Prolonged impacts of sodium glucose cotransporter-2 inhibitors on metabolic dysfunction-associated steatotic liver disease in type 2 diabetes: a retrospective analysis through magnetic resonance imaging

Abstract

The beneficial effects of sodium-glucose cotransporter 2 (SGLT2) inhibitors in people with type 2 diabetes (T2D) and metabolic dysfunction-associated steatotic liver disease (MASLD) have been suggested in several reports based on serological markers, imaging data, and histopathology associated with steatotic liver disease. However, evidence regarding their long-term effects is currently insufficient. In this retrospective observational study, 34 people with T2D and MASLD, treated with SGLT2 inhibitors, were examined by proton density fat fraction derived by magnetic resonance imaging (MRI-PDFF) and other clinical data before, one year after the treatment. Furthermore, 22 of 34 participants underwent MRI-PDFF five years after SGLT2 inhibitors were initiated. HbA1c decreased from 8.9 ± 1.8% to 7.8 ± 1.0% at 1 year (p = 0.006) and 8.0 ± 1.1% at 5 years (p = 0.122). Body weight and fat mass significantly reduced from baseline to 1 and 5 year(s), respectively. MRI-PDFF significantly decreased from 15.3 ± 7.8% at baseline to 11.9 ± 7.6% (p = 0.001) at 1 year and further decreased to 11.3 ± 5.7% (p = 0.013) at 5 years. Thus, a 5-year observation demonstrated that SGLT2 inhibitors have beneficial effects on liver steatosis in people with T2D and MASLD.

Introduction

Metabolic dysfunction-associated steatotic liver disease (MASLD) is the most common chronic liver disease, affecting 15–40% of the population worldwide owing the obesity epidemic [1-5]. Type 2 diabetes mellitus (T2D) is a major risk factor for MASLD, with 30–90% of people with T2D developing this complication [6-9]. MASLD may progress to metabolic dysfunction-associated steatohepatitis (MASH) and hepatocellular carcinoma [10, 11]. The higher the grade of MASLD, the greater the risk of progression to MASH and hepatocellular carcinoma [12]. Therefore, preventing or improving MASLD progression in people with T2D is important.

Previous reports have shown that thiazolidine [9], glucagon-like peptide-1 receptor agonists (GLP-1RA) [13, 14], and statins [15] significantly reduce liver adipogenesis and inflammation. Sodium glucose cotransporter-2 (SGLT2) inhibitors are a class of oral glucose-lowering drugs approved for diabetes treatment. SGLT2 inhibitors lower plasma glucose levels independently of insulin action and has some positive metabolic benefits by reducing renal glucose reabsorption [16-18]. Recently, large randomized controlled trials have shown that SGLT2 inhibitors also exert favorable long-term effects on the risk of major cardiovascular events, including heart failure and diabetic kidney disease, in people with T2D [19-23]. Although several reports have revealed the short-term efficacy of SGLT2 inhibitors on MASLD in people with T2D [24-27], their long-term efficacy remains unknown. Therefore, in the present study, we aimed to retrospectively analyze the effects of SGLT2 inhibitors on MASLD in people with T2D over 5 years by measuring metabolic profiles and magnetic resonance imaging-derived proton density fat fraction (MRI-PDFF).

Materials and Methods

Study design and participants

Using our medical record database, we retrospectively extracted data from people with T2D and MASLD aged 20 and above who were treated with SGLT2 inhibitors and underwent MRI before, 1 and 5 year(s) after administration of SGLT2 inhibitors between April 2014 and August 2021. In total, 35 people with T2D and MASLD on continuous treatment with SGLT2 inhibitors and underwent liver MRI before initiating SGLT2 inhibitor treatment were included in this study (Fig. 1). One case was lost to follow-up at the one-year mark, and eight cases were lost at the five-year follow-up, as they opted not to undergo liver MRI. Additionally, four cases were lost to follow-up because they transferred to another hospital. Finally, 34 and 22 subjects were included for further analysis at the one- and five-year follow-up, respectively. All people had been receiving dietary and/or exercise therapy at the time of first diagnosis and were subsequently treated with insulin and/or oral hypoglycemic agents before administration of SGLT2 inhibitors. The participants continued to take other glucose-lowering drugs; however, their dosages were adjusted as necessary by their physicians to minimize the risk of hypoglycemia or hyperglycemia.

Fig. 1

Flow chart for the inclusion of populations for analyses.

The protocol for this observational study was approved by the Kitasato University Medical School Ethics Committee (B20-292). All study methods were performed in accordance with the relevant guidelines and regulations of Kitasato University Hospital as well as the Ethical Guidelines for Medical and Health Research Involving Human Subjects in Japan and under the Code of Ethics of the Helsinki Declaration.

