Dapagliflozin in Metabolic Dysfunction–Associated Steatotic Liver Disease: Clinical Evidence and Therapeutic Potential

Abstract

Metabolic dysfunction–associated steatotic liver disease (MASLD) has emerged as the most common chronic liver disease worldwide, affecting nearly one-third of the global population and disproportionately impacting individuals with type 2 diabetes mellitus (T2DM) and obesity. The disease is driven by complex metabolic disturbances, including insulin resistance (IR), hepatic fat accumulation, oxidative stress, and chronic inflammation, which together promote progression from simple steatosis to metabolic dysfunction–associated steatohepatitis (MASH), fibrosis, and cirrhosis. Despite the growing burden of MASLD, therapeutic options remain limited and often constrained by cost, accessibility, or tolerability. Dapagliflozin, a sodium–glucose cotransporter-2 inhibitor originally developed for glycemic control in T2DM, has gained attention for its potential hepatoprotective effects. By inducing glucosuria and creating a caloric deficit, dapagliflozin promotes weight reduction, improves insulin sensitivity, and shifts energy metabolism toward fatty acid oxidation. These metabolic effects reduce hepatic de novo lipogenesis, attenuate oxidative stress, and may limit inflammatory pathways involved in MASLD progression. Emerging clinical evidence supports these mechanistic benefits. Prior randomized trials and observational studies have demonstrated improvements in liver enzymes, anthropometric parameters, and metabolic parameters, along with reduction in hepatic fat. Recent biopsy-based evidence, including findings from the DEAN trial, indicates significant improvements in histological endpoints of MASH. Collectively, current data suggest that dapagliflozin may offer a multifaceted therapeutic approach for MASLD by targeting both metabolic and hepatic pathways while providing established cardiovascular and renal benefits. Further large, multicenter trials involving diverse ethnicities, studying key histological endpoints as per regulatory guidance, may confirm these findings and define their role in the evolving treatment landscape of MASLD.

Introduction

Metabolic dysfunction–associated steatotic liver disease (MASLD) now accounts for the majority of chronic liver disease cases worldwide, currently impacting 30% of the global population.^1,2 About 65% of patients with type 2 diabetes mellitus (T2DM) and 75% of obese individuals have MASLD.^3,4 Expectedly, as the prevalence of T2DM increases globally, there will be a parallel significant increase in the burden of MASLD in these sub-populations.²

The pathophysiology of MASLD is complex, involving insulin resistance (IR), hepatic fat accumulation, oxidative stress, and chronic inflammation.⁵ Current therapeutic approaches are limited. Despite the latest approval of resmetirom, it is not available in many countries and imposes additional cost limitations wherever it is available.^6,7 Dapagliflozin, a sodium–glucose cotransporter 2 inhibitor (SGLT2i) originally developed for the management of T2DM, has demonstrated benefits beyond glycemic control and is guideline-recommended for patients with chronic kidney disease, heart failure (across the ejection fraction spectrum), and atherosclerotic cardiovascular disease with T2DM.^8–11 Its unique mechanism of action, promoting glucosuria, visceral fat accumulation, and subsequent metabolic reprogramming, has garnered interest in investigating its potential hepatoprotective effects in MASLD.^12,13 This has been further bolstered with recent biopsy-proven evidence in metabolic dysfunction–associated steatohepatitis (MASH).^14,15 This review summarizes the current evidence regarding dapagliflozin's role in MASLD and explores its therapeutic potential in addressing various facets of the disease.

MASLD Pathogenesis

Metabolic dysfunction–associated steatotic liver disease, earlier called nonalcoholic fatty liver disease (NAFLD), comprises a spectrum of steatotic liver disease characterized by evidence of hepatic steatosis (>5%) with the presence of at least one out of five cardio-metabolic criteria: Overweight/obesity, presence of prediabetes/DM, increased triglycerides, decreased high-density lipoprotein (HDL), or hypertension. Furthermore, the diagnosis of MASLD also requires exclusion of other secondary causes for steatotic liver disease. Metabolically, prolonged cumulative intake of a diet more than 10% of the body's energy requirement disrupts metabolic homeostasis and results in a state of IR. This results in lipolysis with increased free fatty acids (FFAs) in the blood being delivered to the liver with resultant hepatic de novo lipogenesis. This primary steatosis and lipotoxicity in the hepatocytes culminates in increased endoplasmic reticulum stress, mitochondrial dysfunction, oxidative stress, and immune-cell infiltration.¹⁶ Insulin resistance plays a central role in MASLD pathogenesis, creating a vicious cycle where hepatic steatosis exacerbates IR, which in turn promotes further lipid accumulation.¹⁷

Histologically, disease progression is characterized by hepatocellular damage and lobular inflammation. Severity of liver fibrosis is measured on a histological scale, namely F0 (no fibrosis), F1 (perisinusoidal or portal fibrosis), F2 (perisinusoidal and portal or periportal fibrosis), F3 (septal and bridging fibrosis), and F4 (cirrhosis). Clinically significant liver fibrosis (≥F2) is a powerful predictor of all-cause mortality and hepatic complications. Notably, regression of hepatic fibrosis has been associated with a better prognosis.¹⁸

Current Therapeutic and Regulatory Landscape

Despite the recent United States Food and Drug Administration (USFDA) approval of resmetirom for MASH, it is unavailable in most countries. Notably, where available, resmetirom may not be affordable and imposes a significant economic burden.^6,7,16 Moreover, resmetirom has no effect on multiple metabolic syndrome factors, which affect mortality in MASLD.^19–24

Recently, semaglutide demonstrated evidence of histologic benefits in MASH with moderate or advanced hepatic fibrosis in the ESSENCE Trial.²⁵ This resulted in the USFDA approving semaglutide too, for the management of MASH.²⁶ Limitations to semaglutide use include limited accessibility and the high cost of therapy.^27,28 Gastrointestinal side effects are common with semaglutide, occurring in up to 73% of patients, with it being a leading cause for treatment discontinuation.²⁹ In addition, adverse effects such as the development of gallstones and pancreatitis remain a concern.^30,31

Pioglitazone 30 mg daily given over 96 weeks demonstrated a reduction in hepatic steatosis and lobular inflammation in patients with nonalcoholic steatohepatitis (NASH). However, no improvement in fibrosis was noted, and patients experienced significant weight gain.³² Use of pioglitazone 30 mg is often precluded by tolerability concerns in real-world settings.^33,34 In the absence of affordable and readily available therapies, off-label use of metformin, ursodeoxycholic acid, pioglitazone, and Vitamin E has been in use with some levels of associated benefits.^16,35

Regulatory approvals for drugs in MASH are stringent and require targeting patients with an NAFLD activity score (NAS) of at least four with fibrosis stage 2 or 3. In these patients, two surrogate histological endpoints need to be studied, namely “resolution of MASH without worsening fibrosis” and/or “improvement in fibrosis without worsening of MASH.” While the USFDA requires satisfaction of any one of the above histological outcomes, the European Medicines Agency (EMA) requires that both endpoints be met for regulatory approval.³⁶

