With limited treatment options for liver cancer, a nationwide team of genomics researchers sought to gain a better understanding of the causes of this disease while identifying potential targets for therapy (Cell 2017; doi: http://dx.doi.org/10.1016/j.cell.2017.05.046).
This comprehensive analysis of hepatocellular carcinoma (HCC), which was co-led by David Wheeler, PhD, Director of Cancer Genomics and Professor in the Human Genome Sequencing Center at Baylor College of Medicine, Houston, and Lewis Roberts, PhD, Professor of Medicine at the Mayo Clinic, identified several genomic changes that may represent new therapeutic targets.
This study, which is a part of The Cancer Genome Atlas project (TCGA), is the first large-scale, multi-platform analysis of HCC looking at multiple dimensions of the tumor.
"We reported on one facet of the TCGA project, which aimed to study 30 of the most common cancers using six genomic platforms: Genome (or Exome) sequencing, expression via RNA-seq, micro RNA expression, DNA methylation, DNA copy number alteration via SNP array, and protein expression via reverse phase protein array," said Wheeler. "This data is coupled with clinical data for the patients, and very careful quality control on the purification of DNA, RNA, and proteins for analysis."
Research Methods
Utilizing whole-genome sequencing and DNA copy number analyses, investigators analyzed 363 HCC cases. Additionally, 196 HCC cases were analyzed by DNA methylation, RNA, miRNA, and proteomic expression.
"The experimental design was to carefully monitor the acquisition of primary untreated tumors of sufficient size and tumor cell content to enable a single DNA purification and RNA purification to be distributed to the data production centers such that all DNA platforms would use aliquots of the same DNA prep; all RNA platforms would use aliquots of the same RNA prep," explained Wheeler. "All tumors were required to pass pathology review to ensure tumor cell content, absence of necrosis, and confirmation of diagnosis.
"Thus each tumor and the molecular isolates of DNA and RNA and protein had to pass stringent quality control criteria before they were distributed to data production centers," he continued. "This ensured that comparisons among the various analytical platforms and even between tumors would be accurate and valid."
Study Results
The findings revealed a number of genomic alterations that influence development of HCC, including mutations in the TERT gene promoter, TP53, and CTNNB1 ([beta]-catenin) genes, as well as elevated expression of several immune checkpoint genes, including CTLA4, PD-1, and PD-L1.
"Intriguingly, we found that the urea cycle enzyme carbamoyl phosphate synthase is downregulated by hypermethylation, while cytoplasmic carbamyl phosphate synthetase II is upregulated," said Karl-Dimiter Bissig, MD, PhD, Assistant Professor of Molecular and Cellular Biology at Baylor and co-author of the study, in a statement. "This might be explained by the anabolic needs of liver cancer, reprogramming glutamine pathways to favor pyrimidine production potentially facilitating DNA replication, which is beneficial to the cancer cell."
As the one of the most common cancers worldwide, liver cancer has been studied genomically in several reports prior to this study. However, Wheeler noted, "all previous studies were limited in scope to just analysis of somatic mutations in DNA.
"Very large-scale studies of whole-exome sequencing involving as many as 500 patients, or whole genome of 350 patients, have been reported, but no study has taken into account somatic mutation, gene expression, DNA methylation, and protein expression simultaneously," he added. "So, although larger cohorts have been studied, none have been so broadly investigated."
Combining all of these data types in a well-controlled cohort allowed researchers to demonstrate for the first time the profound importance of activation of the sonic hedgehog pathway, and the inactivation of urea production in the transformation of the liver cell to the malignant state, according to Wheeler.
"Production of urea as a waste product of protein metabolism throughout the body is one of the chief functions of the liver," he explained. "By shutting down this activity, the liver is able to use nitrogen to synthesize nucleotides, which are essential building blocks of DNA, in order to more rapidly replicate the DNA of the tumor cells. The enhanced replication of DNA enables tumors to grow more rapidly.
"We also uncovered an intriguing link between hepatocellular carcinoma and the bile duct cancers. The latter are even more lethal than hepatocellular carcinomas, and we discovered that approximately 12 percent of hepatocellular carcinomas share molecular features very similar to cholangiocarcinomas," Wheeler reported. "This insight may improve treatment by flagging patients that have a more lethal variant of hepatocellular carcinoma."
According to Wheeler, the greatest limitation of the study is the ability to analyze such large and complex data types bioinformatically. "One of the main goals of the TCGA is to make these high-quality datasets available to researchers worldwide so that, as new bioinformatics methods are developed, this data can be reanalyzed and new discoveries can be made," he noted.
Practice Implications
Given the limited therapeutic options available to HCC patients, this study has significant practice implications.
"There is one targeted therapy approved for use in this cancer, sorafenib, which extends the life expectancy of these patients by only a few months," Wheeler noted. "Our results suggest inhibitors of the sonic hedgehog pathway that are under development in academia and by pharmaceutical companies may be beneficial for these patients.
"Sonic hedgehog has been recognized as an important cancer driving pathway in many cancers, and there were even reports of its involvement in hepatocellular carcinoma," he emphasized. "But those previous reports involved only a small number of patients. Our study showed that the sonic hedgehog pathway is active in at least 30 percent of patients."
"This is outstanding research analyzing a cancer that's increasing in frequency, especially in Texas. Notably, the observation of gene expression signatures that forecast patient outcome, which we validate in external cohorts, is a remarkable achievement of the study," concluded Richard Gibbs, PhD, Director of the Human Genome Sequencing Center at Baylor. "The results have the potential to mark a turning point in the treatment of this cancer."
Catlin Nalley is associate editor.