Clinical and biochemical analyses

Body composition, glucose and lipid metabolic parameters, hepatic function, and hepatic fibrosis markers were evaluated before and 1 and 5 years after the administration of SGLT2 inhibitors. Body composition, including body weight, fat mass (FM), muscle mass (MM), lean body mass (LBM), and total body water was measured using a body composition analyzer equipped with eight-electrode multifrequency bioelectrical impedance analysis technology (body composition analyzer MC-180; Tanita, Tokyo, Japan). After urination, participants positioned themselves on a floor scale, ensuring that the ball and heel of each foot were in contact with the electrodes. Subsequently, their body mass was recorded. They were then instructed to grasp the hand grips and hold them down by their sides, with metabolic electrodes in contact with their palms and thumbs. The participants kept their arms extended and away from their bodies in accordance with the manufacturer’s instructions, as described in our previous study [25]. The coefficient of variance of the impedance measure was reported to be 0.4% [28]. Glucose metabolic parameters included glycated hemoglobin (HbA1c) and glycated albumin (GA). GA was measured using an enzymatic synthesis method using with a glycated albumin-L assay kit (Lucica^TM; Asahi Kasei Pharma, Tokyo, Japan; coefficient of variation <0.3%). Lipid metabolic parameters included triglyceride, high-density lipoprotein (HDL) cholesterol, and low-density lipoprotein (LDL) cholesterol levels. Hepatic function was assessed by measuring the aspartate aminotransferase (AST), alanine aminotransferase (ALT), alkaline phosphatase (ALP), gamma-glutamyl transpeptidase (GGT), serum ferritin, cholinesterase, and serum albumin levels. The indirect non-invasive fibrosis scores of the Fibrosis-4 (FIB4) index and MASLD fibrosis score were calculated according to the following formula: FIB4 index = age (years) × AST (U/L)/(platelet count [10⁹/L] × √ALT (U/L)) [29]. The FIB4 index is a non-invasive fibrosis marker, and indices ≥2.67 were reported to exhibit an 80% positive predictive value, compared with liver biopsy results [30]. The MASLD fibrosis score was calculated as follows: –1.675 + 0.037 × age (years) + 0.094 × body mass index (BMI; kg/m²) + 1.13 × impaired fasting glucose/diabetes (yes = 1, no = 0) + 0.99 × AST/ALT ratio – 0.013 × platelet count (×10⁹/L) – 0.66 × albumin (g/dL) [31]. The MASLD fibrosis score accurately distinguishes patients with MASLD into those with and without advanced fibrosis. By applying the lower cut-off score of –1.455, advanced hepatic fibrosis can be excluded with high accuracy. Hepatic fibrosis markers include serum type IV collagen 7s, serum hyaluronic acid, and blood platelet count.

Measurement of MRI-PDFF

To measure intrahepatic fat content, we conducted 3.0-T MRI scans of the liver and measured MRI-PDFF using a modified Dixon method with a previously reported methodology (IDEAL-IQ; GE Healthcare, Waukesha, WI, USA) [32, 33] before, 1 and 5 year(s) after SGLT2 inhibitors administration. IDEAL-IQ images representing MRI-PDFF were acquired during a single breath hold using a Discovery MR750w Expert 3.0 Tesla or a Discovery MR750 3.0 Tesla (GE Healthcare). Imaging parameters on the MR750w scanner were as follows: repetition time/first echo time/Δecho time: 8.3/1.0/0.9 ms; number of echoes, six; flip angle, 4°; matrix, 160 × 160; slice thickness, 6 mm; bandwidth, ±111.11 kHz; field of vision, 36–50 cm; and acquisition time, 22 s. When using the MR750 scanner after modification, the following parameters were applied: repetition time/first echo time/Δecho time, 6.3/1.0/0.8 ms; flip angle, 3°; and acquisition time, 19 s, based on the manufacturer’s recommendation.

Statistical analysis

Statistical analysis was performed using SPSS 27.0 for Windows (International Business Machines Corp, Armonk, NY, USA). Data are presented as the means and standard deviation (SD), unless otherwise indicated. Clinical parameters were compared among the three time points using a Bonferroni-corrected paired t-test. p-values <0.05 after Bonferroni correction were considered statistically significant. Furthermore, we analyzed the correlation between changes in MRI-PDFF values and changes in clinical parameters.

Results

Participants

The baseline characteristics of the 34 participants were shown in Table 1. Twenty-six people (76%) were treated with canagliflozin (100 mg once daily) and eight (24%) with luseogliflozin (5 mg once daily). The mean MRI-PDFF was 15.3 ± 7.8%, and 23 people (68%) were classified as having moderate or severe MASLD (MRI-PDFF >11.3%) [33]. Eighteen people (53%) had dyslipidemia, and 14 (41%) had hypertension. None of the 34 people changed their medication for 1 year. Additionally, 22 people were monitored for 5 years after the start of treatment (Fig. 1). Twelve people (55%) began at least one new drug affecting MASLD 1 year after initiating the administration of SGLT2 inhibitors, which included thiazolidine (n = 5), GLP-1RA (n = 8), and statin (n = 4).

Table 1

Baseline characteristics of enrolled patients

	baseline
n	34
Sex (male/female)	14/20
Age (years)	52.9 ± 11.6
Height (cm)	162.6 ± 9.5
Body weight (kg)	79.9 ± 20.3
Body mass index	30.1 ± 7.3
<18.5: Underweight (n)	1
18.5–24.9: normal range (n)	4
25.0–29.9: Pre-obese (n)	14
30.0–34.9: obese class 1 (n)	10
35.0–39.9: obese class 2 (n)	3
>40.0: obese class 3 (n)	2
Duration of diabetes (years)	7.5 (1–30)
Diabetic complications (n)
Neuropathy (–)/(+)	23/11
Retinopathy (none/SDR/prePDR/PDR)	25/5/3/1
Nephropathy (NA/MA/ON/CRF)	15/12/8/0
Complications treated with medication
Dyslipidemia (n)	18 (53%)
Hypertension (n)	15 (47%)
Type of SGLT2 inhibitor
Canagliflozin (n)	26 (76%)
Luseogliflozin (n)	8 (24%)
Concomitant oral antidiabetic agents (n)
BG/TZD/DPP4i/SU/Glinide/αGI	25/1/26/5/6/3
Concomitant inject antidiabetic agents (n)
Insulin/GLP-1RA	9/0
Concomitant antilipidemic agents (n)
Statins/Bezafibrate/Ezetimibe	18/1/1
HbA1c (%)	8.9 ± 1.8
GA (%)	20.9 ± 7.0
MRI-PDFF (%)	15.3 ± 7.8
MASLD fibrosis score
<–1.455 (n)	13
–1.455 to 0.675 (n)	18
>0.675 (n)	3
FIB4 index	1.20 ± 0.59