SGLT2i – Key Mechanisms and Metabolic Effects in MASLD

Sodium–glucose transporter 2 inhibitor blocks the sodium–glucose transporter 2 situated in the proximal tubule of the nephron, which is responsible for reabsorption of approximately 80–90% of the filtered glucose load.³⁷ This leads to significant glucosuria (typically 60–80 gm of glucose per day), thereby leading to a caloric deficit of approximately 240–320 calories per day.³⁸ This results in a weight loss of 0.9–1.3 kg per month, predominantly driven by fat loss in both visceral and subcutaneous adipose tissues.^30,38–40 The reduction in blood glucose, visceral fat, and IR leads to enhanced hepatic fatty acid oxidation and reduced hepatic de novo lipogenesis. Sodium–glucose transporter 2 inhibitors induce a metabolic shift that prioritizes fatty acid oxidation over glucose utilization, which results in increased ketogenesis and more oxygen-efficient energy production. Improved insulin sensitivity reduces hepatic glucose production and de novo lipogenesis, while enhanced peripheral glucose uptake reduces glucose availability for hepatic fat synthesis.⁴¹ This effect assumes importance in the context of MASLD, where impaired fatty acid oxidation contributes to hepatic fat accumulation.¹⁶ Preclinical studies have demonstrated that SGLT2i, like dapagliflozin, reduces hepatic fat accumulation, oxidative stress, inflammatory response, and liver fibrosis.^42,43

Clinical Evidence for Dapagliflozin in MASLD

Clinical evidence with dapagliflozin evolved in the course of time in lieu with advances in diagnostics and therapeutic perspectives of MASLD. Earlier evidence from individual clinical studies showed improvement in hepatocellular enzymes, metabolically related biomarkers, body weight, and imaging parameters.^44–58 Recently, interest has been ignited with biopsy-proven evidence of dapagliflozin's benefits in MASH.^14,15 The gamut of clinical evidence exploring the benefits of dapagliflozin in MASLD has been summarized in Table 1 (Clinical Studies) and Table 2 (Meta-analysis), respectively. In the evidence summary below, we have also used older definitions of the MASLD spectrum (namely NAFLD and NASH) as per the defined study population/inclusion criteria in each of those studies.

Table 1.

