Page 343 - Read Online

Liu et al. Novel predictive and prognostic strategies of HBV-related HCC

oxide synthase. [40] Activation of Wnt/β-catenin pathway was usually evaluated in cohort studies. High risk
contributes to HCC development. The hallmarks patients that were identified through cluster analysis
of Wnt/β-catenin pathway, Wnt-1 and Wnt3a, have or score model based on gene signatures were prone
both predictive and prognostic value. [37,41,42] Likewise, to have unfavourable clinical outcomes, such as poor
signaling pathways such as phosphatidylinositol-3 overall survival and early recurrence.
kinase (PI3K)/protein kinase B (AKT)/mammalian
target of rapamycin (mTOR) pathway, and insulin-like Although the tumor gene signatures were identified
growth factor pathway also play an important role in by different studies with various comparison
hepatocarcinogenesis. [43] strategies, they shared some genes conferring cancer
stemness. For instance, a group of genes related to
Genetic polymorphisms of immune/inflammatory proliferation and epithelial cell adhesion molecule
molecules (EpCAM)-positive phenotype were included in 8 gene
Genetic polymorphisms of immune/inflammatory signatures summarized in different studies and all
molecules can also serve as predictive biomarkers associated with poor prognosis. [48] Gene signatures
for HCC development. For example, genetic from adjacent non-tumor tissues were also reported
polymorphisms of signal transducer and activator to be significantly associated with HCC recurrence,
of transcription 3 (STAT3), class II human leukocyte indicating that the histological “normal” adjacent tissue
antigen DP (HLA-DP), HLA-DQ, miRNA-122-binding may be at the early stage of cancer evolution. That
site, pre-miR-218, nuclear factor-kappaB (NF-κB), and highlights the need of biopsy-based gene signature
its inhibitor IkappaBalpha are significantly associated detection for specific individuals, like HBV-infected
with HCC risk. [12,17,18,44-47] patients. However, signatures from adjacent tissues
obtained in different studies are lack of genes in
IDENTIFYING SIGNATURES OF SIGNALING common. Cross validations are needed to consolidate
PATHWAY ALTERATION the criteria. Altered expression patterns of the genes
in HCC are usually caused by epigenetic modifications
Gene signatures in their regulatory elements and somatic mutations of
The alteration of signaling pathways confers stemness their repressors.
characteristics and competitive advantages to cancer
cells. These alterations usually affect complex Somatic mutation profiles
signaling networks that cannot be represented by Somatic mutations are genetic basis of carcinogenesis.
a signal gene. More than 300 published microarray The values of somatic mutations depend on their
studies of human HCC samples provide sufficient impacts on related signaling pathways. By changing
information regarding tumor gene expression profiles. patterns of signaling transduction, somatic mutations
[48]
The accumulation of data regarding differentially on a small proportion of genes can promote cancer
[19]
expressing genes makes it possible to conduct meta- evolution, which are categorized as “driver mutations”.
analysis and subsequently determine gene signatures. As a matter of fact, some outstanding somatic
Recent gene signature studies are summarized in mutations in HBV-HCC occur in the genes responsible
Table 1. [49-66] Gene signatures developed in those for epigenetic modifications-chromatin remodeling
studies were used to separate patients into 2 or including ARID1A and ARID2 and methylation such
more subgroups with different clinical outcomes, MLL4. [67,68] Due to survival competition and the
phenotypes, and altered signaling pathways. The positive selection of inflammatory microenvironment,
methods of developing gene signatures fall into two driver mutations accumulate sufficiently to promote
major groups. The first group of gene signatures malignant transformation of hepatocytes.
was generated in case-control studies with the data
of training cohort or published gene expression The distribution, combination, and dynamic patterns
data. Most of the gene signature studies belong of driver mutations reflex the pressure of
to this group. [50,52,53,55,57,59,61-65] The second group of microenvironmental selection and growth advantage
gene signatures concerning defined phenotypes or of hepatocyte subsets. The high frequent mutations
signaling pathways was derived from the data of cell can have clinical values as biomarkers for targeted
or animal model studies. [49,51,56,58,60] For examples, therapy, classification, and prognostic prediction. [67-71]
Lee et al. [49] developed a gene signature of stemness For instance, homozygous deletions were detected in
from the gene profiling data of rat fetal liver tissue and 40% of HCC patients and were significantly associated
Kaposi-Novak et al. [51] developed a gene signature of with poor survival (P < 0.0001). [68]
Met signaling pathway using the Met deficient mouse
model. The predictive value of novel gene signatures Using next generation sequencing technology, some
334 Hepatoma Research ¦ Volume 2 ¦ December 23, 2016