What this is
- This analysis compares the efficacy and safety of Resmetirom, FGF21 analogs, and GLP-1 agonists for treating and .
- and are liver diseases linked to obesity and metabolic syndrome, impacting health and healthcare costs.
- The study synthesizes data from clinical trials to inform treatment decisions and optimize patient management.
Essence
- Resmetirom and FGF21 analogs show promise in treating and , with Resmetirom particularly effective in reducing liver fat and improving liver enzymes.
Key takeaways
- FGF21 analogs significantly improve resolution with a relative risk (RR) of 4.84, making them the most effective treatment in this regard.
- Resmetirom achieves the greatest reduction in liver fat, with a mean difference (MD) of -18.41 in MRI-PDFF, indicating its strong efficacy in fat reduction.
- Adverse events are more frequent with Resmetirom, which has an RR of 1.47 for increased risk, highlighting safety concerns despite its efficacy.
Caveats
- Variability in study populations and methodologies introduces heterogeneity, complicating direct comparisons of treatment effects.
- The analysis relies on published data, which may lead to publication bias and limits detailed subgroup analyses.
- Included trials used NAFLD diagnostic criteria, while emphasizes metabolic dysfunction, potentially affecting generalizability.
Definitions
- MASLD: A spectrum of liver conditions linked to metabolic dysfunction, ranging from simple steatosis to steatohepatitis.
- MASH: Metabolic Dysfunction-Associated Steatohepatitis, characterized by liver inflammation and damage due to fat accumulation.
AI simplified
1. Introduction
Metabolic Dysfunction-Associated Steatotic Liver Disease (MASLD) and Metabolic Dysfunction-Associated Steatohepatitis (MASH) are increasingly recognized as significant global health concerns, affecting millions of individuals worldwide [1]. MASLD, which encompasses a spectrum of liver conditions ranging from simple steatosis to MASH, is closely associated with obesity, type 2 diabetes, and metabolic syndrome. These conditions not only increase the risk of liver-related morbidity and mortality but are also associated with cardiovascular diseases and other metabolic complications, significantly burdening healthcare systems [2]. The economic impact is substantial, with direct medical costs attributed to MASLD and MASH estimated to be in the billions of dollars annually in the United States alone. Moreover, patients with MASLD and MASH often face higher healthcare utilization, including increased hospitalizations and outpatient visits, further straining healthcare resources [3].
Recent advancements in the pharmacological treatment of MASLD and MASH have shown promising results, particularly with medications such as Resmetirom and Fibroblast Growth Factor 21 (FGF21) analogs. Resmetirom, a selective thyroid hormone receptor-β agonist, enhances hepatic fat metabolism and significantly reduces liver lipid content, making it a potent agent for liver fat reduction and improved liver function [4]. On the other hand, FGF21 analogs, such as Pegbelfermin, Efruxifermin, and Pegozafermin, have demonstrated strong anti-inflammatory and insulin-sensitizing effects in the liver [5]. FGF21 enhances fatty acid oxidation, reduces hepatic lipid accumulation, and improves insulin sensitivity by activating the FGFR1c and β-Klotho (KLB) receptor complex. This activation increases energy expenditure and reduces lipotoxicity, alleviating hepatic steatosis and improving liver function [5]. Comparatively, Glucagon-like Peptide-1 (GLP-1) agonists, such as liraglutide and semaglutide, primarily work by enhancing insulin secretion, inhibiting glucagon release, promoting weight loss, and improving lipid levels. These mechanisms are beneficial for reducing liver fat and improving overall metabolic health [6]. The distinct mechanisms and therapeutic effects of these agents highlight the evolving landscape of treatment options for liver diseases. Distinctive mechanisms of action are depicted in Figure 1.