Summary of clinical evidence

Author, year	Patients (n) and stratification	Study design, duration, and endpoint(s)	Results
Lin et al., 202514 DEAN trial	Biopsy diagnosed MASH (with or without T2DM) Age: >18 yrs Total: n = 154 Dapa 10 mg O.D. (n = 78) Placebo (n = 76)	Multicenter, double-blind, randomized, placebo-controlled trial (phase III) Duration: 48 wks Primary outcome: MASH improvement (defined as a decrease in at least two points in NAS or a NAS of ≤3 points) without worsening of liver fibrosis (defined as without an increase in fibrosis stage) Confirmatory secondary outcomes: MASH resolution without worsening of fibrosis. fibrosis improvement without worsening of MASH.	MASH improvement without worsening of fibrosis was: – Dapa grp = 53% (n = 41) participants – Placebo grp = 30% (n = 23) participants – RR = 1.73 (95% CI = 1.16–2.58) (p = 0.006) Mean diff of NAS (intergrp) was: – −1.39 (95% CI = −1.99 to −0.79) (p < 0.001) Individual components of the NAS score also showed reduction (improvement) in the Dapa grp compared to the placebo. Mean diff was: – For steatosis = −0.45 (95% CI = −0.71 to −0.19) (p < 0.001) – For ballooning = −0.43 (95% CI = −0.63 to −0.24) (p < 0.001) – For lobular inflammation = −0.49 (95% CI = −0.74 to −0.23) (p < 0.001) MASH resolution without worsening of fibrosis occurred in – Dapa grp = 23% (n = 18) participants – Placebo grp = 8% (n = 6) participants – RR = 2.91 (95% CI = 1.22–6.97) (p = 0.01) Fibrosis improvement without worsening of MASH occurred in – Dapa grp = 45% (n = 35) – Placebo grp = 20% (n = 15) – RR = 2.25 (95% CI = 1.35 to 3.75) (p = 0.001) Non-invasive parameters (assessed by TE) showed improvement in Dapa vs placebo – Liver steatosis = −15.71 dB/m (95% CI = −27.49 to −3.94) (p = 0.009) – Liver stiffness = −1.51 kPa (95% CI = −2.15 to −0.88) (p < 0.001) Anthropometric measures: – Body weight reduced in Dapa grp by −4.26 kg and in placebo grp by −0.75 kg (p < 0.001) – Significant reduction was also noted in BMI, WC, and abdominal visceral fat area (p < 0.001 for all) in the Dapa grp compared to the placebo. AE-related discontinuations – Dapa grp = 1% (n = 1) – Placebo grp = 3% (n = 2)
Weng et al., 202544	NAFLD and CAP score: ≥252 dB/m. Age: 20–74 yrs Total: n = 150 Dapa 10 mg/day grp (n = 75) Control grp (n = 75)	Randomized, open-label, two-arm, parallel-grp trial Duration: 24 wks Primary outcome: Δ CAP scores at 24 wks	Δ CAP score at 24 wks (mean) Dapa grp: −46.4 dB/m (↓14.8%) (p < 0.001) Control grp: −18.1 dB/m (↓5.7%) (p = 0.029) Intergrp diff (p = 0.002) Δ Fatty liver grade at 24 wks Dapa grp: ↓34.6% (from 2.6 ± 0.5 to 1.7 ± 0.7) (p < 0.001) Control grp: ↓15.3% (from 2.6 ± 0.5 to 2.2 ± 0.8) (p < 0.001) Intergrp diff (p < 0.001) LSM, WC, and ALT decreased in both Dapa and control grps but intergrp diff was non-significant.
Monem et al., 202515	Biopsy-proven NASH Age: 18–65 yrs Total: n = 100 DM on Pio 30 mg O.D. (PD grp) (n = 25) DM on Dapa 10 mg O.D. (DD grp) (n = 25) Non-DM on Pio 30 mg O.D. (PND grp) (n = 25) Non-DM on Dapa 10 mg O.D. (DND grp) (n = 25)	Four-grp, prospective, randomized, parallel, open-label study Duration: 24 wks Primary outcomes: Δ NAS score (and individual histological features of steatosis, hepatocellular ballooning, and lobular inflammation)	NAS and hepatic steatosis (by biopsy) significantly improved within DND, DD, and PD grps (p < 0.05) Δ NAS scores and the change in most individual histological parameters were comparable between Dapa and Pio in both DM and non-DM patients (p > 0.05) except for hepatocellular ballooning. Δ Hepatocellular ballooning: Dapa showed superiority vs Pio in DM patients (DD vs PD grp (p = 0.048)) Δ Liver fibrosis grade (assessed by TE) showed: In DM patients – Comparable improvement with Dapa vs Pio [PD vs DD grp (p = 0.287)] In non-DM patients – Superiority for Dapa vs Pio (DND vs PND grp (p = 0.018)) Δ LSM (in kPa assessed by TE) in non-DM patients showed: better improvement with Dapa vs Pio (DND vs PND grp (p = 0.021)) In a subgrp of DD with baseline LSM > 8 kPa: Dapa reduced LSM by 25.6% (p = 0.024) (from 15.58 ± 8.05 kPa at baseline to 11.59 ± 5.82 kPa after 24 wks) Δ NFS score: Improved in DND grp (p = 0.034) Negative changes in PD grp (p = 0.016) Δ FIB-4 score: Improvement in both DD (p = 0.037) and DND (p = 0.035) grps Negative changes in PD grp (p = 0.015) Effect on liver enzyme and metabolic parameters compared to baseline: In DD grp: Significant improvement in ALT, AST, ALP, GGT, TGs, and HbA1c (p < 0.05) In DND grp: Significant improvement in AST, GGT, TC, TGs, and HOMA-IR (p < 0.05) In PND grp: Significant improvement in AST, TGs, and HDL (p < 0.05) Dapa showed significant superiority in: all anthropometric measures (p < 0.001) (both DM and non-DM patients) QoL (p < 0.05) (both DM and non-DM patients)
Fukada et al., 202554	T2DM with MASLD Age: 20–80 yrs Total: n = 24 Dapa 5 mg/day grp (n = 13) Vit. E 150 mg/day grp (n = 11)	Prospective RCT Duration: 24 wks Primary outcomes: FIB-4 Index Liver enzymes (AST, ALT, ALP, and GGT) Type IV collagen levels	No changes in FIB-4 index, type IV collagen levels or ALP levels in both grps. Both grps had similar reduction in AST, ALT. GGT only decreased in Dapa grp. No significant intergrp diff in primary endpoints.
Shi et al., 202349	NAFLD with T2DM (HbA1c uncontrolled on Met 1.5 gm/day) Age: 18–75 yrs Total: n = 84 Dapa 10 mg/day (n = 42) Control grp (any other AHA than SGLT2i) (n = 42)	Prospective, open-label, randomized, controlled trial Duration: 24 wks Primary outcomes: Δ LFC (%) (assessed by MRI-PDFF) Δ PFC (%) (assessed by 3T-MRI)	Δ LFC: In Dapa grp. LFC (%) reduced by −4.18% (95% CI = −7.47 to −2.05) (p < 0.001) Change in Control grp (ns) Intergrp diff: (p < 0.001) % Patients with LFC <5%: 25% in Dapa grp 0% (no patients) in control grp Δ PFC: ↓ in Dapa grp. PFC (%) reduced by −1.16 (95% CI = −1.93 to −0.37) (p < 0.001) Change in control grp (ns) Intergrp diff: (p = 0.033) Δ FIB-4 Index: improved significantly in the Dapa grp (p = 0.001) no improvement in the control grp Intergrp diff barely missed statistical significance (p = 0.058) At 24 wks, there was a significant intergrp diff in the below (in favor of the Dapa grp): BW (p < 0.001) BMI (p = 0.001) Serum TNF-α (p = 0.004) IL-6 (p = 0.001) ALT (p < 0.001) Uric acid (p < 0.001)
Phrueksotsai et al., 202153	T2DM with NAFLD Age: 18–75 yrs Total: n = 40 Dapa 10 mg/day grp (n = 20) Placebo grp (n = 20)	Prospective, double-blinded, placebo-controlled, randomized, single-center study Duration: 12 wks Primary endpoint: Δ intrahepatic lipid content evaluated by LAI (observed in noncontrast CT scan)	At 12 wks, Dapa grp showed an increase in LAI (p = 0.006) compared to placebo (signifying a significant reduction in intrahepatic lipid content) Δ LAI: Dapa grp: ↑ 5.8 ± 5.1 HU Placebo grp: ↑ 0.5 ± 6.1 HU Significant reduction in Dapa grp vs placebo grp was also observed in Δ BW: −3 vs −0.17% (p < 0.001) Δ body fat: −1 vs −0.2% (p = 0.008) Δ V/S ratio: −0.1 vs 0.1 (p = 0.003) Δ ALT: −4.5 U/L vs −1 U/L (p < 0.024) Δ HbA1c: −1.3 vs −0.2 (p = 0.002)
Hussain et al., 202155	T2DM with NAFLD with BMI > 28 Mean age: 29 ± 16 yrs in the Dapa grp and 31 ± 14 yrs in the placebo grp Total: n = 146 Dapa 5–10 mg/day (n = 67) Placebo (n = 71)	double-blind randomized placebo-controlled trial Duration: 12 wks Relevant outcomes: BW, BMI, AST, and ALT	Δ BW: Dapa grp: ↓6.6% (from 90 ± 13.5 to 84 ± 11.6 kg) (p = 0.002) Placebo grp: ↑0.5% (from 85 ± 17.8 to 85.5 ± 13.7 kg) Intergrp diff vs placebo (p = 0.005) Δ BMI: Dapa grp: ↓10.1% (from 29.5 ± 2.5 to 26.5 ± 3.5 kg/m2) Placebo grp: ↓6.3% (from 31.5 ± 3.0 to 29.5 ± 4.2 kg/m2) Intergrp diff vs placebo (p = 0.002) Δ ALT: Dapa grp: −17 ± 14.2 Placebo grp: +4 ± 17.4 Intergrp diff vs placebo (p = 0.001) Δ AST: Dapa grp: −27 ± 5.4 Placebo grp: −6 ± 16.5 Intergrp diff vs placebo (p = 0.002)
Cho et al., 202159	T2DM with NAFLD on Pio for >12 wks. Age: 20–80 yrs Total: n = 53 Dapa grp (n = 27) Pio grp (n = 26)	Multicenter, open-label, prospective, randomized, parallel-grp comparative trial Duration: 24 wks Outcomes: FLI, FIB-4 Index	Δ FLI: Dapa grp: ↓16.2% (from 58.3 ± 18.3 to 48.8 ± 19.5) Pio grp: ↑4.2% (from 49.0 ± 26.1 to 51.1 ± 25.8) Intergrp diff (p < 0.01) Δ FIB-4: Dapa grp: ↓12.4% (from1.37 ± 0.59 to 1.20 ± 0.50) Pio grp: ↑2.2% (from1.32 ± 0.5 to 1.35 ± 0.52) Intergrp diff (p < 0.01)
Das et al., 202145	T2DM with NAFLD Mean age: 44.11 ± 8.24 yrs Total: n = 100 (all received Dapa 10 mg/day)	Single-center, prospective, open-label, uncontrolled, interventional, cohort study Duration: 6 months Outcomes of interest: Hepatic fibrosis score (measured by TE), Hepatic steatosis (by abdominal USG), non-invasive scores, NFSs, FIB-4 index, and APRI score.	TE measured hepatic fibrosis score: ↓13.6% (6.95 ± 1.42 kPa at baseline to 6 ± 1.44 kPa after 6 months) (p = 0.001) Hepatic steatosis improved (p < 0.05) Non-invasive scores like NFS, FIB-4 score, and APRI scores did not show any improvement (p > 0.05)
Kinoshita et al., 202052	T2DM with NAFLD Age: >20 yrs Total: n = 98 Dapa 5 mg/day grp (n = 32) Pio 7.5 to 15 mg/day grp (n = 33) Glim 0.5 to 1 mg/day grp (n = 33)	Prospective, randomized, open-label, three-arm, active control study Duration: 28 wks Primary outcome: L/S ratio (measured by abdominal CT)	L/S ratio Dapa and Pio (but not Glim) significantly increased the L/S ratio – Δ L/S ratio with Dapa = ↑0.17 ± 0.04 (p < 0.05) – Δ L/S ratio with Pio = ↑0.22 ± 0.04 (p < 0.05) – Δ L/S ratio with Glim = ↑0.03 ± 0.04 Effects of Dapa and Pio on L/S ratio were comparable. % patients whose L/S ratio improved after treatment was – 84.4% for Dapa – 81.8% for Pio – 48.5% for Glim Serum type IV collagen 7S level tended to decrease after treatment with Pio or Dapa. Patients in the Dapa and Pio arms also experienced improvements in AST, ALT, GGT, and adiponectin. VFA decreased in Dapa arm by −19.4 ± 4.1 cm2 (p < 0.05).
Gastaldelli et al., 202056 DURATION-8 post hoc analysis	Uncontrolled T2DM with HbA1c 8–10% despite 1.5 gm met monotherapy Total: n = 695 Exe 2 mg once weekly grp (n = 227) Dapa 10 mg O.D. grp (n = 230) Exe 2 mg once weekly + Dapa 10 mg O.D. grp (n = 228)	Post hoc analysis of multicenter, double-blind, randomized, active-controlled, phase III trial Duration: 104 wks Outcomes: At 28 and 52 wks Non-invasive markers of hepatic steatosis (FLI and NLFS), fibrosis (FIB-4), severe fibrosis (NFS) Liver enzymes IR	At wk 28, biomarkers of steatosis and fibrosis were reduced from baseline in all treatment grps. For a decrease in FLI at 28 wks: Exe + Dapa grp effects were stronger vs Exe grp. −2.92, (95% CI = −5.11 to −0.73) (p = 0.0092) Exe + Dapa grp effects were stronger vs Dapa grp −2.77 (95% CI = −4.93 to −0.62) (p = 0.0119) For the decrease in FLI at 52 wks: Exe + Dapa grp effects were stronger vs Exe grp −3.23 (95% CI = −5.79 to −0.68) (p = 0.0134). FIB-4 showed a reduction vs baseline only in the Exe + Dapa grp at both: 28 wks: −0.06 (95% CI = −0.11 to −0.01) (p = 0.0135) 52 wks: −0.05 (95% CI = −0.09 to −0.004) (p = 0.0308). ALT was reduced significantly in both Exe + Dapa grp and Dapa grp at both 28 wks and 52 wks
Ribeiro Dos Santos and Baer Filho, 202057	NAFLD Age: 21–74 yrs Total: n = 14 All patients used Dapa 10 mg/day for an avg of 75 days (60–90 days).	Retrospective, observational study Duration: Avg 75 days (60–90 days) Outcomes: ALT, AST, GGT, HOMA-IR, BW	ALT, AST, GGT, HOMA-IR, and BW were significantly lower after Dapa treatment Δ ALT = ↓41% (from 62.29 to 36.71 U/L) (p < 0.0001) Δ AST = ↓33.4% (from 37 to 24.64 U/L) (p = 0.0006) Δ GGT = ↓34.7% (from 68.5 to 44.71 U/L) (p = 0.0009) Δ HOMA-IR = ↓37.2% (from 5.21 to 3.27 U/L)↓ (p = 0.0005) Δ BW = ↓5.3% (from 98.6 to 93.3 kg) (p = 0.0001)
Aso et al., 201946	T2DM with NAFLD Total: n = 57 Dapa 5 mg/day grp (n = 33) Control grp (n = 24)	Randomized, active-controlled, open-label trial Duration: 24 wks Outcomes: CAP, LSM	At wk 24, Dapa grp showed: ↓CAP by 7.6% (from 314 ± 61 dB/m at baseline to 290 ± 73 dB/m) (p = 0.042) ↓LSM by 15.5% (from 9.49 ± 6.05 kPa at baseline to 8.01 ± 5.78 kPa) ↓ Visceral Fat mass (significant) In subgrp with LSM ≥8.0 kPa on Dapa: ↓LSM by 25.1% (from 14.7 ± 5.7 kPa at baseline to 11.0 ± 7.3 kPa) (p = 0.0158) Dapa grp (but not control grp) had ↓ALT and ↓GGT.
Kato et al., 201947	T2DM with NAFLD Total: n = 57 Dapa 5 mg/day (n = 33) Placebo (n = 24)	Randomized, placebo-controlled trial Duration: 24 wks Outcomes of interest: VAT – using dual bioelectrical impedance analysis CAP – using TE	At 24 wks, Dapa grp showed Significant ↓ VAT Significant ↓ AST, ↓ ALT, ↓ GGT Significant ↓ CAP Although both grps showed significant reduction in serum sDPP4 at 24 wks, magnitude of reduction was greater in Dapa grp. Changes in liver enzymes after Dapa treatment correlated positively with those in serum sDPP4 levels.
Morino et al., 201951 SUMS-ADDIT-2 Study	Overweight (BMI > 23), uncontrolled T2DM treated with OADs Age: (not mentioned) Total: n = 52 Dapa grp Control grp	Randomized controlled, open-label trial Duration: 24 wks Primary outcome: Δ BW Secondary outcome: IHTG (measured by MRS)	Δ BW: Mean diff between Dapa vs Control grps was −1.72 kg (p = 0.004) with greater reduction in Dapa grp Δ IHTG: ↓ Dapa grp by 18% (p = 0.033) ↑ Control grp by 13%
Shimizu et al., 201848	T2DM with NAFLD Age: >20 yrs Total: n = 57 Dapa 5 mg/day grp (n = 33) Control grp (n = 24)	Randomized, active-controlled, open-label trial Duration: 24 wks Outcomes: CAP (measure of hepatic steatosis) LSM (measure of hepatic fibrosis) Both outcomes were measured by TE	Δ CAP: Dapa grp: ↓7.6% (from 314 ± 61 to 290 ± 73 dB/m) (p = 0.04) Control grp: (ns) Δ LSM: Dapa grp: ↓15.5% (from 9.49 ± 6.05 kPa to 8.01 ± 5.78 kPa) (p = 0.22) In the subgrp with baseline LSM > 8 kPa, LSM decreased significantly in Dapa grp by 25.7% (from 14.7 ± 5.7 kPa to 11.0 ± 7.3 kPa) (p = 0.0158)
Eriksson et al., 201850 EFFECT-II Study	T2DM with NAFLD Mean age: 65.5 yrs Total: n = 84 Dapa 10 mg/day (n = 21) Om-3CA 4 gm/day (n = 20) Combination of both (n = 22) Placebo (n = 21)	Multicenter, randomized, placebo-controlled, double-blind, parallel-grp study Duration: 12 wks Primary outcome: LFC (assessed by MRI-PDFF)	Δ LFC: Om-3CA grp: −15% (ns) Dapa grp: −13% (ns) Combination grp: −21% (p = 0.046) Dapa monotherapy (but not combination therapy) reduced ALT, AST, GGT, CK 18-M30, CK 18-M65, and FGF-21 levels
Tobita et al., 201758	T2DM with biopsy-confirmed NASH Age: 46–78 yrs Total n = 16 (results evaluated in 11 patients) All patients received Dapa 5 mg/day	Single-arm, prospective, non-randomized, open-label, pilot study Duration: 24 wks Outcomes of interest: BW, BMI, WC, WHR, body fat mass, Body fat percentage, AST, ALT, GGT, and adiponectin	After 24 wks of Dapa treatment: Δ BW: ↓4.7% (−3.8 kg) (p < 0.01) Δ BMI: ↓11.9% (from 31 to 27.3 kg/m2) (p < 0.01) Δ WC: ↓7.1% (from 101.4 to 94.2 cm) (p < 0.01) Δ WHR: ↓2% (from 1.02 to 1.00) (p < 0.01) Δ Body fat mass: ↓21.5% (from 28.3 to 22.2 kg) (p < 0.01). Notably, there were no changes in lean body mass Δ Body fat percentage: ↓ from 42.4 to 38.2% (p < 0.01) Δ AST: ↓50% (from 52 to 26 U/L) (p < 0.01) Δ ALT: ↓49% (from 59 to 30 U/L) (p < 0.01) Δ GGT: ↓48.4% (from 64 to 33 U/L) (p < 0.01) Δ Adiponectin: ↑29.6% (from 5.4 to 7.00 µg/mL) (p < 0.01)

