Correspondence to editorial on “Comparison of glucagon-like peptide-1 receptor agonists and thiazolidinediones on treating nonalcoholic fatty liver disease: a network meta-analysis”

Article information

Clin Mol Hepatol. 2024;30(4):989-991

Publication date (electronic) : 2024 August 27

doi : https://doi.org/10.3350/cmh.2024.0705

Hayeon Kim¹, Min Jeong Park², Myeong Gyu Kim³, Kyungim Kim^,1,4

¹College of Pharmacy, Korea University, Sejong, Korea

²Division of Endocrinology & Metabolism, Department of Internal Medicine, Korea University College of Medicine, Seoul, Korea

³College of Pharmacy, Ewha Womans University, Seoul, Korea

⁴Institute of Pharmaceutical Science, Korea University, Sejong, Korea

Corresponding author : Kyungim Kim College of Pharmacy, Korea University, 2511 Sejong-ro, Sejong, 30019, Korea Tel: +82-44-860-1624, Fax: +82-44-860-1606, E-mail: kim_ki@korea.ac.kr

Editor: Han Ah Lee, Chung-Ang University College of Medicine, Korea

Received 2024 August 23; Accepted 2024 August 27.

Dear Editor,

We appreciate Ren et al. for their interest in our study and for providing valuable insights into incretin-based therapy for patients with nonalcoholic fatty liver disease (NAFLD), which was referred to as metabolic dysfunctionassociated steatotic liver disease (MASLD) in the editorial [1,2]. As the authors indicated, we acknowledged that there are still some issues to be addressed when using incretinbased therapy to treat NAFLD. In our study, glucagon-like peptide-1 receptor agonists (GLP-1RAs) were significantly more effective than thiazolidinediones (TZDs) in reducing liver fat content measured by proton magnetic resonance spectroscopy (1H-MRS) in overweight or obese patients with NAFLD or nonalcoholic steatohepatitis (NASH). For controlled attenuation parameter (CAP) and biopsy-based outcomes including fibrosis stage, the superiority of GLP-1RAs was not statistically significant. In the editorial, the authors noted inconsistencies in study outcomes and called for studies targeting non-obese or lean populations. We would like to go over these issues in detail.

Regarding the disparity between non-invasive and invasive liver outcomes in our study results, the editorial especially focused on liver fibrosis, which can be detected by invasive methods. The authors stated that GLP-1RAs may have a positive effect on metabolic dysfunction-associated steatohepatitis (MASH)-related fibrosis, based on the evidence from previous TZD studies and our results showing a trend toward better efficacy of GLP-1RAs than TZDs [2,3]. We agree with this suggestion, yet some points need consideration. Previous meta-analysis has shown that TZDs have the potential to improve or prevent the progression of liver fibrosis [3]. TZDs can improve liver fibrosis in patients with NASH, regardless of whether they have advanced (F3-F4) fibrosis at baseline [3]. However, GLP-1RAs have shown inconsistent results pertaining to improving liver fibrosis in patients with NAFLD [4,5]. A randomized controlled trial reported that patients with more severe baseline fibrosis (F3-F4) are less likely to respond to the GLP-1RAs [4]. Reportedly, early liver fibrosis is likely to regress and resolve when extracellular matrix crosslinking and angiogenesis are absent [6]. Therefore, to make an accurate assessment of the treatment effects on fibrosis reduction, patients should be analyzed according to their baseline fibrosis stage before starting the drugs. However, due to a lack of trials reporting the effects of GLP-1RAs on liver histological endpoints, we could not identify how the drugs affected liver fibrosis based on patients’ baseline fibrosis stage in our study. Furthermore, because type 2 diabetes is a potential risk factor for the development and progression of advanced fibrosis, the patients’ diabetic status should also be considered [7]. More clinical trials analyzing the effect of GLP-1RAs on liver fibrosis according to baseline fibrosis stage and diabetes in patients with NAFLD will lead to more solid conclusions.

When it comes to non-invasive liver outcomes, there was an imbalance between the results from 1H-MRS and CAP in our study. GLP-1RAs had a greater effect than TZDs on reducing liver fat when detected by 1H-MRS in obese or overweight patients with NAFLD/NASH. However, according to CAP, GLP-1RAs only showed a trend of better effects than TZDs and did not reach statistical significance. This inconsistency could be attributed to obese people’s lower CAP accuracy compared to non-obese people [8,9]. CAP may struggle to differentiate between steatosis grades S2 and S3 in patients with a BMI ≥28 kg/m [2.10]. Meanwhile, MRS has been reported as a precise and effective method for assessing treatment response in the absence of iron deposition, fibrosis, or coexisting liver pathologies [9].

In the editorial, the authors emphasized the importance of further research into incretin-based therapy in the management of non-obese or lean subjects with NAFLD. We completely agree with the suggestions. Non-obese individuals with NAFLD are expected to account for approximately 40% of the NAFLD/NASH population, but the majority of the studies on pharmacological treatments for NASH patients have focused on overweight and obese patients [11]. Due to data scarcity, it was not able to compare the effects of GLP-1RAs and TZDs in non-obese or lean patients with NAFLD/NASH in our study. However, based on previous research, we anticipate that GLP-1RAs can have a promising effect on improving NASH through visceral fat loss in non-obese patients with NAFLD/NASH. There is a strong link between visceral fat obesity and non-obese NAFLD, and GLP-1RAs have been shown to reduce visceral fat mass [12,13]. From a mechanistic standpoint, there is evidence that GLP-1RAs may help improve NASH in non-obese individuals. A previous study found that GLP-1RAs can lower leptin levels, a proinflammatory cytokine, regardless of fat mass reduction [14]. Leptin causes liver fibrosis by inhibiting peroxisome proliferator-activated receptor-γ (PPAR-γ), which prevents hepatic stellate cells from differentiating into myofibroblasts [15]. Thus, GLP-1RAs may have a therapeutic effect on NASH in non-obese patients by modulating proinflammatory cytokines like leptin, which are involved in NASH development. However, given the lack of evidence, more research into non-obese or lean populations with NAFLD/NASH is needed.

In conclusion, our findings suggest that GLP-1RAs, an incretin-based glucose-lowering drug, may be more effective in overweight or obese NAFLD/NASH patients than TZDs, particularly in terms of reducing liver fat content. For invasive outcomes associated with liver fibrosis, clinical trials investigating the effects of GLP-1RAs in NAFLD/NASH patients based on baseline fibrosis stage and diabetic status are warranted. Furthermore, future research should focus on the effects of GLP-1RAs and TZDs in non-obese or lean NAFLD/NASH patients. Because incretin may have a positive effects on multiple organs involved in NAFLD/NASH pathophysiology, research into the use of various incretinbased therapies in these patients will help to expand the limited therapeutic options.