Data are presented as number (n), means ± standard deviation (SD), or median (minimum–maximum).

SDR, simple diabetic retinopathy; prePDR, pre-proliferative diabetic retinopathy; PDR, proliferative diabetic retinopathy; NA, normoalbuminuria; MA, microalbuminuria; ON, overt nephropathy; CRF, chronic renal failure; SGLT2, sodium-glucose cotransporter 2; BG, biguanide; TZD, thiazolidine; DPP4i, dipeptidyl-peptidase-4 inhibitor; SU, sulfonylurea; αGI, α-Glucosidase inhibitor; GLP-1RA, glucagon like peptide-1 receptor agonist; GA, glycated albumin; MRI-PDFF, proton density fat fraction derived by magnetic resonance imaging; MASLD, metabolic dysfunction-associated steatotic liver disease.

Efficacy of SGLT2 inhibitors

After administration of SGLT2 inhibitors, body weight and FM significantly decreased from 79.9 ± 20.3 kg and 30.3 ± 15.8 kg at baseline to 77.2 ± 20.7 kg (p < 0.001) and 27.8 ± 15.6 kg (p < 0.001) at 1 year and 79.2 ± 22.5 kg (p = 0.003) and 25.8 ± 12.5 kg (p = 0.002) at 5 years, respectively. However, MM and LBM did not change (Table 2). The FM of the trunk and both arms and legs exhibited a significant decrease at 1 year and 5 years, respectively. No significant changes were observed in trunk MM. Although a significant decrease was noted in MM in both arms at 1 year, no significant change was found at 5 years. The MM of the right legs showed no change at 1 year and 5 years, whereas that of the left leg remained unchanged at 1 year and significantly decreased at 5 years (Table 3).

Table 2

Changes in clinical parameters of enrolled patients

	Baseline	1 year	p value	5 years	p value
Body weight (kg)	79.9 ± 20.3 (45.1–135.5)	77.2 ± 20.7 (43.9–132.9)	<0.001	79.2 ± 22.5 (45.3–133.0)	0.003
Body mass index (kg/m2)	30.1 ± 7.3 (16.8–56.8)	29.1 ± 7.3 (17.1–54.4)	<0.001	29.9 ± 9.0 (21.5–60.7)	0.006
Fat mass (kg)	30.3 ± 15.8 (10.3–87.6)	27.8 ± 15.6 (9.2–84.3)	<0.001	25.8 ± 12.5 (9.8–56.0)	0.002
Muscle mass (kg)	46.8 ± 10.5 (32.5–68.1)	46.6 ± 10.9 (32.2–68.2)	0.573	47.4 ± 11.8 (30.2–68.5)	0.924
Lean body mass (kg)	49.6 ± 10.9 (34.3–71.8)	49.3 ± 11.4 (34.1–71.9)	0.542	50.8 ± 12.0 (31.9–72.2)	0.646
Total body water (kg)	36.3 ± 7.1 (23.0–55.4)	36.0 ± 7.6 (24.2–52.5)	0.382	36.7 ± 8.4 (23.0–53.8)	0.725
BWC (%)	46.5 ± 6.6 (23.7–61.4)	47.8 ± 6.7 (23.4–59.1)	<0.001	48.6 ± 5.4 (40.4–59.1)	0.042
HbA1c (%)	8.9 ± 1.8 (6.5–15.3)	7.8 ± 1.0 (5.3–9.7)	0.006	8.0 ± 1.1 (5.7–9.9)	0.122
GA (%)	20.6 ± 6.9 (13.6–47.4)	18.1 ± 3.7 (13.2–31.8)	0.083	17.8 ± 3.0 (11.7–22.2)	0.247
Triglycerides (mg/dL)	206.9 ± 128.4 (65–600)	202.7 ± 122.3 (57–553)	0.801	224.4 ± 192.3 (83–917)	0.888
LDL cholesterol (mg/dL)	114.1 ± 30.3 (50–181)	114.5 ± 33.2 (48–206)	0.921	103.0 ± 31.6 (43–171)	0.037
HDL cholesterol (mg/dL)	50.0 ± 9.5 (33–70)	55.1 ± 11.0 (32–88)	<0.001	54.0 ± 12.4 (32–79)	0.013
AST (U/L)	42.8 ± 26.1 (13–127)	35.6 ± 30.8 (12–182)	0.097	33.1 ± 19.5 (14–73)	0.002
ALT (U/L)	65.0 ± 48.9 (8–269)	47.7 ± 37.2 (10–146)	0.008	49.5 ± 33.9 (7–133)	0.002
ALP (U/L)	286.7 ± 98.9 (136–531)	251.3 ± 74.1 (130–447)	0.005	91.0 ± 21.0 (49–124)	0.090
GGT (U/L)	98.6 ± 123.6 (9–730)	73.5 ± 84.5 (10–452)	0.011	75.0 ± 76.3 (14–285)	0.162
Blood platelet count (×104/μL)	25.8 ± 7.4 (17.8–52.8)	25.0 ± 7.9 (16.0–56.4)	0.099	24.5 ± 7.2 (16.4–48.9)	0.400
Serum albumin (g/dL)	4.4 ± 0.4 (3.6–5.2)	4.4 ± 0.4 (3.9–4.8)	0.634	4.4 ± 0.5 (3.1–5.1)	0.917
Serum ferritin (ng/mL)	192.1 ± 196.6 (4–863)	121.7 ± 112.8 (14–475)	0.121	89.5 ± 80.8 (9–283)	0.044
Cholinesterase (U/L)	395.8 ± 80.0 (247–610)	384.8 ± 74.0 (241–608)	0.087	377.7 ± 87.1 (173–594)	0.002
Serum type IV collagen 7s (ng/mL)	5.2 ± 1.6 (3.0–11.0)	4.9 ± 1.4 (2.3–7.6)	0.297	4.6 ± 1.8 (2.7–9.7)	0.003
Serum hyaluronic acid (ng/mL)	46.6 ± 37.5 (9–151)	48.2 ± 40.9 (9–170)	0.755	51.3 ± 45.3 (9–149)	0.864
FIB4 index	1.2 ± 0.6 (0.29–2.87)	1.2 ± 0.7 (0.37–3.82)	0.911	1.2 ± 0.5 (0.39–2.36)	0.267
MASLD fibrosis score	–1.3 ± 1.4 (–4.71–2.29)	–1.2 ± 1.4 (–5.71–1.86)	0.297	–0.9 ± 1.8 (–4.73–3.81)	0.121