*Statistically significant; Δ, change from baseline; 3T-MRI, three tesla magnetic resonance imaging; AHA, antihyperglycemic agent; ALP, alkaline phosphatase; ALT, alanine transaminase; APRI, AST to Platelet Ratio Index; ASH, metabolic dysfunction–associated steatohepatitis; AST, aspartate transaminase; avg, average; BMI, body mass index; BW, body weight; CAP, continuous attenuation parameter; CK, cytokeratin; CT, computerized tomography; Dapa, dapagliflozin; diff, difference; DM, diabetes mellitus; FGF-21, fibroblast-growth factor 21; FIB-4, fibrosis index based on four factors; FLI, fatty liver index; GGT, gamma-glutamyl transpeptidase; Glim, glimepiride; grp, group; HbA1c, glycosylated hemoglobin; HDL, high-density lipoprotein; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; HU, Hounsfield unit; IHTG, intrahepatic triglyceride; IL-6, interleukin 6; Intergrp, intergroup; IR, insulin resistance; kg, kilogram; kPa, kilo-pascals; L/S ratio, liver spleen ratio; LAI, liver attenuation index; LFC, liver fat content; LSM, liver-stiffness measurement; MASLD, metabolic dysfunction–associated steatotic liver disease; Met, metformin; MRI-PDFF, magnetic resonance imaging estimated proton density fat fraction; MRS, magnetic resonance spectroscopy; NAFLD, nonalcoholic fatty liver disease; NAS, NAFLD activity score; NASH, nonalcoholic steatohepatitis; NFS, NAFLD fibrosis score; NLFS, NAFLD liver fat score; ns, non-significant; O.D., once daily; OAD, oral antidiabetic drug; Om-3CA, omega-3 carboxylic acids; PFC, pancreatic fat content; Pio, pioglitazone; RCT, randomized controlled trial; RR, relative risk; sDPP4, soluble dipeptidyl peptidase four; subgrp, subgroup; T2DM, type 2 diabetes mellitus; TC, total cholesterol; TE, transient elastography; Tene, teneligliptin; TG, triglyceride; TNF-α, tumor necrosis factor α; V/S ratio, visceral fat/subcutaneous fat ratio; VAT, visceral adipose tissue; VFA, visceral fat area; vs, versus; WC, waist circumference; wks, weeks; WHR, waist-to-hip ratio; yrs, years

