In a move in line with its continued shift towards molecular diagnostics, Helicos Biosciences has developed a method for capturing and sequencing the BRCA1 gene directly from genomic DNA.
A proof-of-principle study of this capture sequencing approach — which forgoes gene isolation or amplification in favor of single-molecule sequencing of the gene from sheared genomic DNA — was published in the advance, online version of the journal Genome Research in July.
Using samples from the Coriell biorepository, researchers from the company showed that it was possible to detect both mutations and rearrangements in BRCA1, a gene implicated in inherited breast and ovarian cancer risk. Those involved in the study said the same approach is also showing promise for detecting genetic glitches in a related cancer risk gene, BRCA2, though that work has not yet been published.
"Because no gene isolation or amplification is required for sequencing, the exceptionally low costs of sample preparation and analysis could make genetic tests more accessible to those who wish to know their own disease susceptibility," the Helicos team wrote.
Patrice Milos, senior vice president and chief scientific officer at Helicos and the study's senior author, told Clinical Sequencing News that she and her colleagues eventually hope to see their BRCA1/2 sequencing method offered by a CLIA-certified laboratory.
"Our goal would be to have a CLIA-validated laboratory running the service in partnership with us," Milos said, noting that additional, validation studies are needed before that happens.
As part of a series of restructuring steps, Helicos last year dropped its plans to commercialize its single-molecule sequencer in favor of developing the system as a diagnostic platform. The company initially planned to run its molecular diagnostics business through its own CLIA lab, but abandoned that plan in late 2010 due to cost considerations (IS 11/30/2010). It said in its first-quarter financial report that it plans to "evaluate the best modality of bringing high-value MDx tests based on our technology to the market through a CLIA laboratory in partnership, through out-licensing, or our own efforts should the financial outlook for us improve." (IS 5/17/2011).
The company has said in the past that it expects its first test to be for inherited breast and ovarian cancer, but the Genome Research paper marks the first time it has disclosed its plans to specifically target BRCA testing — a strategy that pits it directly against Myriad Genetics, which dominates the BRCA 1 and BRCA 2 testing market and wields a substantial IP portfolio protecting the two genes.
Though Helicos's direct capture method can also be used to capture and sequence sites where retroviruses, transposons, or other mobile bits of DNA have become integrated in the genome — or even to directly capture and sequence plant genes — Milos said the company's current focus will be on using the approach to find clinically informative mutations in BRCA1, BRCA2, and other medically important genes.
Milos said that the current study stemmed from efforts to continue innovating on the company's sequencing platform, the Heliscope Genetic Analysis System, which sequences molecules by capturing them on sequencing flow cell surfaces that have oligonucleotides attached to them.
"We kept thinking about the ability to sort of reverse the surface and make it a capture region for repetitive regions of the human genome, transposons, and then all the way down to the level of individual genes of interest," she said.
Once they'd found that it was feasible to capture and sequence repetitive elements and transposons, which are abundant in the genome, Milos recalled, "We said, 'Oh, what the heck, let's try it with a gene or genes of really high medical relevance and see whether it could be used for clinical diagnostic sequencing … And lo and behold, it worked absolutely beautifully."
The Direct Capture Method
To directly sequence genes of interest, the Helicos researchers use custom flow cells that capture these sequences from acoustically fragmented genomic DNA using gene-specific primers within the flow cell. Once annealed to the primers, this DNA can be sequenced with the Helicos single molecule platform.
Unlike whole-genome Helicos sequencing, the strategy does not require the addition of a poly-A tail or 3'-end blocking step during sample preparation. Instead, the researchers wrote, "The entire gene can … be sequenced without copying the DNA or using any enzymes prior to sequencing."
Though the method involves little in the way of sample preparation — just extracting and shearing genomic DNA from patient samples — it does require some strategizing about primer sequences.
"The flow cell is specific for the DNA region that you're interested in," lead author John Thompson, senior director of genomic sciences at Helicos, told CSN. "So you have to make a different flow cell for each set of genes that you want to look at."
"But typically in a clinical setting you're looking at particular genes anyway," he added, "so it's not really an issue."