Data are expressed as the means ± standard deviation (SD).

p-value vs. baseline, the paired Student’s t test with pre-hoc Bonferroni correction of the level of significance (0.05/3 = 0.017) for multiple comparisons.

FM, fat mass; MM, muscle mass; TBW, total body water; BWC, body water content (TBW/BW × 100); GA, glycated albumin; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; GGT, gamma glutamyl transpeptidase; MASLD, metabolic dysfunction-associated steatotic liver disease.

Table 3

Changes in body composition of enrolled patients

	Baseline	1 year	p value	5 years	p value
Truncal MM (kg)	24.1 ± 4.5 (18.0–32.5)	24.4 ± 4.7 (18.3–32.5)	0.378	24.7 ± 5.8 (16.9–42.4)	0.507
Truncal FM (kg)	16.0 ± 6.5 (5.1–32.5)	14.5 ± 6.7 (4.2–32.8)	<0.001	14.6 ± 6.8 (5.4–29.2)	0.004
MM in right arm (kg)	2.6 ± 0.7 (1.4–4.1)	2.5 ± 0.7 (1.5–4.0)	0.014	2.6 ± 0.8 (1.3–4.3)	0.789
FM in right arm (kg)	1.3 ± 0.7 (0.3–3.1)	1.1 ± 0.7 (0.3–2.5)	<0.001	1.1 ± 0.7 (0.2–2.5)	0.008
MM in left arm (kg)	2.4 ± 0.6 (1.3–3.8)	2.4 ± 0.6 (1.4–3.6)	0.023	2.4 ± 0.7 (1.3–3.7)	0.535
FM in left arm (kg)	1.4 ± 0.7 (0.4–3.0)	1.2 ± 0.7 (0.3–2.7)	<0.001	1.5 ± 1.3 (0.1–5.4)	0.004
MM in right leg (kg)	9.1 ± 2.7 (4.9–16.7)	9.0 ± 2.8 (5.3–15.9)	0.166	8.8 ± 3.1 (5.4–16.4)	0.144
FM in right leg (kg)	5.1 ± 2.2 (2.0–13.0)	4.7 ± 2.2 (2.0–11.8)	<0.001	4.7 ± 2.5 (2.1–11.4)	0.001
MM in left leg (kg)	9.0 ± 2.7 (4.9–17.1)	8.8 ± 2.7 (5.3–16.2)	0.080	8.8 ± 2.9 (5.4–16.5)	0.003
FM in left leg (kg)	5.0 ± 2.3 (2.0–13.2)	4.6 ± 2.2 (2.0–11.9)	<0.001	4.6 ± 2.4 (1.9–11.3)	<0.001

Data are expressed as the means ± standard deviation (minimum–maximum). p-value vs. baseline, the paired t-test with pre-hoc Bonferroni correction of the level of significance (0.05/3 = 0.017) for multiple comparisons. FM, fat mass; MM, muscle mass.

