Two independent research teams have analyzed the genomes of pancreatic cancer and glioblastoma multiforme, both particularly fatal malignancies, and discovered new molecular alterations that may lead to novel diagnostics and therapeutics.
The studies, which involved both Sanger sequencing and second-generation sequencing technologies, lay the groundwork for large-scale cancer genomics studies such as the remaining pilot projects of the National Institutes of Health’s Cancer Genome Atlas and projects planned by members of the International Cancer Genome Consortium.
The first study, published online last week in Nature by the TCGA Research Network, included the sequence analysis of several hundred known or suspected cancer genes in almost 100 glioblastomas. Separately, a research consortium led by the Ludwig Center for Cancer Genetics and Therapeutics at the Johns Hopkins Kimmel Cancer Center, published two studies online in Science last week that involved sequencing almost 21,000 genes in 22 glioblastomas and 24 pancreatic cancers.
The Nature paper is the first publication resulting from the TCGA pilot project, which launched in late 2005 with $100 million in funding over three years from the National Human Genome Institute and the National Cancer Institute. Its aim is to gauge the feasibility of a large-scale cancer genomics project by characterizing the genomes of several hundred lung, brain, and ovarian cancer samples.
The TCGA study analyzed approximately 600 genes in 91 human glioblastomas as well as DNA copy number changes, gene expression, and DNA methylation in 206 glioblastoma samples.
The sequencing data was generated by the TCGA’s three genome sequencing centers: the Genome Sequencing Center at Baylor College of Medicine, the Broad Institute of MIT and Harvard, and the Genome Center at Washington University School of Medicine in St. Louis.
After PCR-amplifying the exons, the researchers sequenced them on ABI 3730xl capillary electrophoresis sequencers. They validated putative SNPs on Sequenom or Illumina Golden Gate genotyping platforms, with TaqMan or Biotage assays, or by 454-based resequencing.
They also used 454’s technology to validate putative indels, noting in their paper that “since the 3730 and 454 sequencing technologies utilize different chemistries and detection methods, the 454 provides an independent platform for validating sequence variants.”
The study results showed that most genetic changes in glioblastoma patients affect three core cellular pathways: those controlling the cell cycle, cell growth, and cell death.
“While these pathways have been known to be involved in cancer, what has never been observed before is the extent to which the pathways are knocked out or deregulated in individual patients,” David Wheeler, associate professor at Baylor’s GSC and an author of the study, told In Sequence in an e-mail. “What made this view possible is the comprehensive sequencing of a large number of genes, known or suspected to play a role in cancer, in so many people.”
The study also discovered that three genes known before to be involved in other cancers also play a key role in glioblastoma. As a result, therapies directed at those genes or their products that have been used in other cancers can now also be tested in glioblastoma patients, Wheeler said.
Finally, the study unearthed connections between genes involved in DNA damage repair and chemotherapeutics that can give insight into resistance mechanisms and may help devise ways to overcome resistance, he said.
“What made this view possible is the comprehensive sequencing of a large number of genes, known or suspected to play a role in cancer, in so many people.”
“This answers the big question about whether the cancer genome project is worthwhile,” said Richard Gibbs, director of Baylor’s genome sequencing center, in a press statement. “The result shows that it is, definitely.”
According to the statement, the TCGA researchers decided to publish their results before completing the glioblastoma project because of the significance of the results.
According to Wheeler, the aim is to sequence the exons of 1,300 genes “with the possibility of adding a few hundred more” in 500 glioblastomas and matched controls, of which 130 pairs have been completed.
The greatest challenge has been acquiring tumor and matching control samples that meet the criteria for the study, Wheeler said. “The tumor samples must meet stringent criteria for tumor content, and the accompanying clinical data must be thorough,” he said. “Over half the potential samples donated to the project fail these quality control criteria.”
Second-generation sequencing technologies will play a larger role for the remainder of the glioblastoma project, as well as for the pending lung and ovarian cancer pilot projects of the TCGA.
“We are planning a transition to next-generation platforms by January ’09,” Wheeler said. “If all goes according to plan, the majority of the ovarian cancer samples and all of the lung squamous cell carcinoma samples will be sequenced using next-gen platforms.”
The TCGA’s study results overlapped with one of the two studies published by the Hopkins team in Science last week. In that study, researchers characterized 22 glioblastomas, sequencing almost 21,000 protein-coding genes and analyzing genome amplifications, deletions, and gene-expression profiles. They also sequenced genes that they found to be mutated more than once in 83 additional cancers.
“They sampled many more genes and, as a result, found a very unexpected gene, isocitrate dehydrogenase 1, a gene known heretofore only for its role in cellular metabolism,” Wheeler said. “However, because their discovery set had only 22 patients, their study was less sensitive overall.”
The Hopkins researchers found mutations in IDH1, which were associated with longer survival, in a large fraction of young glioblastoma patients as well as in those patients with secondary glioblastoma.
“The discovery of IDH1 and other genes previously not known to play a role in human tumors validates the utility of genome-wide genetic analysis of tumors in general and opens new avenues of basic and clinical brain tumor research,” the authors noted.
In their pancreatic cancer study, they discovered genetic alterations that “defined a core set of 12 cellular signaling pathways and processes that were each genetically altered in 67 percent to 100 percent of the tumors.”
“Our studies suggest that the best hope for therapeutic development may lie in the discovery of agents that target the physiologic effects of the altered pathways and processes rather than their individual gene components,” they write.
Agencourt Bioscience provided Sanger sequencing services for both Hopkins-led studies. The company had already delivered similar services for previous cancer genome studies, for example on breast and colorectal cancer, by the same Hopkins group. Last year, Agencourt’s parent Beckman Coulter said that it had obtained two options from Hopkins to license cancer mutations discovered in sequencing-service projects that Agencourt had conducted for the Johns Hopkins Kimmel Cancer Center since 2002 (see In Sequence 11/6/2007).
But the latest glioblastoma and pancreatic cancer studies also studied the transcriptomes of the samples by serially analyzing gene expression, or SAGE, using Illumina’s Genome Analyzer. It was not immediately clear who among the research team or its collaborators performed this analysis. The TCGA researchers, on the other hand, used Affymetrix and Agilent microarrays to generate mRNA and miRNA expression profiles of the glioblastomas they studied.
According to Hopkins researchers, their SAGE sequencing approach “is similar to that used in recent RNA-Seq studies, but SAGE has the advantage that the quantification does not depend on the length of the transcript.”