Conducting a meta-analysis on the efficacy and safety of novel therapeutic agents like Resmetirom, FGF21 analogs, and GLP-1 agonists for MASLD and MASH is essential for offering a comparative evaluation among these therapeutic options. Despite numerous clinical trials, comprehensive comparative analyses that offer clear guidance for clinical practice are limited. Our meta-analysis aims to address this gap by systematically evaluating available evidence on these therapies, providing insights into their relative benefits and risks. This analysis will aid clinicians in making informed decisions tailored to individual patient profiles, ultimately improving the management of MASLD and MASH.
2. Methods
The study protocol is registered on the Open Science Framework (OSF) at (https://osf.io/hxvp3↗, accessed on 15 June 2024) [7]. The results of this meta-analysis adhere to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [8]. The PRISMA checklist was followed throughout the study.
2.1. Search Strategy
A comprehensive literature search was conducted across PubMed, Scopus, Web of Science, and the Cochrane Library to identify relevant studies from inception to 8 June 2024. The search terms included: ("Resmetirom" OR "MGL-3196" OR "Thyroid hormone receptor-β agonist") OR ("FGF21 analogs" OR "Pegozafermin" OR "BIO89-100" OR "Pegbelfermin" OR "BMS-986036" OR "Efruxifermin" OR "AKR-001") OR ("GLP-1 agonists" OR "liraglutide" OR "semaglutide" OR "dulaglutide") AND ("MASLD" OR "MASH" OR "NASH" OR "MAFLD" OR "NAFLD") AND ("clinical trials" OR "randomized controlled trials" OR "RCTs" OR "trial"). The detailed search strategy is reported in. Supplement S1
2.2. Screening Process
Two authors (HA and SS) independently screened the titles and abstracts of retrieved studies. Full-text articles were reviewed for eligibility. Discrepancies were resolved through discussion or by consulting a third author (SA) if necessary.
2.3. Study Selection
We included clinical trials addressing MASLD or MASH investigating Resmetirom, FGF21 analogs (Pegozafermin, Pegbelfermin, Efruxifermin), GLP-1 agonists, or combinations thereof, reporting outcomes related to liver histology, liver fat reduction, and biochemical markers. Studies were excluded if they focused on pediatric populations, did not report relevant outcomes, or were non-comparative studies or conference abstracts without full-text availability. Post-hoc analysis and extension studies were excluded.
2.4. Data Extraction
Data extracted from each study included participant demographics, baseline characteristics, intervention details, and outcome measures. Specifically, data on age, sex, body mass index (BMI), diabetes duration, liver enzyme levels, and liver fat content were collected. Information on study design, duration, and adverse events was also extracted. For studies reporting medians and interquartile ranges, these were converted to means and standard deviations using the method by Wan et al. [9]. Combined means and standard deviations were calculated following the Cochrane Handbook for Systematic Reviews of Interventions guidelines [10].
2.5. Statistical Analysis
We implemented a network meta-analysis to compare multiple treatments by analyzing data from various studies, allowing for both direct and indirect comparisons. This method synthesizes evidence to determine the relative effectiveness of each treatment, even if some treatments were not directly compared in any individual study [11]. Statistical analyses were conducted using a random-effects model to account for heterogeneity among studies. Mean differences (MD) for continuous outcomes and relative risks (RR) for dichotomous outcomes were calculated. We used standardized mean difference (SMD) for outcomes with different measurement methods. Heterogeneity was evaluated using τ2 (tau-squared), I2 statistics, and Q statistics [12]. All analyses were performed using RStudio with the meta and netmeta packages [13,14]. P-scores were used to rank treatments based on their estimated effect sizes, with higher P-scores associated with better outcomes [15]. The outcomes evaluated include the following: MASH resolution is defined as the complete absence of ballooning and mild or no inflammation without worsening fibrosis. Improvement in fibrosis refers to a reduction of at least one stage in fibrosis severity without worsening MASH. MRI-PDFF (Magnetic Resonance Imaging-Proton Density Fat Fraction) measures the alteration in liver fat content over the study period, with a specific evaluation for more than 30% fat reduction. VCTE (Vibration-Controlled Transient Elastography) assesses changes in liver stiffness, a surrogate marker for fibrosis. Changes in liver enzyme levels include ALT (Alanine Aminotransferase), AST (Aspartate Aminotransferase), and GGT (Gamma-Glutamyl Transferase), which are indicators of liver injury. Safety outcomes include adverse events, treatment discontinuation due to TEAEs (treatment-emergent adverse events), and the incidence of nausea and vomiting among participants. Sensitivity analysis and meta-regression were performed using different statistical techniques.