As for changes in glucose metabolism, HbA1c levels significantly improved from 8.9 ± 1.8% at baseline to 7.8 ± 1.0% at 1 year (p = 0.006) and 8.0 ± 1.1% at 5 years (p = 0.122); GA decreased from 20.6 ± 6.9% at baseline to 18.1 ± 3.7% at 1 year (p = 0.083) and 17.8 ± 3.0% at 5 years (p = 0.247). As for the changes in lipid metabolism, serum triglyceride levels changed from 206.9 ± 128.4 mg/dL at baseline to 202.7 ± 122.3 mg/dL at 1 year (p = 0.801) and 224.4 ± 192.3 mg/dL at 5 years (p = 0.888); LDL-cholesterol levels changed from 114.1 ± 30.3 mg/dL at baseline to 114.5 ± 33.2 mg/dL at 1 year (p = 0.921) and 103.0 ± 31.6 mg/dL at 5 years (p = 0.037). Notably, serum HDL-cholesterol levels were significantly increased from 50.0 ± 9.5 mg/dL at baseline to 55.1 ± 11.0 mg/dL at 1 year (p < 0.001) and 54.0 ± 12.4 mg/dL at 5 years (p = 0.013).

MRI analysis of hepatic fat levels revealed a significant decrease in MRI-PDFF values, declining from 15.3 ± 7.8% at baseline to 11.9 ± 7.6% at 1 year (p = 0.001) and further to 11.3 ± 5.7% at 5 years (p = 0.013) (Fig. 2). Regarding changes in serum levels of hepatic enzymes, AST changed from 42.8 ± 26.1 U/L at baseline to 35.6 ± 30.8 U/L at 1 year (p = 0.097) and 33.1 ± 19.5 U/L at 5 years (p = 0.002); ALT changed from 65.0 ± 48.9 U/L at baseline to 47.7 ± 37.2 U/L at 1 year (p = 0.008) and 49.5 ± 33.9 U/L at 5 years (p = 0.002); ALP changed from 286.7 ± 98.9 U/L at baseline to 251.3 ± 74.1 U/L at 1 year (p = 0.005) and 91.0 ± 21.0 U/L at 5 years (p = 0.090); GGT changed from 98.6 ± 123.6 U/L at baseline to 73.5 ± 84.5 U/L at 1 year (p = 0.011) and 75.0 ± 76.3 U/L at 5 years (p = 0.162). The serum levels of ferritin, cholinesterase and type IV collagen 7s were not significantly altered at 1 year, but showed significant improvement at 5 years. Serum albumin, blood platelet count and serum hyaluronic acid showed no significant changes. In addition, the FIB4 index and MASLD fibrosis score did not show significant change (Table 2).

Fig. 2

Magnetic resonance imaging-derived proton density fat fraction (MRI-PDFF) of liver fat before and after the administration of SGLT2 inhibitors. MRI-PDFF was assessed using 3-T magnetic resonance imaging immediately at baseline, and at 1 and 5 years after the initiation of SGLT2 inhibitor treatment. Each circle represents a single patient measurement. Parametric data are expressed and were analyzed using the paired Student’s t-test with pre-hoc Bonferroni correction for multiple comparisons (0.05/3 = 0.017). * p < 0.05, ** p < 0.005 vs. baseline.

We further analyzed the correlation of changes in MRI-PDFF values with changes in clinical parameters (Table 4). Five-year change in MRI-PDFF values significantly correlated with changes in body weight (r = 0.593, p = 0.004), BMI (r = 0.593, p = 0.004), FM (r = 0.525, p = 0.015), AST (r = 0.516, p = 0.014), GGT (r = 0.578, p = 0.005), triglycerides (r = 0.458, p = 0.032), LDL cholesterol (r = 0.456, p = 0.033), type IV collagen 7s concentration (r = 0.516, p = 0.017), and FIB4 index (r = 0.424, p = 0.049), but not with changes in HbA1c, GA, LBM, and MM.

Table 4

Associations of change in MRI-PDFF with changes in other clinical parameters

	Correlation coefficients	p values
ΔBody weight (kg)	0.593	0.004
ΔBMI	0.593	0.004
ΔAST (U/L)	0.516	0.014
ΔALT (U/L)	0.256	0.250
ΔALP (U/L)	0.209	0.364
ΔGGT (U/L)	0.578	0.005
ΔCholinesterase (U/L)	0.298	0.190
ΔSerum albumin (mg/dL)	–0.271	0.223
ΔPlatelet count (*103/μL)	0.076	0.736
ΔTriglycerides (mg/dL)	0.458	0.032
ΔLDL cholesterol (mg/dL)	0.456	0.033
ΔHDL cholesterol (mg/dL)	–0.123	0.586
ΔHbA1c (%)	0.369	0.091
ΔGA (%)	0.364	0.126
ΔCreatinine (mg/dL)	0.079	0.726
ΔEstimated glomerular filtration rate (mL/min/1.73 m2)	0.068	0.764
ΔSerum ferritin (ng/mL)	0.422	0.172
ΔFIB4 index	0.424	0.049
ΔMASLD fibrosis score	0.324	0.142
ΔType IV collagen 7s (ng/mL)	0.516	0.017
ΔFM (kg)	0.525	0.015
ΔMM (kg)	0.157	0.535
ΔLBM (kg)	0.145	0.531

Pearson’s correlation coefficients or Spearman’s correlation coefficients are shown. Changes (Δ) were calculated by subtracting the baseline value from the value 5 years after treatment with SGLT2 inhibitors.

BMI, body mass index; AST, aspartate aminotransferase; ALT, alanine aminotransferase; ALP, alkaline phosphatase; GGT, gamma glutamyl transpeptidase; GA, glycated albumin; MASLD, metabolic dysfunction-associated steatotic liver disease; FM, fat mass; MM, muscle mass; LBM, lean body mass.