He said primers are typically spaced roughly 60 to 70 bases apart on a given sequence of interest, though the team has been working to optimize read length and, in turn, stretch out the distance between primers.
For instance, the researchers reported that they could use so-called 'JumpStart' sequencing to extend primers after genomic DNA is captured in the flow cell but before sequencing to address low coverage in parts of the gene with poor hybridization or repetitive regions that are more difficult to capture.
For the current study, the team designed flow cells containing 564 oligonucleotides, in primer sequences between 24 and 41 nucleotides long that corresponded to regions of BRCA1 containing clinically significant mutations documented in the Breast Cancer Information Core database.
"Though the current flow cell is directed at the entire coding sequence and 20 [nucleotides] of adjacent intronic sequence," the researchers wrote, "it is straightforward to incorporate any other sequences that might be found to be medically relevant."
The team demonstrated that they could generate reads from both strands of DNA for the BRCA1 gene using genomic DNA from 22 HapMap samples from the Coriell Institute for Medical Research biorepository.
After shearing this genomic DNA into bits that were each around 200 base pairs long, investigators hybridized between 0.1 and 20 micrograms of genomic DNA in each channel of a 25-channel flow cell overnight and then sequenced the gene through 120 nucleotide-addition cycles.
The researchers said they tended to get more reads corresponding to BRCA1 when higher concentrations of genomic DNA were used, but noted that lower genomic DNA concentrations could also be used in a pinch.
Once they'd tossed out sequence artifacts and reads shorter than 25 nucleotides long, the team aligned the reads to a BRCA1 reference sequence. More than 20 percent of reads aligned to the reference gene when the greatest concentrations of genomic DNA were used, which they said represents an enrichment of greater than 100,000-fold since BRCA1 exons represent only 0.0002 percent of the genome. In addition, nearly all the reads that did not align with the gene aligned "sporadically" to other parts of the genome.
Although they noted that there were some differences in coverage at different positions across the gene due to variable hybridization efficiency for the primers used, the researchers found that coverage at specific sites in the gene was consistent from one sample to the next. For 19 samples that they ran simultaneously on a single flow cell, for instance, they reported a median coverage of 47-fold for each of the BRCA1 regions targeted.
Within the 22 samples tested, the researchers found all of the BRCA1 variants that had been previously identified in the well-characterized samples, using the Helicos SNP-calling algorithm SnpSniffer to find substitutions and comparisons between coverage levels within multiple samples to track down larger rearrangements.
[ pagebreak ]
"The advantage with this technique is that you can also see much larger insertions and deletions than you can with standard sequencing," Thompson said, noting that studies have shown that between about 10 and 12 percent of high-risk patients harbor large insertions and deletions in BRCA1 that can be missed by Sanger sequencing or amplification-based testing methods.
Liying Zhang, director of Memorial Sloan-Kettering Cancer Center's diagnostic molecular genetics laboratory, who was not involved in the research, said the study "is a very nice proof of principle that this technology allows detection of all previously identified variants and mutations in BRCA1."
But while the method seems like it should be scalable to allow testing of more genes in one run and has the potential to detect large structural variants, Zhang told CSN in an e-mail message that "more careful and complete validations are necessary before it can be introduced into clinical molecular diagnostics laboratories."
In the future, Thompson and his colleagues plan to look at even more samples while comparing their BRCA1 mutation results to internal standards to ensure the approach continues to yield results that are consistent with those other techniques. They are also looking at suspicious patient samples with no obvious BRCA1 or BRCA2 mutations to see if they can find novel mutations not detected by other methods.
"The challenge for us is having the clinical specimens to query all of the different possible types of mutations that you might find in BRCA1 and BRCA2," Milos said. "And frankly, I don't think we know the full extent of those."
She said the company has been building relationships with some unnamed cancer centers that are interested in exploring alternative methods for BRCA1 and BRCA2 testing. Helicos hopes to forge partnerships at these centers that will provide access to more samples from individuals who may have BRCA1/2 mutations. Ideally, Milos said, the company would like to work with collaborators on prospective, head-to-head studies comparing the Helicos Direct Capture BRCA1/2 assay with conventional BRCA1/2 testing methods.