2.6. Bias Assessment and Certainty of Evidence
The risk of bias was assessed using the RoB 2 (Risk of Bias 2) tool, which evaluates bias across five domains: randomization, deviations from intended interventions, missing outcome data, outcome measurement, and selection of reported results [16]. Two authors (HA and SS) independently assessed the risk of bias. Disagreements were resolved through discussion or by consulting a third author (SA). The certainty of evidence was assessed using the Confidence in Network Meta-Analysis (CINeMA) framework [17].
3. Results
3.1. Study Characteristics
A total of 16 studies (Table 1) with a total of 3535 participants were included in the analysis [4,18,19,20,21,22,23,24,25,26,27,28,29,30,31,32] (Figure 2) (Supplement S1). The network for the included interventions is presented in Figure 3. The mean age of participants was 55.60 years (SD 11.35), with a mean BMI of 35.62 kg/m2 (SD 6.43). Liver function tests showed a mean ALT level of 42.76 U/L (SD 44.18) and a mean AST level of 33.09 U/L (SD 37.28). Glycemic control, measured by HbA1c, had a mean of 6.55% (SD 1.12), with 48.6% of participants being male (Supplement S2 and S3); baseline characteristics are summarized in Table 2. The risk of bias was generally low, with some domains having some concerns (Supplement S4). Publication bias was generally low except for VCTE, which showed a high risk of bias (Supplement S6). The certainty of evidence was generally low due to concerns of incoherence and imprecision (Supplement S8).
3.2. Sensitivity Analysis and Meta-Regression
To ensure the robustness of our findings and validate the transitivity assumption, we conducted a meta-regression analysis. Sensitivity analysis was further performed using a Bayesian approach to strengthen the reliability of our results. In our meta-regression, we employed the Bayesian method, incorporating key variables such as HbA1c, age, and gender (male) for a comprehensive subgroup analysis. The Bayesian analysis revealed consistent treatment rankings across all outcomes, underscoring the reliability and stability of the results. Additionally, the subgroup analysis confirmed the uniformity of treatment rankings across different strata, as detailed below (). Supplement S10
3.3. Biopsy Outcomes
3.3.1. NASH Resolution
In the random effects model, the relative risk (RR) of NASH resolution compared to placebo was significantly higher for all treatments (Figure 4): FGF21 (RR 4.84, 95% CI: 2.59 to 9.03, p < 0.0001), GLP-1 agonists (RR 2.48, 95% CI: 1.30 to 4.72, p = 0.006), and Resmetirom (RR 3.06, 95% CI: 1.91 to 4.91, p < 0.0001). The heterogeneity analysis showed an I2 of 11.70%, indicating low to moderate heterogeneity. The tests for heterogeneity within designs (Q = 6.79, df = 6, p = 0.34) and between designs (Q = 0.00, df = 0) were not significant, suggesting consistent treatment effects across studies. P-scores for ranking the treatments were as follows: FGF21 (0.93), Resmetirom (0.61), and GLP-1 (0.46), indicating that FGF21 agonists had the highest probability of being the most effective treatment for NASH resolution. There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.3.2. Improvement in Fibrosis