None of the participants had severe hypoglycemia, severe genital or urinary tract infections, or euglycemic diabetic ketoacidosis during the observation period. There were no cases of discontinuation of SGLT2 inhibitors due to adverse events.

Discussion

The present study revealed that the five-year administration of SGLT2 inhibitors significantly improved MRI-PDFF values and serum levels of hepatic enzymes. This suggests the beneficial effects of SGLT2 inhibitors on liver steatosis in people with T2D and MASLD. A previous study showed that a 6% reduction in body weight resulting from a 3-month hypocaloric diet was accompanied by a significant reduction in intrahepatic lipid content [34]. Recently, some studies have suggested that SGLT2 inhibitors ameliorate hepatic fibrosis, insulin resistance and lipotoxicity in animal models [35, 36]. Moreover, they demonstrate a more effective reduction in body weight and visceral fat volume more effectively than thiazolidine in people with T2D [37]. Additionally, the combination of SGLT2 inhibitors and incretin-based treatments, such as dipeptidyl peptidase-4 inhibitors or GLP-1RA, has been shown to ameliorate of MASH [34, 38]. In a 3-year post-marketing surveillance study involving Japanese people treated with ipragliflozin, significant improvements in hepatic parameters, such as AST, ALT, ALP, and fatty liver index, were reported in people with T2D and impaired liver function [39]. Additionally, a study using canagliflozin in a small number of cases indicated improvements in liver histological findings at 5 years [40]. Our study not only further supports the effectiveness of SGLT2 inhibitors, which is consistent with the findings of these studies, but also represents the first long-term assessment of liver fat content using MRI. Although some reports showed a 13–39% reduction in liver fat at 8–52 weeks [25, 41], this study showed a 32% reduction in liver fat even 5 years after the initiation of SGLT2 inhibitors. In addition, the mean body weight was reduced by 3.1% at 1 year and by 4.3% at 5 years, and the reduction in liver fat correlated with the reduction in body weight and FM (Tables 2 and 3). This result contrasts with a previous report that showed an increase in appetite and a recovery of body weight during the treatment with SGLT2 inhibitors [42]. Moreover, in this study, the five-year alteration in MRI-PDFF demonstrated correlation with variations in body weight and BMI, whereas no significant correlation was observed with glycemic markers such as HbA1c and GA. This suggests that changes in MRI-PDFF may depend on improvements in body weight rather than improvements in blood glucose levels.

No individuals added any drugs during the first year of SGLT2 inhibitors treatment. However, 12 people were received additional drugs, including thiazolidine (n = 5), GLP-1RA (n = 8) and statins (n = 4) between the second and fifth years of follow-up. Therefore, these additional drugs may affect liver function after the second year of follow-up, as previously described [43]. The lack of uniformity in medications and/or interventions during the observational period is a limitation in this retrospective study. Further studies are required to investigate the interactions between these drugs and SGLT2 inhibitors in people with T2D and MASLD.

Although this study possesses the strength of consecutively monitoring hepatic fat storage by MRI after a 5-year administration of SGLT2 inhibitors, several limitations should be considered. First, this study is retrospective in nature and has a small sample size. Therefore, our results may have been influenced by confounding factors which could affect the improvement of liver steatosis. Second, hepatic fat storage was evaluated using MRI-PDFF values without hepatic biopsy or histological evaluation.

In conclusion, long-term retrospective data on people with T2D and MASLD demonstrated that SGLT2 inhibitors have beneficial effects on hepatic fat storage and liver function. This is evident from the results of laboratory tests measuring liver enzymes for 5 years. Such improvements could contribute to the optimal management of those people.

Acknowledgment

Disclosure

TM is a member of Endocrine Journal’s Editorial Board.

Funding

This work was supported in part by a grant from JSPS KAKENHI Grant Number 22K15674 and the Terumo Life Science Foundation Grant Number 21-III5030. No additional external funding was received for this study. The authors declare no conflicts of interest.

Contribution statement

All authors have significantly contributed to the study. AS, AH, and TM had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. AS and AH are responsible for the conception and design of the study, evaluated the data and are the primary authors of the manuscript. AS, AH, SO, RF, NS, KM, TTa and TTo collected the data. AS, AH, SO, TTa and TTo calculated and analyzed the HFF using MRI. AS and AH analyzed the data. All authors reviewed the data and took responsibility for the final assembly and editing of the manuscript.