Table 2.

Meta-analysis evidence

Authors, year	Number of studies and participants	Key findings
Gomez et al., 202660	14 RCTs n = 855	Compared with SOC, SGLT2 is – ↓CAP (mean diff = 10.48 dB/m) – ↓LSM (mean diff = 0.57 kPa) Dapa showed the most consistently favorable estimates compared to other SGLT2i across both outcomes.
Duan and Chen, 202461	5 RCTs n = 395 patients	Compared with the comparator grp, Dapa reduced (expressed as SMD): ALT: −1.10 (p < 0.00001) AST: −1.32 (p < 0.00001)
Hu et al., 202362	4 RCTs n = 275 patients	Compared with the comparator grp, Dapa reduced (expressed as SMD): ALT = −1.27 (p < 0.00001) AST = −1.37 (p = 0.0002) HOMA-IR = −0.36 (p = 0.16)
Sun et al., 202263	7 RCTs n = 390 patients	Compared with the comparator grp, Dapa reduced (expressed as WMD): ALT: −6.62U/L (p = 0.03) AST: −4.20 U/L (p = 0.03) HOMA-IR: −0.88 (p = 0.002) BW: −3.79 kg (p < 0.00001) BMI: −1.33 kg/m2 (p = 0.01) LDLc: −2.66 mg/dL (p < 0.00001) TG: −16.77 mg/dL (p = 0.03)
He et al., 202264	11 studies n = 839 patients	Compared with the comparator grp, Dapa reduced (expressed as MD): ALT: −5.58 U/L (p = 0.001) AST: −6.17 U/L (p = 0.009) GGT: −4.87 U/L (p < 0.00001) TG: −0.16 mmol/L (p = 0.05) BW: −3.6 kg (p = 0.0003) BMI: −1.2 kg/m2 (p < 0.00001)

