Gene Signatures from a Liver Cancer Mouse Model Predict Survival in Patients
Author Information
Author(s): Ivanovska Irena, Zhang Chunsheng, Liu Angela M., Wong Kwong F., Lee Nikki P., Lewis Patrick, Philippar Ulrike, Bansal Dimple, Buser Carolyn, Scott Martin, Mao Mao, Poon Ronnie T. P., Fan Sheung Tat, Cleary Michele A., Luk John M., Dai Hongyue
Primary Institution: Merck & Co., Inc.
Hypothesis
The study aims to analyze the molecular signatures of liver cancer in a c-MET-transgenic mouse model and investigate its prognostic relevance to human hepatocellular carcinoma (HCC).
Conclusion
The study provides evidence that a disease model can serve as a possible platform for generating hypotheses to be tested in human tissues and highlights an efficient method for generating biomarker signatures before extensive clinical trials.
Supporting Evidence
- The study identified gene signatures that were down-regulated in both mouse tumors and human HCC, which had significant predictive power on overall and disease-free survival.
- Mouse tumors showed parallels with human liver tumors, including down-regulation of metabolic pathways.
- The predictive power of the mouse-derived signatures stems from their tumor properties rather than c-MET-driven properties.
Takeaway
Scientists studied mice with liver cancer to find patterns that could help predict how long patients with liver cancer might live.
Methodology
The study involved molecular profiling of liver tissues from a c-MET-driven mouse model and human HCC samples, using whole genome microarray expression profiling.
Potential Biases
Potential bias due to the reliance on a single mouse model and the limited number of human samples.
Limitations
The predictive power of the mouse-derived signatures may not fully represent human disease due to differences in tumor initiation mechanisms.
Participant Demographics
272 HBV-associated and 9 HCV-associated HCC patients.
Statistical Information
P-Value
6.3×10−6
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website