A team of scientists led by the Wellcome Trust Sanger Institute has discovered approximately 100 new cancer genes in a large-scale sequencing study of human cancers.
The research, part of the Sanger Institute’s Cancer Genome Project, paves the way for an international cancer genome-sequencing effort within the next two years, according to one of the study leaders.
The Sanger researchers are also currently exploring new sequencing technologies for cancer genome sequencing, hoping they will lower the cost of whole-genome sequencing.
At the moment, sequencing entire cancer genomes is cost-prohibitive. “The genome is extremely large, and it is not possible on current technologies to screen a large number of cancers all the way through the genome,” Mike Stratton, one of the leaders of the Sanger Institute’s Cancer Genome Project, said in a press briefing last week.
For that reason, the researchers decided to confine their study to about 3 percent of the protein-coding fraction of the genome, and “explore what large-scale cancer genome sequencing might bring us by looking at one gene family,” in this case protein kinases, he said.
The reason they chose this gene family is that “several members have been shown to be involved in cancer,” added Andy Futreal, co-leader of the project. Also, kinases “make good drug targets,” he said. For example, Gleevec from Novartis targets the BCR-ABL kinase in the treatment of chronic myeloid leukemia.
Using standard capillary electrophoresis methods, the researchers sequenced the coding regions of 518 protein kinase genes in 210 human cancer samples that included breast, lung, colorectal, gastric, testicular, ovarian, and renal cancer, as well as samples of melanoma, glioma, and acute lymphoblastic leukemia.
In total, they sequenced 274 megabases of DNA and found 1,000 somatic mutations. Analyzing these further, they characterized 350 of them as so-called “driver mutations” in approximately 100 new cancer genes.
“We call them drivers because they drive the cells to stop behaving normally and to start behaving in that abnormal fashion that we call a cancer,” Stratton explained.
The remaining mutations, so-called passenger mutations, “have nothing to do with causation of cancer, but hitchhike along for the ride,” he said. “A major challenge is always going to be to separate out those driver mutations from the passenger mutations.”
Overall, the results were a surprise. “This is a lot larger number of cancer genes than we really expected to find,” Stratton said, adding that he believes “there are many more cancer genes out there.”
With mutation data from an entire gene family in hand, the researchers can now start analyzing not only individual cancer genes but “ask questions about the role of kinase networks,” Futreal said.
The sequencing study, which covered about 3 percent of the human genome’s protein-coding sequence, is part of the Sanger Institute’s Cancer Genome Project, a multi-study effort launched in 1999.
This was not the first large-scale cancer sequencing study. Last fall, scientists at Johns Hopkins University published a study in Science in which they sequenced 13,023 genes in 22 cancer samples, covering breast and colorectal cancer.
“The major conclusions of the two studies are similar,” Victor Velculescu, who led the Hopkins study, told In Sequence this week. Both showed that more different genes than expected are mutated in cancer, and that there are “pretty dramatic differences” between different cancer types, and even between individual tumors of the same type, he said.
But because their study did not look at a particular gene set but analyzed two thirds of all human genes, the Hopkins researchers were able to find “lots of different cellular processes” that are involved in cancer, he said.
ABI is collaborating with the Hopkins researchers to validate its SOLiD sequencing platform for cancer sequencing (see GenomeWeb News 09/12/2006, an In Sequence sister publication).
But these studies are only the beginning, Stratton stressed. “We really ought to be aiming for the complete sequence of the genomes of maybe 10,000 to20,000 individual human cancers, covering all the common and many of the less common types of cancer,” he said.
The next phase “undoubtedly will be an international consortium, all the groups with interest in it, working in a coordinated and connected way, in a similar fashion to the way that the human genome was sequenced in the first instance,” Stratton said. “And we look forward to that happening in the next couple of years.”
The Sanger Institute and the NIH’s National Cancer Institute did not respond to requests from In Sequence about who might be involved in this international project, and who will coordinate it.
“This is a lot larger number of cancer genes than we really expected to find. There are many more cancer genes out there.”
Velculescu, who is an advisor for the NCI/NHGRI’s Cancer Genome Atlas, said he did not know of any plans for an organized international effort. “I think with time, there will be more coordination between these groups in order to try to make this a productive project overall,” he said. “But there isn’t anything formal, as far as I know, along these lines.”
However, any international effort will likely include the Sanger Institute, the Hopkins group, and the Cancer Genome Atlas, which launched at the end of 2005.
As part of TCGA’s three-year pilot phase, researchers will map genomic changes in lung, brain, and ovarian cancers. NHGRI’s three large-scale sequencing centers will sequence a “substantial number” of selected gene targets in these cancers [see GenomeWeb News, 11/20/2006).
Both the Sanger Institute’s and the Hopkins study “will guide and will inform the Cancer Genome Atlas project as it grows and progresses,” Velculescu said.
Next-generation sequencing technologies will likely play an important effort in any large-scale cancer genome sequencing project. “Ourselves and lots of other groups are investigating the current wave of new technologies, which are literally light years ahead of what we are currently using,” Futreal said in last week’s briefing.
However, he believes that not this generation of new technologies “but probably generation 1.2” will allow researchers to tackle entire genome from large numbers of individual samples. “But certainly there is an interim series of technologies which allow us to expand these types of work, which will drive the cost down, which will allow us to do more tumors.”