In the random effects model assessing the improvement in fibrosis, FGF21 demonstrated a significant relative risk (RR) of 2.47 (95% CI: 1.35 to 4.53, p = 0.003), which was statistically significant in comparison to placebo (Figure 4). GLP-1, however, showed no significant effect with an RR of 0.99 (95% CI: 0.43 to 2.26, p = 0.97), and Resmetirom had an RR of 1.67 (95% CI: 0.79 to 3.52, p = 0.18), also not reaching significance. The heterogeneity analysis revealed moderate to substantial heterogeneity with I2 of 62.50%. The tests for heterogeneity within designs were significant (Q = 18.65, df = 7, p = 0.009), suggesting variability in the effect estimates. P-scores, which rank treatments based on their effectiveness, were highest for FGF21 (0.92), followed by Resmetirom (0.65) and GLP-1 (0.23), indicating FGF21 analogs as the most effective treatment for fibrosis improvement among the interventions studied. There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.4. Imaging Outcomes
3.4.1. Change in MRI-PDFF
In the random effects model assessing the change in MRI-PDFF, Resmetirom demonstrated the most significant effect with a mean difference (MD) of −18.41 (95% CI: −23.60 to −13.22, p < 0.0001) compared to placebo, indicating a substantial reduction in liver fat content (Figure 4). FGF21 also showed a significant effect with an MD of −8.38 (95% CI: −11.93 to −4.84, p < 0.0001), followed by GLP-1 with an MD of −4.99 (95% CI: −8.72 to −1.25, p = 0.009). The heterogeneity analysis revealed significant variability across studies, with an I2 of 89.00%. Tests of heterogeneity within designs were significant (Q = 72.44, df = 8, p < 0.0001), indicating considerable inconsistency among the study results. The P-scores, which rank treatments based on their effectiveness, were highest for Resmetirom (1.00), followed by FGF21 (0.63) and GLP-1 (0.36). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.4.2. >30% Fat Reduction on MRI-PDFF
In the random effects model assessing a >30% reduction in fat on MRI-PDFF, Resmetirom showed the most significant effect with a relative risk (RR) of 3.56 (95% CI: 2.41 to 5.26, p < 0.0001) compared to placebo (Figure 4). FGF21 also demonstrated a significant effect with an RR of 2.93 (95% CI: 2.00 to 4.30, p < 0.0001), followed by GLP-1 with an RR of 1.83 (95% CI: 1.16 to 2.90, p = 0.010). The heterogeneity analysis revealed moderate heterogeneity, with an I2 of 15.20%. The tests for heterogeneity within designs (Q = 10.62, df = 9, p = 0.30) were not significant, suggesting consistency among the study results. The P-scores, which rank treatments based on their effectiveness, were highest for Resmetirom (0.91), followed by FGF21 (0.73) and GLP-1 (0.36). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.4.3. Change in VCTE
In the random effects model assessing change in VCTE, FGF21 demonstrated a significant standardized mean difference (SMD) of −1.71 (95% CI: −3.11 to −0.31, p = 0.017) compared to placebo, while Resmetirom showed a borderline significant effect with an SMD of −1.68 (95% CI: −3.38 to 0.02, p = 0.053) (Figure 4). GLP-1 agonists did not show a significant effect, with an SMD of −0.67 (95% CI: −2.02 to 0.68, p = 0.33). The heterogeneity analysis indicated substantial variability with an I2 of 73.10%. Tests of heterogeneity within designs were significant (Q = 26.03, df = 7, p = 0.0005), indicating inconsistency among the study results. P-scores ranked FGF21 (0.78) and Resmetirom (0.76) as the most effective treatments for reducing liver stiffness measured by VCTE, followed by GLP-1 (0.39).
