced by higher doses of GLP-1 allowing the latter to be used at higher doses. Dual GIP/GLP-1 receptor agonist tirzepatide is being studied in phase III trials for NASH (SYNERGY-NASH). Initial data in T2D and NAFLD showed a significantly higher decrease in liver, subcutaneous, and visceral fat as compared to insulin degludec[63]. Once weekly semaglutide (0.5 mg/wk) is being studied with sodium glucose co-transporter 2 inhibitors (Luseogliflozin 2.5 mg/d) for T2D with NASH[62].

GLP-1 is also being combined with glucagon for NASH. Glucagon has direct liver receptors whereas effects of GLP-1 RAs are based on extra-hepatic effects. Glucagon has hepatic receptors where it has anti-inflammatory effects, reduces bile acid production/hepatic stellate cell activation, inhibits transforming growth factor signaling, reduces collagen production, and increases turnover. Dulaglutide, a GLP-1 RA, is being studied in NASH. Earlier studies in T2D and NAFLD showed a reduction in liver fat[64]. GLP-1/glucagon receptor co-agonist efinopegdutide 10 mg weekly resulted in a greater reduction in liver fat than weekly semaglutide (1 mg) in a phase IIa study[52]. Histology-based trials in NASH are warranted.

Challenges and promise of incretin-based therapies for NASH

Selecting the right category of patients for incretin-based therapy needs further exploration. Disease activity (inflammation), stage (fibrosis), and comorbidities (e.g., T2D) need to be considered. The long-term safety of these therapies in NASH needs further study. The holistic approach of combination treatment including incretin-based therapies beyond glycemic control is a potential paradigm shift in NASH management and is an area of ongoing research. Collaborative efforts among healthcare providers such as primary care doctors, hepatologists, and endocrinologists are warranted.

CONCLUSION

In conclusion, specific targeted interventions are essential for NAFLD in T2D. This is because of the alarmingly high prevalence of NAFLD in T2D, their synergistic effect on hepatic and extra-hepatic complications, and the bidirectional link driven by alterations in glucose/Lipid metabolism along with insulin resistance. PPAR agonists such as pioglitazone have been shown to improve liver outcomes by regulating lipid metabolism, glucose homeostasis, and inflammation. Newer dual and pan-PPAR agonists show promising results with minimized side effects in NAFLD/NASH with T2D. Incretin-based therapies, (GLP-1RAs and DPP-4 inhibitors) act in T2D-related NAFLD/NASH by glycemic control as well their exhibit pleiotropic effects such as anti-inflammatory and anti-fibrotic actions. Liraglutide and semaglutide have demonstrated remarkable efficacy in resolving NASH without worsening fibrosis apart from improving metabolic parameters. Given the multifactorial nature of NASH pathogenesis, a combination of incretin-based therapies with small molecules targeting de novo lipogenesis and sodium glucose co-transporter 2 inhibitors are upcoming approaches for treating NASH. Dual GLP-1/glucagon receptor agonists with direct hepatic effects of glucagon affecting fibrosis could indicate a paradigm shift in NASH management. The current challenges include the proper selection of patients for incretin-based therapies, concerns regarding long-term safety, and the need for a multifaceted approach. As we better understand the intricate interplay between T2D and NAFLD, PPAR agonists and incretin-based therapies offer hope for more effective and targeted management of NAFLD/NASH in T2D.