MASH Mouse Models 2026: GAN vs CDA-HFD vs STAM Comparison
FAQ on choosing the right MASH preclinical model (GAN Diet, CDA-HFD, STAM). Align models with drug Mechanism of Action (MoA) to avoid clinical failures.
The recent shift in nomenclature (from NAFLD to MASLD and NASH to MASH) reflects a deeper understanding of the disease's metabolic roots. This shift also redefines how we evaluate preclinical efficacy.
Because no single animal model can perfectly replicate the complex human pathology, evaluating compounds across multiple, complementary models based on Mechanism of Action (MoA) is essential. (For instance, the development of Semaglutide famously utilized both the GAN model for metabolic evaluation and the CDA-HFD model for fibrosis pathway validation). We have restructured our guide into an FAQ format to help you combine and select the most clinically relevant models.
FAQ: Understanding MASH Preclinical Failures
Q1: Why do so many MASH drugs work in mice but fail in humans?
A: The primary reason is a lack of External Validity. Historically, the industry favored models that induced severe fibrosis rapidly, often through extreme nutrient deprivation (e.g., choline deficiency). While these animals develop fibrotic livers, they achieve this purely through acute dietary stress. They completely lack the complex metabolic background—overnutrition, clinical obesity, and insulin resistance—that characterizes true human MASH patients. Evaluating experimental metabolic drugs in models without "Metabolic Dysfunction" severely limits clinical predictability.
FAQ: Comparing the "Big 3" MASH Models
Q2: What is the "GAN Diet" model, and when should I use it?
A: The GAN Diet (the trans-fat-free successor to the legacy AMLN diet) represents the "Gold Standard" for systemic obesity and metabolic MASH.
- How it works: Mice fed a Western diet naturally develop obesity, severe insulin resistance, and MASH pathology. Given enough time, they also progress to Liver Cancer (HCC). However, induction takes a long time (12-24 wks) and has high inter-individual variability. Therefore, researchers often must perform an intermediate liver biopsy to screen and select mice with established pathology before starting treatment.
- When to use it: When your drug is a Metabolic Modifier (e.g., GLP-1 RAs, GIP/GLP-1 dual agonists, THR-β agonists). If your MoA aims to improve liver pathology by correcting systemic metabolism, the GAN model is absolutely essential.
Q3: If my drug targets fibrosis directly, should I still use the GAN diet?
A: Not necessarily. If you need to quickly evaluate a direct anti-fibrotic effect, the CDA-HFD (Choline-Deficient, L-Amino Acid-defined, HFD) model is often better.
- How it works: It acts as a "Fibrosis Accelerator," rapidly developing fibrosis due to VLDL secretion blockage. Critically, mice experience weight loss and lack comprehensive metabolic syndrome. Therefore, while insufficient for metabolic evaluation on its own, it is frequently combined with other metabolic models (like GAN) to specifically evaluate a drug's purely anti-fibrotic efficacy.
- When to use it: For Anti-fibrotics & Anti-inflammatories (e.g., TGF-β antagonists). Warning: This model lacks the human metabolic profile and often causes paradoxical peripheral insulin sensitivity. Use it only when screening drugs that target fibrosis pathways independently of metabolism.
Q4: What is the STAM™ model, and what makes it unique?
A: The STAM™ Model is uniquely positioned for studies focusing on late-stage disease progression and HCC.
- How it works: Neonatal STZ injection plus HFD creates a phenotype of impaired insulin secretion. Importantly, it does not induce massive obesity like the GAN diet. While unsuitable for evaluating typical obese MASH, it serves as a T2DM-driven non-obese MASH/HCC progression model, reliably progressing from steatosis to fibrosis, and eventually HCC (note: it does not represent the full spectrum of human Lean MASH, which is defined by BMI criteria across a heterogeneous population).
- When to use it: When testing drugs aimed at halting the irreversible progression to HCC, or when developing therapies specifically intended for patient subgroups with severe, insulin-depleted diabetes.
For researchers tracking fibrosis & inflammation R&D
FDA approval alerts, trial readouts, preclinical model selection, and assay optimization — curated signal for bench-to-pipeline readers. 2 emails/month max.
FAQ: Practical Model Selection
Q5: How do I quickly decide which model fits my project?
A: Ask yourself what your primary clinical mechanism is:
- "I want to improve the liver by addressing systemic metabolism (weight loss, insulin resistance)."
$\to$ Choose the 【GAN DIO Model】. - "I am ignoring metabolism. I need to powerfully and directly stop liver fibrosis progression."
$\to$ Choose the 【CDA-HFD Model】. - "I want to stop the progression from MASH to liver cancer (HCC)."
$\to$ Choose the 【STAM™ Model】.
Q6: Can I see a side-by-side comparison of the models?
A: Yes. The table below summarizes the critical differences:
| Feature | GAN Diet (AMLN) | CDA-HFD | STAM™ Model |
|---|---|---|---|
| Concept | Systemic Obese MASH | Fibrosis Accelerator | T2DM-driven Non-obese MASH / HCC |
| Induction | Western Diet Mimic | Nutrient Deficiency (Choline) | Beta-cell destruction + HFD |
| Metabolic Profile | Severe Insulin Resistance | None or Paradoxical Sensitivity | Reduced Insulin Secretion |
| Pathology | Mild-Mod Fibrosis, Ballooning | Severe Fibrosis (Rapid) | Steatosis → Fibrosis → HCC |
| Clinical Relevance | High (Metabolically identical) | Limited (Only matches fibrotic phenotype) | Relatively High (T2D base + HFD) |
| Recommended MoA | GLP-1 RA, THR-β, GIP/GLP-1 | TGF-β inhibitors, ASK1 inhibitors | Anti-carcinogenic, Antioxidant |
| Timeline (MASH Phase) | ~12-24 weeks | ~6-12 weeks | ~9 weeks |
Q7: Is there a more detailed guide for selecting models based on my drug's MoA?
A: Yes. While this article focuses on the "Big 3," real-world drug development often requires evaluating additional models such as CCl4, TAA, or Western Diet + CCl4 combinations. We have prepared a comprehensive MoA-based selection matrix:
👉 MoA-Based MASH Model Selection Guide: Choosing the Optimal Model by Drug Mechanism
Q8: We want to outsource these studies. What is the next step?
A: Setting up advanced models like GAN DIO or STAM from scratch requires massive investments in time, diet procurement, housing optimization, and standardizing pathological scoring systems. Fibrosis evaluation also demands specialized pathological analysis including Sirius Red staining and Hydroxyproline quantification, which require considerable expertise to standardize.
Experienced CROs can provide end-to-end support—from MoA-aligned model selection to pathological scoring and AI-powered image analysis. Contact us to discuss your project requirements.
References & Clinical Linkages
1. Clapper JR, et al. Diet-induced mouse model of fatty liver disease and nonalcoholic steatohepatitis reflecting clinical disease progression and methods of assessment. Am J Physiol Gastrointest Liver Physiol. 2013;305(7):G483-G495. PMID 23886860 (GAN/AMLN Diet primary reference)
2. Matsumoto M, et al. An improved mouse model that rapidly develops fibrosis in non-alcoholic steatohepatitis. Int J Exp Pathol. 2013;94(2):93-103. (CDA-HFD)
3. Fujii M, et al. A murine model for non-alcoholic steatohepatitis showing evidence of association between diabetes and hepatocellular carcinoma. Med Mol Morphol. 2013;46(3):141-152. (STAM)