3.5. Biochemical Markers
3.5.1. Change in ALT
In the random effects model assessing the change in ALT, Resmetirom demonstrated the most significant effect with a mean difference (MD) of −15.71 (95% CI: −23.30 to −8.13, p < 0.0001) compared to placebo (Figure 5). FGF21 also showed a significant reduction in ALT with an MD of −13.32 (95% CI: −18.49 to −8.15, p < 0.0001), followed by GLP-1 with an MD of −10.30 (95% CI: −16.24 to −4.36, p = 0.0007). The heterogeneity analysis revealed moderate heterogeneity with an I2 of 55.80%. Tests of heterogeneity within designs were significant (Q = 22.61, df = 10, p = 0.012), indicating variability among the study results. The P-scores, which rank treatments based on their effectiveness, were highest for Resmetirom (0.85), followed by FGF21 (0.69) and GLP-1 (0.45). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.5.2. Change in AST
In the random effects model assessing the change in AST, Resmetirom demonstrated the most significant reduction with a mean difference (MD) of −12.28 (95% CI: −21.07 to −3.49, p = 0.006) compared to placebo (Figure 5). GLP-1 also showed a significant reduction with an MD of −8.71 (95% CI: −14.73 to −2.68, p = 0.005), followed by FGF21 with an MD of −7.91 (95% CI: −13.79 to −2.02, p = 0.009). The heterogeneity analysis revealed substantial variability across studies, with an I2 of 77.80%. Tests of heterogeneity within designs were significant (Q = 45.07, df = 10, p < 0.0001), indicating inconsistency among the study results. The P-scores, which rank treatments based on their effectiveness, were highest for Resmetirom (0.84), followed by GLP-1 (0.61) and FGF21 (0.54). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.5.3. Change in GGT
In the random effects model assessing the change in GGT, Resmetirom demonstrated a significant reduction with a mean difference (MD) of −19.56 (95% CI: −34.68 to −4.44, p = 0.011) compared to placebo (Figure 5). GLP-1 also showed a significant reduction with an MD of −18.73 (95% CI: −30.55 to −6.91, p = 0.002), while FGF21 did not show a significant effect with an MD of −11.44 (95% CI: −29.22 to 6.34, p = 0.207). The heterogeneity analysis revealed substantial variability across studies, with an I2 of 75.50%. Tests of heterogeneity within designs were significant (Q = 20.41, df = 5, p = 0.001), indicating inconsistency among the study results. The P-scores, which rank treatments based on their effectiveness, were highest for Resmetirom (0.76), followed closely by GLP-1 (0.74), then FGF21 (0.47). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.6. Safety Outcomes
3.6.1. Adverse Events
In the random effects model assessing adverse events, Resmetirom showed a significant increase in the risk of adverse events with a relative risk (RR) of 1.47 (95% CI: 1.24 to 1.74, p < 0.0001) compared to placebo (Figure 6). GLP-1 agonists did not show a significant effect, with an RR of 1.23 (95% CI: 0.70 to 2.17, p = 0.474), indicating a non-significant trend towards increased risk. FGF21 also showed a non-significant increase in the risk of adverse events with an RR of 1.22 (95% CI: 0.89 to 1.67, p = 0.220). The heterogeneity analysis revealed no significant heterogeneity, with I2 = 0%. Tests for heterogeneity within designs were not significant (Q = 2.29, df = 10, p = 0.994), indicating consistency among the study results. The P-scores, which rank treatments based on their effectiveness with higher scores indicating more favorable outcomes, were highest for FGF21 (0.4892), followed by GLP-1 (0.4821), and Resmetirom (0.1445). There was no difference in the subgroups (Supplement S10), indicating the robustness of the results.