References

• 1   Alisi  A,  Manco  M,  Pezzullo  M,  Nobili  V (2011) Fructose at the center of necroinflammation and fibrosis in nonalcoholic steatohepatitis. Hepatology 53: 372–373.
• 2   Targher  G,  Byrne  CD (2013) Clinical review: nonalcoholic fatty liver disease: a novel cardiometabolic risk factor for type 2 diabetes and its complications. J Clin Endocrinol Metab 98: 483–495.
• 3   Wong  VW,  Wong  GL,  Yeung  DK,  Lau  TK,  Chan  CK, et al. (2015) Incidence of non-alcoholic fatty liver disease in Hong Kong: a population study with paired proton-magnetic resonance spectroscopy. J Hepatol 62: 182–189.
• 4   Rinella  ME,  Lazarus  JV,  Ratziu  V,  Francque  SM,  Sanyal  AJ, et al. (2023) A multisociety Delphi consensus statement on new fatty liver disease nomenclature. Hepatology 78: 1966–1986.
• 5   Younossi  ZM,  Golabi  P,  Paik  JM,  Henry  A,  Van Dongen  C, et al. (2023) The global epidemiology of nonalcoholic fatty liver disease (NAFLD) and nonalcoholic steatohepatitis (NASH): a systematic review. Hepatology 77: 1335–1347.
• 6  (2016) EASL-EASD-EASO Clinical Practice Guidelines for the management of non-alcoholic fatty liver disease. J Hepatol 64: 1388–1402.
• 7   Amarapurkar  DN,  Hashimoto  E,  Lesmana  LA,  Sollano  JD,  Chen  PJ, et al. (2007) How common is non-alcoholic fatty liver disease in the Asia-Pacific region and are there local differences? J Gastroenterol Hepatol 22: 788–793.
• 8   Kwok  R,  Choi  KC,  Wong  GL,  Zhang  Y,  Chan  HL, et al. (2016) Screening diabetic patients for non-alcoholic fatty liver disease with controlled attenuation parameter and liver stiffness measurements: a prospective cohort study. Gut 65: 1359–1368.
• 9   Targher  G,  Lonardo  A,  Byrne  CD (2018) Nonalcoholic fatty liver disease and chronic vascular complications of diabetes mellitus. Nat Rev Endocrinol 14: 99–114.
• 10   Ascha  MS,  Hanouneh  IA,  Lopez  R,  Tamimi  TA,  Feldstein  AF, et al. (2010) The incidence and risk factors of hepatocellular carcinoma in patients with nonalcoholic steatohepatitis. Hepatology 51: 1972–1978.
• 11   Wong  VW,  Wong  GL,  Choi  PC,  Chan  AW,  Li  MK, et al. (2010) Disease progression of non-alcoholic fatty liver disease: a prospective study with paired liver biopsies at 3 years. Gut 59: 969–974.
• 12   Taylor  RS,  Taylor  RJ,  Bayliss  S,  Hagstrom  H,  Nasr  P, et al. (2020) Association between fibrosis stage and outcomes of patients with nonalcoholic fatty liver disease: a systematic review and meta-analysis. Gastroenterology 158: 1611–1625. e1612.
• 13   Armstrong  MJ,  Gaunt  P,  Aithal  GP,  Barton  D,  Hull  D, et al. (2016) Liraglutide safety and efficacy in patients with non-alcoholic steatohepatitis (LEAN): a multicentre, double-blind, randomised, placebo-controlled phase 2 study. Lancet 387: 679–690.
• 14   Newsome  PN,  Buchholtz  K,  Cusi  K,  Linder  M,  Okanoue  T, et al. (2021) A placebo-controlled trial of subcutaneous semaglutide in nonalcoholic steatohepatitis. N Engl J Med 384: 1113–1124.
• 15   Dongiovanni  P,  Petta  S,  Mannisto  V,  Mancina  RM,  Pipitone  R, et al. (2015) Statin use and non-alcoholic steatohepatitis in at risk individuals. J Hepatol 63: 705–712.
• 16   Hasan  FM,  Alsahli  M,  Gerich  JE (2014) SGLT2 inhibitors in the treatment of type 2 diabetes. Diabetes Res Clin Pract 104: 297–322.
• 17   Scheen  AJ (2020) Sodium-glucose cotransporter type 2 inhibitors for the treatment of type 2 diabetes mellitus. Nat Rev Endocrinol 16: 556–577.
• 18   Cusi  K (2020) Time to include nonalcoholic steatohepatitis in the management of patients with type 2 diabetes. Diabetes Care 43: 275–279.
• 19   Schernthaner  G,  Schernthaner-Reiter  MH,  Schernthaner  GH (2016) EMPA-REG and other cardiovascular outcome trials of glucose-lowering agents: implications for future treatment strategies in type 2 diabetes mellitus. Clin Ther 38: 1288–1298.
• 20   Wu  JH,  Foote  C,  Blomster  J,  Toyama  T,  Perkovic  V, et al. (2016) Effects of sodium-glucose cotransporter-2 inhibitors on cardiovascular events, death, and major safety outcomes in adults with type 2 diabetes: a systematic review and meta-analysis. Lancet Diabetes Endocrinol 4: 411–419.
• 21   Zelniker  TA,  Braunwald  E (2020) Clinical benefit of cardiorenal effects of sodium-glucose cotransporter 2 inhibitors: JACC state-of-the-art review. J Am Coll Cardiol 75: 435–447.
• 22   Zelniker  TA,  Braunwald  E (2020) Mechanisms of cardiorenal effects of sodium-glucose cotransporter 2 inhibitors: JACC state-of-the-art review. J Am Coll Cardiol 75: 422–434.
• 23   Zelniker  TA,  Wiviott  SD,  Raz  I,  Im  K,  Goodrich  EL, et al. (2019) Comparison of the effects of glucagon-like peptide receptor agonists and sodium-glucose cotransporter 2 inhibitors for prevention of major adverse cardiovascular and renal outcomes in type 2 diabetes mellitus. Circulation 139: 2022–2031.