ALT, alanine transaminase; AST, aspartate transaminase; BMI, body mass index; BW, body weight; CAP, controlled attenuation parameter; Dapa, dapagliflozin; Diff, difference; GGT, gamma-glutamyl transpeptidase; Grp, group; HOMA-IR, Homeostatic Model Assessment of Insulin Resistance; LDLc, low-density lipoprotein cholesterol; LSM, liver-stiffness measurement; MD, mean difference; RCT, randomized controlled trials; SGLT2i, sodium–glucose transporter 2 inhibitors; SMD, standard mean difference; SOC, standard of care; TG, triglycerides; WMD, weighted mean difference

Histological Evidence

Until recently, there was no biopsy-proven evidence with SGLT2i for key histological outcomes of MASH. In the recently published DEAN trial, 154 adult patients with biopsy-proven MASH (irrespective of presence or absence of T2DM) were randomized to receive either dapagliflozin 10 mg daily or placebo for 48 weeks. The primary endpoint of this phase III trial was MASH improvement without worsening of liver fibrosis. Histologically, “MASH improvement” was defined as a decrease in at least two points in NAS or NAS of ≤3 points. “Worsening of liver fibrosis” was defined by no increase in fibrosis stage.¹⁴ Confirmatory secondary endpoints were “MASH resolution without worsening of fibrosis” and “fibrosis improvement without worsening of MASH,” which were in alignment with regulatory guidance for drug development.^14,36 The primary endpoint was achieved in 53% of patients on dapagliflozin compared with 30% in the placebo group (p = 0.006). The mean difference in NAS was −1.39 in favor of the dapagliflozin group (p < 0.001), with statistically significant improvement in individual components too, namely steatosis, lobular inflammation, and hepatocyte ballooning (p < 0.001 for all). Metabolic dysfunction–associated steatohepatitis resolution without worsening of fibrosis occurred 23% of patients in the dapagliflozin group compared to 8% with placebo (p = 0.01). Similarly, fibrosis improvement without worsening of MASH also occurred in 45% of patients in the dapagliflozin group compared to 20% with placebo (p = 0.001).¹⁴ The unequivocal benefit noted in key secondary endpoints of this trial opens new vistas for the therapeutic potential and further evidence generation for dapagliflozin (and potentially other SGLT2i) for managing patients with MASH. Recently, Monem et al. studied the comparative effects of dapagliflozin and pioglitazone in NASH patients via a four-group open-label study. A total of 100 patients with biopsy-proven NASH (n = 50 with diabetes and n = 50 without diabetes) were randomized 1:1:1:1 (25 patients in each group) to receive either dapagliflozin 10 mg o.d. or pioglitazone 30 mg o.d. for 24 weeks. Though the change in NAS scores (primary endpoint) was comparable across groups who received either agent, dapagliflozin demonstrated borderline superiority for improvement in hepatocellular ballooning in patients with diabetes.¹⁵

Imaging Parameters Evidence

Imaging parameters for MASLD have moved beyond the earlier era with an ultrasonography (USG) diagnosis of a “fatty liver,” i.e., hepatic steatosis. Current imaging modalities like magnetic resonance imaging–proton density fat fraction (MRI-PDFF) and transient elastography (TE) have changed the landscape of diagnosis of MASLD. Studies that explored the therapeutic effect of dapagliflozin in MASLD have largely utilized TE, MRI, computerized tomography (CT), and USG.

By TE

Transient elastography has emerged as an important non-invasive modality in the diagnosis and management of MASLD. For assessment of liver steatosis, the controlled attenuation parameter (CAP) provides a semi-quantitative assessment.⁶⁵ Controlled attenuation parameter is expressed in decibels per meter (dB/m) with values ranging between 100 and 400 dB/m (normal values typically being under 247 dB/m).⁶⁶ Consequently, a reduction in CAP values from a higher baseline value implies a reduction in hepatic steatosis. Liver stiffness increases with fibrosis severity, with TE being the most common method to assess it. Consequently, it may be used to exclude significant hepatic fibrosis.⁶⁵ The stiffer the liver tissue is, the higher the values of liver-stiffness measurement (LSM) are. Lower LSM values indicate a more elastic liver, with the healthy population having an LSM of 4.5–5.5 kPa.⁶⁶ A TE-derived LSM < 8 kPa can be used to rule out advanced fibrosis.⁶⁵ In the DEAN trial described earlier, other than biopsy parameters, TE also demonstrated that 48-week treatment with dapagliflozin 10 mg in MASH patients reduced both liver steatosis (−15.71 dB/m) and liver stiffness (−1.51 kPa) significantly compared to placebo.¹⁴ Weng et al. conducted a randomized, open-label, two-arm, parallel-group trial in NAFLD patients with a CAP score > 252 dB/m. A total of 150 patients between 20 and 74 years of age were randomized 1:1 to receive either dapagliflozin 10 mg daily or placebo for 24 weeks. Baseline CAP score was 319.9 ± 45.5 dB/m in the control group and 312 ± 40.3 dB/m in the dapagliflozin group. Controlled attenuation parameter-derived fatty liver grade at baseline was 2.6 ± 0.5. At 24 weeks, the mean change in CAP score from baseline (primary outcome) was greater in the dapagliflozin group (−46.4 dB/m) than in the control group (−18.1 dB/m) (intergroup p = 0.002). Similarly, fatty liver grade was also reduced in the dapagliflozin group by 34.6% (−1.7 ± 0.7), while in the control group, it was reduced by 15.3% (intergroup p < 0.001).⁴⁴ In the earlier described study by Monem et al., though the primary endpoint was histological, the change in liver fibrosis grade and LSM by TE were additionally studied. In patients with diabetes, comparable improvement in liver fibrosis grade was noted with dapagliflozin and pioglitazone. However, in patients without diabetes, dapagliflozin significantly reduced the liver fibrosis grade compared to pioglitazone (p = 0.018). Evidently, for change in LSM, patients without diabetes showed significantly better improvement with dapagliflozin compared to pioglitazone (p = 0.021). In a subgroup of diabetic patients on dapagliflozin with baseline LSM > 8 kPa, the LSM reduced significantly by 25.6%.¹⁵