3.6.2. Treatment Discontinuation
In the random effects model assessing treatment discontinuation, Resmetirom had a significant increase in the risk of discontinuation with a relative risk (RR) of 1.71 (95% CI: 1.08 to 2.71, p = 0.022) compared to placebo (Figure 6). GLP-1 agonists did not show a significant effect with an RR of 1.84 (95% CI: 0.82 to 4.13, p = 0.142), although there was a trend towards increased risk. FGF21 showed a non-significant increase in the risk of discontinuation with an RR of 2.19 (95% CI: 0.98 to 4.91, p = 0.058). The heterogeneity analysis revealed no significant heterogeneity, with I2 = 0%. Tests for heterogeneity within designs were not significant (Q = 3.20, df = 10, p = 0.976), indicating consistency among the study results. The P-scores, which rank treatments based on their effectiveness with higher scores indicating more favorable outcomes, were highest for Resmetirom (0.42), followed by GLP-1 (0.38), and FGF21 (0.24). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
3.6.3. Nausea
In the random effects model assessing adverse events, GLP-1 agonists showed the highest significant increase in risk with a relative risk (RR) of 2.48 (95% CI: 1.35 to 4.57, p = 0.0035) compared to placebo (Figure 6). Resmetirom followed with a significant increase in risk, presenting an RR of 1.77 (95% CI: 1.08 to 2.88, p = 0.0223). FGF21 exhibited a non-significant increase in the risk of nausea with an RR of 1.38 (95% CI: 0.83 to 2.30, p = 0.2130). The heterogeneity analysis indicated moderate heterogeneity with tau^2 = 0.0957 and I^2 = 39.4%. Tests for heterogeneity within designs were not significant (Q = 14.86, df = 9, p = 0.0949), suggesting consistency among the study results. The P-scores, which rank treatments based on their likelihood of causing adverse events, were highest for FGF21 (0.5955), followed by Resmetirom (0.3532) and GLP-1 (0.0911). This ranking indicates that GLP-1 had the highest risk of nausea, followed by Resmetirom, with FGF21 showing the least increase in risk compared to placebo. There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results. However, it showed a significantly higher incidence in those over 55 years old across all treatment groups.
3.6.4. Diarrhea
In the random effects model assessing the incidence of diarrhea, Resmetirom demonstrated a significant increase in the risk with a relative risk (RR) of 1.96 (95% CI: 1.61 to 2.38, p < 0.0001) compared to placebo (Figure 6). FGF21 also showed a significant increase in the risk of diarrhea with an RR of 1.89 (95% CI: 1.23 to 2.89, p = 0.004), followed by GLP-1 with an RR of 1.77 (95% CI: 1.08 to 2.91, p = 0.024). The heterogeneity analysis revealed no significant heterogeneity, with I2 = 0%. Tests for heterogeneity within designs were not significant (Q = 6.76, df = 10, p = 0.7481), indicating consistency among the study results. The P-scores, which rank treatments based on their likelihood of causing diarrhea, were highest for GLP-1 (0.41), followed by Resmetirom (0.27), and FGF21 (0.20). There was no difference in the subgroups (Supplement S10) in the Bayesian analysis, indicating the robustness of the results.
4. Discussion
The findings of this network meta-analysis reveal significant clinical insights into the efficacy of FGF21 analogs, GLP-1 agonists, and Resmetirom for treating liver conditions. FGF21 analogs showed the highest effectiveness in terms of MASH resolution and fibrosis improvement. Resmetirom also demonstrated substantial efficacy, particularly in reducing liver fat content and improving ALT and AST levels, highlighting its potential in managing liver fat and inflammation. GLP-1 agonists, while effective in reducing liver fat and improving AST levels, were less impactful on fibrosis. However, both Resmetirom and GLP-1 agonists showed significant reductions in ALT and GGT levels. Adverse events analysis indicated that Resmetirom and FGF21 analogs were associated with higher risks of nausea and diarrhea.
To our knowledge, this study is the first to provide a practical comparison of FGF21 analogs, Resmetirom, and GLP-1 agonist treatments for clinical practice, highlighting the significance of the efficacy and safety data. Our analysis provides a detailed assessment of different outcomes, including fibrosis improvement, liver fat reduction, and changes in biochemical markers such as ALT, AST, and GGT. By providing information about adverse events, treatment discontinuation due to adverse events, and some highlights of the common adverse events, these findings underscore the importance of considering adverse event profiles and treatment discontinuation rates when selecting a therapeutic regimen.