Zhang noted that even if the approach is "fully validated and demonstrates high sensitivity and specificity comparable to the existing technologies," there are a number of other factors at play that would determine how it would stack up to existing methods in a real-world setting, including cost, turnaround time, assay design issues, and data analysis.
"Without this information, it is hard to conclude whether this new method is better than the existing technologies currently used in molecular diagnostics laboratories," Zhang wrote.
Thorny IP Landscape
Tackling BRCA1 may seem to be a bold move given the intellectual property issues surrounding the gene. Both the BRCA1 and BRCA2 genes are covered under Myriad's patents, which are at the heart of an ongoing court battle over gene patenting.
But the Helicos researchers believe the nature of their sequencing-based BRCA1/2 diagnostic method — which does not involve isolation of the gene prior to sequencing — may sidestep some of these intellectual property issues.
As reported by CSN sister publication, Pharmacogenomics Reporter, a recent federal appeals court ruling upheld Myriad's right to patent isolated versions of BRCA1 and BRCA2 (PGx Reporter 8/3/2011). But the company's methods patents covering comparisons and analyses of the gene were not included. Furthermore, Myriad's lawyers suggested as part of the appeals court hearing that whole-genome sequencing that doesn’t involve an isolation step would not infringe the company's BRCA patents.
"Because the DNA is not copied or amplified prior to sequencing, it avoids any of the patents that are out there," Thompson said. "This is not isolated, because you're putting down all of the genomic DNA at once and then you're pulling out the sequence that you want."
A representative from Myriad said the company was not sufficiently familiar with the Helicos protocol to comment on its potential IP implications.
For her part, Milos said it remains to be seen how the legal issues related to BRCA1/2 testing shake out. "We've always felt that an individual's DNA sequence in their natural DNA is an individual's and that individual's natural DNA really shouldn't be the subject of patents," she noted. "But I'll leave that to the courts."
Despite the questions that remain about BRCA1/2 IP, Helicos's long-term goal is to see its test running in a CLIA-certified laboratory. That goal likely won't be realized in the near future, Milos noted, since the company must first secure funding for studies demonstrating equivalency to the standard of care.
"The plan would have to be funded," she said. "It's a question of whether we can find the capital to do the work."
As of March 31, Helicos had $2.5 million in cash and cash equivalents. The company has not yet reported its financial results for the three months ended June 30.
Should it reach the clinical diagnostics market, the team believes the Helicos Direct Capture BRCA1/2 assay has an edge over competitors, in part, because of the ease of sample preparation and decreased risk of introducing errors prior to sequencing compared to methods that require DNA amplification.
"When you think about the ability to use sequencing for gene regions that are medically relevant, the workflow that a molecular diagnostics lab really benefits from is 'the simpler the better' — so that you have a streamlined workflow that makes getting from a patient sample to clinically actionable information very simple and straightforward," Milos said.
"In a clinical setting, what's really important is reproducibility and accuracy," Thompson added. "This is really perfect for that kind of situation, because sample handing is minimal and it's very reproducible."
The team also touted the method's ability to detect sequence mutations and copy number changes simultaneously. Myriad's BRCAnalysis test relies on PCR amplification and Sanger sequencing to detect mutations in BRCA1 and BRCA2, but that company uses a separate test, BART Analysis, to find rearrangements in the genes.
Helicos did not disclose how much the Direct Capture BRCA1/2 assay would cost, should it become available, but Milos said the cost of reagents used in the Genome Research study was well under $100 per sample, not including labor costs.
Along with ongoing work with BRCA1, the researchers have also done similar experiments with BRCA2 and other undisclosed genes that are known or suspected to be clinically relevant.
"The novel approach described here holds broad potential not only for BRCA1 but also for any other gene for which accurate, quantitative sequencing is medically important," the study authors wrote. "The minimal sample handling and lack of amplification translate directly to lower sample preparation costs and reduced sample handling errors critical for molecular diagnostic laboratories."
Have topics you'd like to see covered in Clinical Sequencing News? Contact the editor at anderson [at] genomeweb [.] com.