Das et al. conducted a prospective, open-label, single-arm, interventional, cohort study in 100 patients with T2DM and NAFLD. All patients received dapagliflozin 10 mg/day for 6 months. At the end of the study, TE measured hepatic fibrosis score reduced significantly by 13.6% (from 6.95 ± 1.42 kPa at baseline to 6 ± 1.44 kPa at the end of the study). Hepatic steatosis by USG also reduced significantly (p < 0.05).⁴⁵ Aso et al. conducted a randomized, active-controlled, open-label trial, whereby 57 patients with T2DM and NAFLD received either dapagliflozin 5 mg/day or a control group for 24 weeks. Key outcomes were the change in TE-derived CAP and LSM at 24 weeks. At 24 weeks, the dapagliflozin group showed a significant 7.6% reduction in CAP scores and 15.5% reduction in LSM. In a subgroup of patients on dapagliflozin with baseline LSM >8 kPa, a 25% reduction in LSM was noted from baseline (from 14.7 ± 5.7 to 11.0 ± 7.3 kPa).⁴⁶ In a similar study by Kato et al., this effect of dapagliflozin on the reduction in CAP was mirrored.⁴⁷ Shimizu et al. studied the effects of dapagliflozin in 57 adult patients with T2DM and NAFLD. Eligible participants were randomized to receive either dapagliflozin 5 mg/day or a control group. Transient elastography measured change in CAP and LSM at 24 weeks was the key outcome of interest. At the end of the study, CAP was significantly reduced by 7.6% (from 314 ± 61 at baseline to 290 ± 73 dB/m). In the control group, there was no significant reduction of CAP. Similarly, at 24 weeks, LSM was reduced by 15.5% in the dapagliflozin group (from 9.49 ± 6.05 to 8.01 ± 5.78 kPa). In a subgroup of patients taking dapagliflozin with baseline LSM > 8 kPa, there was a 25.7% reduction in LSM compared to baseline (from 14.7 ± 5.7 to 11.0 ± 7.3 kPa).⁴⁸ A recent meta-analysis of 14 randomized controlled trials (RCTs) demonstrated that, compared to other SGLT2i, dapagliflozin had the most consistently favorable estimate for reducing both CAP and LSM in MASLD patients.⁶⁰

By MRI and Complementary Techniques

Magnetic resonance imaging–PDFF is an accurate, reproducible, and precise MRI-based investigation for liver fat quantification. It is highly sensitive across the spectrum of steatosis and exhibits high accuracy to assess dynamic change. Liver fat content (LFC) > 5% by MRI-PDFF is diagnostic of hepatic steatosis.⁶⁵ Shi et al. conducted an open-label RCT of patients with NAFLD and T2DM uncontrolled on metformin 1.5 gm per day. A total of 84 adult patients between 18 and 75 years of age were randomized to either the study group (received dapagliflozin 10 mg daily for 24 weeks) or the control group (which received another antidiabetic agent other than SGLT2i for optimal glycemic control). The primary outcomes were changes in LFC assessed by MRI-PDFF and pancreatic fat content assessed by three-tesla MRI. At 24 weeks, the dapagliflozin group achieved a 4.18% reduction in LFC, with 25% of patients achieving LFC < 5%. Conversely, in the control group, the effect on LFC was non-significant, with none of the patients achieving LFC < 5%. Additionally, the dapagliflozin group also achieved a significant reduction in pancreatic fat content, which was not seen in the control group.⁴⁹ An earlier conducted study by Eriksson et al. also noted the disease-modifying effect of dapagliflozin in patients with T2DM and NAFLD, whereby, in addition to reducing biomarkers for hepatocyte injury and fibroblast-like-growth factor 21 (FGF-21), dapagliflozin also reduced LFC by 13%. A combination of dapagliflozin with omega-3 carboxylic acids resulted in a significant 21% reduction in LFC (p = 0.046).⁵⁰ In a study by Morino et al., they noted a 18% reduction in intrahepatic triglyceride content (measured by magnetic resonance spectroscopy) in overweight patients with T2DM who were administered dapagliflozin for 24 weeks.⁵¹ Recently, Naumova et al. also demonstrated that attenuation of hepatic steatosis by dapagliflozin was independent of weight reduction.⁶⁷

By Abdominal CT

In a noncontrast abdominal CT, a normal liver is denser and has a higher CT attenuation value [measured in Hounsfield units (HUs)] than the spleen. Therefore, a normal liver-to-spleen (L/S) attenuation ratio (also known as the L/S ratio) is greater than 1. However, when fat infiltrates the liver, its density on CT decreases, thereby causing a decrease in the L/S ratio. A decrease in the L/S ratio of less than one is indicative of steatotic liver disease.⁶⁸ Noncontrast abdominal CT can also be used to evaluate the intrahepatic lipid content by assessing the liver attenuation index (LAI). In essence, a higher positive LAI means a healthier liver, while a lower or negative LAI suggests fat accumulation in it. Liver attenuation index of above +5 indicates a normal liver, while an LAI of 0 to −5 is indicative of 10–30% hepatic steatosis.⁶⁹

Kinoshita et al. conducted a prospective, randomized, open-label, triple-arm, active-controlled study in 98 adults with T2DM and NAFLD to study the effects of three antihyperglycemic medications (AHAs) on the L/S ratio. In this study, patients were randomized 1:1:1 to receive either dapagliflozin, pioglitazone, or glimepiride, and L/S ratio was studied in each of the groups at baseline and at 28 weeks. Patients who received either dapagliflozin or pioglitazone (but not glimepiride) showed a significant increase in L/S ratio at 28 weeks compared to baseline (p < 0.05). The effect of the change in L/S ratio for both agents, dapagliflozin (+0.17) and pioglitazone (+0.22), was comparable. A total of 84.4% of patients on dapagliflozin and 81.8% patients on pioglitazone demonstrated improvement in their L/S ratios compared to only 48.5% of patients on glimepiride.⁵² Phrueksotsai et al. conducted a prospective, double-blind, placebo-controlled randomized study, whereby 40 adult patients with T2DM and NAFLD were randomized 1:1 to receive either dapagliflozin 10 mg per day or placebo for 12 weeks. At the end of 12 weeks, patients in the dapagliflozin group had a significantly higher LAI (5.8 ± 5.1 HU) compared to placebo (0.5 ± 6.1 HU), signifying a greater reduction in intrahepatic steatosis. Also, the visceral fat/subcutaneous fat ratio (V/S ratio) showed a significant reduction in the dapagliflozin arm (−0.1) compared with placebo (0.1).⁵³

Biomarkers, Body-composition, and Noninvasive Score Evidence

Aspartate aminotransferase (AST), alanine aminotransferase (ALT), and gamma-glutamyl transpeptidase (GGT) levels, which are markers of hepatocellular injury, typically show modest but significant reductions, reported across studies.^{15,44,46,47,49,50,52–58,61–64} Erikkson et al. in their study noted that T2DM patients with NAFLD on dapagliflozin showed a reduction in cytokeratin 18-M30, cytokeratin 18-M65, and fibroblast-growth factor 21, which are novel biomarkers for hepatocellular injury.⁵⁰ Along with a reduction in commonly measured anthropometric parameters (body weight, body mass index, waist circumference, and waist-hip ratio), studies have consistently noted that dapagliflozin also reduced visceral fat area, visceral adipose tissue, body fat percentage, and V/S ratio.^{14,15,44,46,47,49,52,53,55,57,58,63,64} Hence, weight reduction appears to preferentially affect visceral adipose tissue, as demonstrated by imaging studies showing greater reductions in visceral fat compared to subcutaneous fat.⁷⁰ Notably, based on the histological evidence studies, the hepatic benefits conferred by dapagliflozin clearly exceed a threshold that can be explained by weight loss alone.^14,15 Dapagliflozin's benefit in MASLD has also been studied for its effect on clinically validated non-invasive scores. The beneficial effect on fibrosis index based on four factors (FIB-4), NAFLD liver fat score (NLFS), NAFLD fibrosis score (NFS), and fatty liver index (FLI) is summarized in Table 1.^15,49,56,59