The clinical outcomes observed are closely linked to the mechanisms of action of Resmetirom, FGF21 analogs, and GLP-1 agonists. Resmetirom, a selective thyroid hormone receptor-β agonist, accelerates hepatic fat metabolism, demonstrating substantial efficacy in reducing liver fat (MRI-PDFF) and improving liver enzymes (ALT, AST). This highlights Resmetirom's effectiveness in targeting liver fat accumulation and enhancing liver function [4]. FGF21 analogs' anti-inflammatory properties help reduce liver inflammation and fibrosis [5], which could explain their effectiveness in fibrosis improvement and MASH resolution. GLP-1 agonists primarily work by enhancing insulin secretion, inhibiting glucagon release, promoting weight loss, and improving lipid profiles [6]. This explains their effectiveness in reducing liver fat and improving liver enzymes, although their impact on fibrosis is less pronounced. The adverse events associated with these medications, such as nausea and diarrhea, are consistent with their metabolic and gastrointestinal effects, reflecting their broad systemic actions.
Furthermore, this study highlights the need for long-term follow-up and real-world evidence to validate the clinical benefits and safety profiles of these treatments. The observed heterogeneity across studies emphasizes the importance of individualized patient care, as treatment responses may vary due to genetic, metabolic, and lifestyle factors. Additionally, the high costs of therapies, for example, the annual cost of Resmetirom being around $47,700 [33], necessitate cost-effectiveness analyses to determine their value in routine practice. Future research should focus on head-to-head comparisons, combination therapy strategies, and integrating these agents into clinical guidelines for MASLD and MASH management. These treatments' potential to improve liver health and address comorbid conditions like obesity and diabetes supports their inclusion in a comprehensive treatment approach.
At this time, GLP-1 agonists and FGF21 analogs are not approved for the treatment of MASH/MASLD. These agents are still under investigation in ongoing clinical trials, though early results are promising. For example, the ESSENCE trial is a Phase 3 study evaluating the use of semaglutide, a GLP-1 receptor agonist, in approximately 1200 patients with NASH. The primary endpoint of this study is to assess the efficacy of a weekly 2.4 mg dose of semaglutide in improving NASH without worsening fibrosis after 72 weeks. Pegozafermin, an FGF21 analog, is being evaluated in the ENLIGHTEN program. This includes the ENLIGHTEN-Fibrosis trial, which targets patients with F2-F3 fibrosis, and the ENLIGHTEN-Cirrhosis trial, aimed at patients with compensated cirrhosis (F4). Both trials have co-primary endpoints focused on fibrosis improvement and MASH resolution, aiming for future regulatory approval [34,35].
This study, while comprehensive, faces several limitations. Key challenges include variability across studies in populations, interventions, and outcome measures, introducing heterogeneity and complicating direct comparisons. Differences in the quality and methodological rigor of the studies potentially affect the robustness of pooled estimates. Reliance on published data may lead to publication bias, and the lack of individual patient data restricts detailed subgroup analyses. Variations in study designs, follow-up durations, and outcome definitions add complexity, particularly given the low certainty of evidence for most comparisons. Additionally, it must be noted that this analysis compares the only approved drug for this condition against two other classes of agents, GLP-1 agonists and FGF21 analogs, which are still under investigation. This inherently adds heterogeneity to the comparisons and assumes a similar effect across all studied GLP-1 drugs and FGF21 agents, which may not fully capture the nuances in their individual efficacy and safety profiles. Furthermore, a key limitation is that the included trials were conducted using the diagnostic criteria for NAFLD, whereas the newer classification, MASLD, emphasizes metabolic dysfunction more clearly. While MASLD reclassifies NAFLD with a focus on metabolic drivers, the therapeutic interventions studied, such as GLP-1 agonists and Resmetirom, target shared underlying mechanisms, including insulin resistance, lipid metabolism, and inflammation. Therefore, the results and conclusions from these NAFLD-based trials remain relevant to MASLD populations. However, the shift in diagnostic criteria could influence patient selection in future studies, potentially affecting generalizability. Further research using MASLD-specific criteria may provide more precise insights into this population.