• 24   Sumida  Y,  Yoneda  M (2018) Current and future pharmacological therapies for NAFLD/NASH. J Gastroenterol 53: 362–376.
• 25   Inoue  M,  Hayashi  A,  Taguchi  T,  Arai  R,  Sasaki  S, et al. (2019) Effects of canagliflozin on body composition and hepatic fat content in type 2 diabetes patients with non-alcoholic fatty liver disease. J Diabetes Investig 10: 1004–1011.
• 26   Sumida  Y,  Murotani  K,  Saito  M,  Tamasawa  A,  Osonoi  Y, et al. (2019) Effect of luseogliflozin on hepatic fat content in type 2 diabetes patients with non-alcoholic fatty liver disease: a prospective, single-arm trial (LEAD trial). Hepatol Res 49: 64–71.
• 27   Hsiang  JC,  Wong  VW (2020) SGLT2 inhibitors in liver patients. Clin Gastroenterol Hepatol 18: 2168–2172. e2162.
• 28   Leahy  S,  O’Neill  C,  Sohun  R,  Jakeman  P (2012) A comparison of dual energy X-ray absorptiometry and bioelectrical impedance analysis to measure total and segmental body composition in healthy young adults. Eur J Appl Physiol 112: 589–595.
• 29   Sterling  RK,  Lissen  E,  Clumeck  N,  Sola  R,  Correa  MC, et al. (2006) Development of a simple noninvasive index to predict significant fibrosis in patients with HIV/HCV coinfection. Hepatology 43: 1317–1325.
• 30   Shah  AG,  Lydecker  A,  Murray  K,  Tetri  BN,  Contos  MJ, et al. (2009) Comparison of noninvasive markers of fibrosis in patients with nonalcoholic fatty liver disease. Clin Gastroenterol Hepatol 7: 1104–1112.
• 31   Angulo  P,  Hui  JM,  Marchesini  G,  Bugianesi  E,  George  J, et al. (2007) The NAFLD fibrosis score: a noninvasive system that identifies liver fibrosis in patients with NAFLD. Hepatology 45: 846–854.
• 32   Hope  TA,  Ohliger  MA,  Qayyum  A (2014) MR imaging of diffuse liver disease: from technique to diagnosis. Radiol Clin North Am 52: 709–724.
• 33   Imajo  K,  Kessoku  T,  Honda  Y,  Tomeno  W,  Ogawa  Y, et al. (2016) Magnetic resonance imaging more accurately classifies steatosis and fibrosis in patients with nonalcoholic fatty liver disease than transient elastography. Gastroenterology 150: 626–637. e627.
• 34   Sato  F,  Tamura  Y,  Watada  H,  Kumashiro  N,  Igarashi  Y, et al. (2007) Effects of diet-induced moderate weight reduction on intrahepatic and intramyocellular triglycerides and glucose metabolism in obese subjects. J Clin Endocrinol Metab 92: 3326–3329.
• 35   Tahara  A,  Kurosaki  E,  Yokono  M,  Yamajuku  D,  Kihara  R, et al. (2014) Effects of sodium-glucose cotransporter 2 selective inhibitor ipragliflozin on hyperglycaemia, oxidative stress, inflammation and liver injury in streptozotocin-induced type 1 diabetic rats. J Pharm Pharmacol 66: 975–987.
• 36   Hayashizaki-Someya  Y,  Kurosaki  E,  Takasu  T,  Mitori  H,  Yamazaki  S, et al. (2015) Ipragliflozin, an SGLT2 inhibitor, exhibits a prophylactic effect on hepatic steatosis and fibrosis induced by choline-deficient l-amino acid-defined diet in rats. Eur J Pharmacol 754: 19–24.
• 37   Ito  D,  Shimizu  S,  Inoue  K,  Saito  D,  Yanagisawa  M, et al. (2017) Comparison of ipragliflozin and pioglitazone effects on nonalcoholic fatty liver disease in patients with type 2 diabetes: a randomized, 24-week, open-label, active-controlled trial. Diabetes Care 40: 1364–1372.
• 38   Jojima  T,  Tomotsune  T,  Iijima  T,  Akimoto  K,  Suzuki  K, et al. (2016) Empagliflozin (an SGLT2 inhibitor), alone or in combination with linagliptin (a DPP-4 inhibitor), prevents steatohepatitis in a novel mouse model of non-alcoholic steatohepatitis and diabetes. Diabetol Metab Syndr 8: 45.
• 39   Tobe  K,  Maegawa  H,  Nakamura  I,  Uno  S (2021) Effect of ipragliflozin on liver function in Japanese type 2 diabetes mellitus patients: subgroup analysis of a 3-year post-marketing surveillance study (STELLA-LONG TERM). Endocr J 68: 905–918.
• 40   Akuta  N,  Kawamura  Y,  Fujiyama  S,  Saito  S,  Muraishi  N, et al. (2022) Favorable impact of long-term SGLT2 inhibitor for NAFLD complicated by diabetes mellitus: a 5-year follow-up study. Hepatol Commun 6: 2286–2297.
• 41   Kuchay  MS,  Krishan  S,  Mishra  SK,  Farooqui  KJ,  Singh  MK, et al. (2018) Effect of empagliflozin on liver fat in patients with type 2 diabetes and nonalcoholic fatty liver disease: a randomized controlled trial (E-LIFT trial). Diabetes Care 41: 1801–1808.
• 42   Ferrannini  G,  Hach  T,  Crowe  S,  Sanghvi  A,  Hall  KD, et al. (2015) Energy balance after sodium-glucose cotransporter 2 inhibition. Diabetes Care 38: 1730–1735.
• 43   DeFronzo  RA (2017) Combination therapy with GLP-1 receptor agonist and SGLT2 inhibitor. Diabetes Obes Metab 19: 1353–1362.