Other Newer Evidence with SGLT2i

A large nationwide cohort study from South Korea demonstrated that in patients with T2DM, the use of SGLT2i was associated with reduced risk of development of hepatocellular carcinoma.⁷¹ A recent cohort study of the 2,82,161 patients with T2DM from the Taiwan National Health Insurance database demonstrated that use of SGLT2is was associated with a 22% reduced risk of cirrhosis, a 21% reduced risk of decompensated cirrhosis, and a 48% reduced risk of all-cause mortality.⁷² Intriguingly, a recent retrospective cohort study involving 1,18,751 cirrhotic patients from the TriNetX platform (from 120 healthcare organizations) demonstrated that patients receiving SGLT2i had a 32% lower risk of serious hepatic events, 21% reduced risk of variceal bleeding, 33% reduced risk of all-cause hospitalizations, 46% reduced risk of paracentesis, and 53% reduced risk of hepatorenal syndrome compared to the control group.⁷³ A recent meta-analysis of eight cohort studies studied the effects of SGLT2i on hepatic events in patients with T2DM. During a median of 2.7 years, SGLT2i use was associated with 17% lower risk of major adverse liver outcomes (defined as a composite of hepatic decompensation events, hepatocellular carcinoma, liver transplantation, or liver-related deaths) and 36% lower risk of liver-related deaths. Notably, the significant reduction in major adverse liver outcomes was observed in comparison with dipeptidyl peptidase 4 inhibitors, metformin, or even pioglitazone.⁷⁴

Clinical Considerations

Patients with MASLD often have multiple cardiovascular risk factors, with MASLD itself being an independent risk factor for the same.^75,76 A total of 65% of T2DM patients have MASLD.³ SGLT2i, including dapagliflozin, have demonstrated unequivocal cardio-renal benefits which may confer additional therapeutic value in these patients beyond the hepatic effects.^77–80 In patients with T2DM and simple steatosis, early use of SGLT2i may prevent the development of hepatic fibrosis.⁸¹

It is also noteworthy that dapagliflozin has minimal clinically relevant drug interactions, which assumes importance in patients with multiple comorbidities.⁸² Patients with metabolic syndrome features, including central obesity and IR, may represent ideal candidates for dapagliflozin therapy, with its cardiovascular benefits providing value in this high-risk population.^83–85 Cost-effective generic options of SGLT2i, wherever available, assume further importance from a cost-benefit and patient compliance perspective, especially in resource-crunched settings.⁸⁶ It is also notable that SGLT2is are unlikely to cause hypoglycemia when given as monotherapy. However, the risk of hypoglycemia exists if used with concurrent medications like insulin or sulfonylureas.⁸⁷ Glucosuria due to SGLT2i can increase the risk of genitourinary tract infections. Patients should be counseled on adequate hydration and perineal hygiene to prevent the same.⁸⁸ In especially insulinopenic T2DM patients on SGLT2i, the risk of euglycemic diabetic ketoacidosis (eDKA) must be taken into clinical consideration.⁸⁹ Though extremely rare, eDKA due to SGLT2i may also occur in patients without diabetes.⁹⁰ The safety and efficacy of SGLT2is in severe hepatic impairment is less studied.^82,91 In cirrhotic patients, relatively lower effective circulating volume may exist despite overall fluid overload. Use of SGLT2i in these patients may be associated with further volume depletion risk.⁹¹

Research Considerations

Though approved indications for SGLT2i like dapagliflozin are from the perspective of T2DM, heart failure, and chronic kidney disease, newer evidence in MASLD underscores the therapeutic potential for further research.⁸² Though the positive histological benefits of the DEAN trial are endearing, there are scientific limitations. The primary endpoint of the trial was MASH improvement without worsening of fibrosis, instead of a composite of MASH resolution without worsening fibrosis and fibrosis improvement without worsening MASH. Positively, though, individual components of these were studied as secondary endpoints where dapagliflozin demonstrated unequivocal benefits. Another limitation of the trial was the recruitment of only Chinese patients, which limits its generalizability to the larger global population.¹⁴ A larger multi-national, multicentric phase III trial with primary and key secondary histological endpoints in alignment with the USFDA guidance will serve to close this evidence gap. Furthermore, trials must also explore the role of adding dapagliflozin to recent MASH-approved therapies, i.e., resmetirom and semaglutide, to note if key incremental benefits exist to combination therapy.

Conclusion

Dapagliflozin represents a promising potential therapeutic option for MASLD, providing not only hepatic fat reduction but also improvement in key histological parameters. The medication's unique mechanism of action, combining glucosuria-induced metabolic reprogramming with direct hepatic effects, addresses multiple aspects of MASLD pathophysiology. Current clinical evidence demonstrates consistent reductions in hepatic fat content, improvements in histological parameters, and liver enzymes with beneficial effects on anthropometric and metabolic parameters. Unequivocal benefits conferred to the heart and kidney, too, mandate clinical consideration. The therapeutic potential of dapagliflozin in MASLD appears substantial, particularly when integrated with lifestyle interventions and potentially as part of combination therapy regimens. Further, long-term, multinational, multicentric studies studying key histological outcomes as per regulatory guidance can cement the role of SGLT2i in MASLD. Future directions may also benefit from phenotyping patients to better channelize precision medicine, which could individualize and potentially improve patient care outcomes.

Ethical Approval

Not applicable since it is a literature review.

Data Availability Declaration

Not applicable since it is a literature review.

Artificial Intelligence (AI) Disclosure

No AI-assisted technologies were used.

Acknowledgments

No acknowledgements to declare.

Authors’ Contributions

Rajesh Upadhyay, Mangesh Tiwaskar, Bharat Saboo, and L Sreenivasamurthy contributed to conceptualization, literature review, formal analysis, writing original draft, review, editing, and supervision. Kirti Sonawale, Charmy Prajapati, and Parthasarathy Muralidharan contributed to the literature review, formal analysis, writing the original draft, review, and editing.

Orcid

Mangesh Tiwaskar https://orcid.org/0000-0003-4024-0095

Bharat Saboo https://orcid.org/0000-0002-7014-0143

L Sreenivasamurthy https://orcid.org/0000-0001-5571-7357

Kirti Sonawale https://orcid.org/0009-0008-5440-1711

Charmy Prajapati https://orcid.org/0009-0006-0193-3664

Parthasarathy Muralidharan https://orcid.org/0009-0002-2663-4645