NEW YORK (GenomeWeb News) - At least two teams of scientists have been exploring the ability of 454 Life Sciences’ platform to detect mutations in tumors, testing the potential advantages of massively parallel sequencing over traditional DNA sequencing for this application.
In the short term, these studies might help decide whether the Genome Sequencer 20 platform will become part of the NIH’s Cancer Genome Atlas pilot project.
The research could also bode well for other manufacturers of massively parallel sequencing technologies, among them Solexa, Helicos, and Agencourt. To be sure, these studies are early stage and only involve a small number of samples. But if the technologies hold up their promise in these and other studies, they could pave the way for next-generation DNA sequencing technology to enter the potentially lucrative cancer diagnostic market.
“The promise of the new sequencing technologies is that they may allow accurate mutation detection for cancer genomes to be more rapid and economical,” Greg Riggins, who is involved in one of the studies, sponsored by the National Human Genome Research Institute, told GenomeWeb News in an e-mail message. He is an associate professor of neurosurgery, oncology, and genetic medicine at Johns Hopkins University.
In fact, the “major finding” of the second study, performed by the Dana-Farber Cancer Institute, the Broad Institute, and 454 Life Sciences, was the tool’s ability to detect mutations “in samples where mutations had been invisible to Sanger sequencing,” according to Roman Thomas, a lead author of the study, which appeared online in Nature Medicine last month. The results “suggest a huge impact on clinical decision-making,” added Thomas, who is currently a postdoc in the lab of Matthew Meyerson at Dana-Farber.
Meanwhile, researchers led by Bob Strausberg at the J. Craig Venter Institute have been sequencing glioblastomas, an aggressive form of brain cancer, in an NHGRI-sponsored pilot study that compares traditional Sanger sequencing and 454’s pyrosequencing approach, GenomeWeb News has learned.
NHGRI, which approved and funded the project several months ago, is planning to describe this and another cancer sequencing pilot project -- both part of its medical sequencing program -- later this week on its website.
In the Nature Medicine paper, the team sequenced small regions from the epidermal growth factor receptor gene in 22 lung adenoma samples using the GS20 and found additional mutations in two samples that Sanger sequencing missed. They also confirmed mutations previously identified by Sanger.
Further, they were able to detect EGFR mutations in a pleural fluid-derived sample that contained less than 10 percent tumor cells --- a finding Sanger sequencing missed.
What made 454’s technology stand out in this study was its ability to find mutations that were only present in a fraction of a clinical sample. That is because the platform analyzes many DNA fragments from the same gene region in parallel, rather than sequencing a pool of DNA strands, in which normal DNA can mask a mutation.
Dana-Farber’s Thomas said that he and others have found that Sanger sequencing reaches its limits when a mutated allele is present at less than 25 percent in a sample. That means that heterozygous mutations could only be detected if at least 50 percent of a sample consisted of tumor cells.
However, in the clinic, it is often not possible to obtain samples with such a high tumor content, Thomas said. Also, mutations might only be present in a small portion of a tumor. “In the clinic … you couldn’t send away a patient saying ‘I am sorry, your tumor belongs to the 90 percent of tumors that don’t have enough tumor cells to analyze,’” he said.
Meantime, in the NHGRI-sponsored pilot study, the scientists plan initially to re-sequence selected regions from 37 cancer-relevant receptor genes in 20 human glioblastoma samples and matching normal controls using standard Sanger sequencing on an Applied Biosystems instrument, according to the project proposal. They then plan to use 454’s GS20 to focus on those gene regions that are rich with mutations.
This “relatively small-scale but potentially highly informative pilot with the 454 technology” would validate and quantify the mutations detected by Sanger sequencing. More importantly, the 454 technology, because it is more sensitive, could potentially find “additional somatic mutations not previously detected with the ABI technology,” according to the proposal.
“If this proves to be true, it will allow researchers to more easily sequence directly from tumor DNA to seek mutations, without the sometime laborious purification steps such as laser capture micro-dissection, or growing the DNA as a xenograft,” Riggins wrote.
In the Queue
Other massively parallel next-generation sequencing technologies, such as those developed by Solexa, Helicos, or ABI’s Agencourt platform, would have the same advantage as 454’s. “Our basic finding is that … every sequencing technology that is parallel in nature might in theory do the same thing,” Thomas said. However, 454's is the only technology widely available at the moment -- and has as a distribution partner the world’s largest diagnostic company by revenue, Roche.
The implications of his and other studies might be twofold, according to Thomas. On the one hand, they might help determine the suitability of 454’s technology for mutation discovery projects, such as the NIH’s Cancer Genome Atlas pilot project.
That project, funded by the NCI and NHGRI over three years with $100 million, will analyze a small number of cancer types using sequencing and other genomic techniques. The first awards, for up to four Cancer Genome Characterization Centers, will be made by the end of September, the end of the NIH’s fiscal year.
In fact, data from the Venter Institute-led project “is likely to be integrated into the Atlas,” according to Riggins.
On the other hand, 454’s technology might have direct applications in the clinic, where it could, for example, find rare resistance alleles in tumor samples prior to therapy, Thomas said. He will start his own research group at the University of Cologne in Germany this fall, and plans to establish clinical trials where patients are selected for therapeutic intervention based on results from 454 sequencing.
Julia Karow covers the next-generation genome-sequencing market for GenomeWeb News. E-mail her at [email protected]