NEW YORK (GenomeWeb) – Researchers studying the genomes of brain tumor samples from patients with different stages of disease have identified molecular pathways that likely drive the development of those tumors. They also found that glioma cell cultures showed sensitivity to a potential therapeutic called BET inhibitor, although the mechanism is still unclear.
Reporting today in Nature Genetics, researchers from Yale University, Memorial Sloan Kettering Cancer Center, and elsewhere used exome sequencing and DNA methylation analysis to understand why patients with grade II or grade III gliomas and mutations to the IDH1 locus typically progress to develop glioblastoma multiforme, a much more aggressive form of the disease that has a median survival time of only 12 to 18 months.
Gliomas are typically classified as grades I-IV. Most grade IV gliomas, or glioblastoma multiforme (GBM), occur spontaneously, but about 20 percent arise from the progression of grade II or III gliomas. Nearly 80 percent of grade II and III gliomas harbor IDH1 mutations in conjunction with mutations to either TP53 or ATRX. In addition, the IDH1-mutated tumors tend to have extensive DNA methylation.
In order to piece together the events that drive grade II and III gliomas to progress to GBM, the researchers sequenced the exomes of 41 IDH1-mutated tumor samples at initial diagnosis and progression and compared their DNA methylation profiles. Twenty of the samples also had matched normal samples. In tumors with matched normal samples, the researchers identified an average of 31 somatic mutations per tumor.
The researchers identified previously reported recurrent mutations in driver genes as well as new somatic mutations. For example, they found mutations to NOTCH1 and NOTCH2 that were consistent with mutations that have been reported in squamous cell carcinomas as inactivating, as well as mutations to downstream targets within the NOTCH signaling pathway. They also identified new mutations in the FAT receptor genes, including both protein-coding mutations and copy number alterations. In addition, around 44 percent of all tumors had deletions within the tumor suppressor locus CDKN2A-CDKN2B.
Next, the researchers compared the mutational profiles of the diagnosis and progression samples to findevents that drove the progression. On average, the tumors acquired 21 somatic mutations during progression and lost 27 percent of the mutations present at diagnosis and 30 percent of the copy number alterations. Also, 27 percent of the higher-grade gliomas acquired different mutations to the same glioma driver genes that were mutated at diagnosis, such as TP53, ATRX, CIC, and FUBP1. These findings "suggest that genomic heterogeneity, which is known to occur in gliomas, has a substantial role during progression," the authors wrote.
During progression, the pathways and genes that had the most protein-coding mutations were the RTK-RAS-PI3K pathway, the NOTCH pathway, and the FAT receptors, with mutations to those pathways found in 22 percent, 12 percent, and 10 percent of patients, respectively.
Epigenetics also seemed to play a big role. Histone modifier genes were mutated in 44 percent of patients, while mutations to the SWI/SNF chromatin remodeling complex were found in 5 percent of patients, and genes involved in transcriptional modulation were mutated in 12 percent of patients.
Looking at copy number alteration differences between diagnosis and progression samples, the researchers found that the MYC locus was the most frequently amplified gene during progression, present in 22 percent of patients. Amplification of the FOXM1 gene locus was found in 20 percent of patients. Deletions to CDKN2A-CDKN2B were most common and were found across all subtypes. Deletions to PTEN, FAT1, and FAT2 were also frequently found and associated with progression.
Considering all classes of mutations, the researchers noted that the MYC signaling pathway was the most highly mutated. It "became activated in 56 percent of the patients and showed one of the strongest associations with progression," they wrote.
"Overall, our observations suggest that these pathways converge upon opposing but synergistic cellular effects that stimulate proliferation while inhibiting differentiation during glioma progression," the authors wrote.
Finally, the researchers wanted to look at therapeutic possibilities. Previous studies had indicated that inhibition of Bromodomain and Extra-Terminal (BET) motif proteins could be a potential therapy for GBM. BET inhibitors are thought to target MYC signaling, although the exact target is not known. The researchers tested two BET inhibitors on six patient-derived IDH1-mutant primary glioma cultures and one well-characterized glioma cell line. Gilead Sciences provided one of the compounds for the study. The study was also partly supported through a research agreement between Gilead and Yale.
Although the researchers found that the cell cultures were sensitive to BET inhibition, "further studies will be needed to identify the exact therapeutic target of BET inhibition as well as the effectiveness of this inhibition in vivo in gliomas," they wrote.