RainDance Technologies plans to launch a somatic cancer sequencing panel this spring that focuses on key areas of about 50 genes that are often mutated in cancer. It will likely compete with similar panels offered by Illumina and Ion Torrent, promising advantages in terms of DNA input requirements, ease of workflow, and overall cost per sample.
The cancer panel, called ThunderBolts, runs on a $50,000 component of RainDance's RainDrop digital PCR system, followed by sequencing on an Illumina platform. It is the first application of the company's RainDrop instrument for DNA target enrichment, expanding RainDance's reach into that market.
"This product far and away has generated more buzz, more interest, and frankly opened more doors than anything we've launched in our history," RainDance CEO Roopom Banerjee told Clinical Sequencing News.
ThunderBolts includes all of the content of the Illumina TruSeq Amplicon Cancer Panel, a 48-gene panel that is optimized for use with the MiSeq sequencer, and of the Ion AmpliSeq Cancer Hotspot Panel v2, a 50-gene panel that Ion Torrent sells for use with the PGM. In collaboration with a group of early-access customers, RainDance is currently testing the performance of the panel and finalizing its design. It expects the panel will cover at least 50 genes at launch.
The ThunderBolts panel requires at least 10 nanograms of starting DNA, similar to the input requirement for the AmpliSeq panel, but over 10 times less than the 250 nanograms needed for the TruSeq panel.
It is compatible with DNA from formalin-fixed paraffin-embedded samples, which is often damaged, and can also use DNA extracted from fresh-frozen tumors and free circulating tumor DNA from plasma samples.
According to Banerjee, while Ion Torrent has had "some success" in targeted sequencing of FFPE samples, Illumina has not addressed this market much because of its starting sample requirements. ThunderBolts will allow customers using Illumina sequencers, including the recently launched NextSeq, to analyze those samples. It thus complements rather than competes with Illumina's TruSeq panel, he said, and directly competes with the AmpliSeq panel.
"One of our objectives was to build an Illumina-friendly system that … enables this community," Banerjee said. "Given that Illumina has not historically targeted this market, we're effectively taking a market that they have been locked out of, or haven't looked at, [and] opening it up for them." He said both companies will benefit by increasing the use of Illumina's platform for targeted tumor analysis.
ThunderBolts will also compete favorably with both Illumina's and Ion Torrent's cancer panels in terms of overall cost, he said. Depending on the amount of multiplexing through indexing, combined sample prep and sequencing costs for the RainDance assay can be as low as $100 per sample, compared to $250 to $500 for the other vendors' panels, according to Banerjee.
The ThunderBolts workflow, including PCR enrichment and integration of sequencing adaptors and barcodes, takes about an hour and a half, and a MiSeq run adds another 25 hours, Banerjee said, compared to several days required for target enrichment and sequencing with the Illumina TruSeq panel.
But ThunderBolts requires users to have access to the RainDrop source instrument, which costs about $50,000, to perform the droplet-based target amplification. RainDance has so far only sold the instrument as part of the RainDrop Digital PCR system, which also includes a sensor instrument and costs $125,000.
Existing customers of the RainDrop system can now run the ThunderBolts assay and do not need to upgrade their hardware or software. "The fact that we have a vibrant and fast-growing RainDrop community already pre-populated and able to run ThunderBolts will only accelerate the adoption of that assay into the market," Banerjee said.
RainDance is currently testing the ThunderBolts cancer panel with a small number of customers under a "first-access program" and expects those users to present data at the American Association for Cancer Research annual meeting in April.
After that, it plans to sell the panel to more customers under a limited release program. By the end of the year, customers will also likely be able to order custom sequencing panels that run on the RainDrop.
Early-access users weigh in
Tony Godfrey, a professor of surgery at Boston University School of Medicine, has been testing the ThunderBolts panel with MiSeq sequencing on more than 10 fresh frozen esophageal adenocarcinoma samples so far, using a RainDrop system his lab had originally purchased for digital PCR assays, and starting with as little as 50 nanograms of DNA.
Up until now, his group, which outsources sequencing to an external provider, has been using the AmpliSeq Cancer Hotspot panel, combined with PGM sequencing, for their analyses.
Godfrey told CSN that the ThunderBolts assay "worked out great" the first time his lab used it. The overall workflow of the protocol, while similar in length to the AmpliSeq protocol, is considerably less complex and requires less hands-on time, he said. It also includes fewer clean-up steps, during which DNA can be lost.
ThunderBolts also has smaller costs associated with library preparation, and likely lower sequencing costs, he said, although he cautioned that sequencing costs are difficult to compare between platforms because of the many variables involved, such as the degree of multiplexing. He said he believes the quality of the Illumina MiSeq sequence data is better than that of the PGM.
"You get the advantages of AmpliSeq in terms of sample input … and the advantage over AmpliSeq is really the quality of the sequence data and the ease of the use," he said.
Godfrey's group has not tested ThunderBolts on FFPE samples yet but the amplicon sizes should be compatible with degraded FFPE DNA, he said, though the amount of DNA required might be higher than for fresh frozen samples.
Going forward, Godfrey's team plans to customize the ThunderBolts panel for its needs, omitting genes that are not relevant for their cancer type and adding other genes. "It will be interesting to see how easy that process is and how flexible it is," he said.
"We're going to be sequencing a lot of tumors, so we can identify the mutations, and then we'll be using the RainDrop digital PCR to identify the mutations in plasma samples from patients," he said. "That's really nice for us because we bought the RainDrop system just for the digital PCR, and now we get to use it for the sequencing piece as well."
In addition, he plans to decrease the size of the amplicons in the panel in order to be able to sequence cell-free circulating tumor DNA from plasma samples.
Pierre Laurent-Puig, a professor at the Paris Descartes University Medical School, has tested the ThunderBolts panel in eight colon and lung cancer FFPE samples so far, and the results have been "very satisfactory" in terms of mutation detection and coverage, he told CSN, while the library preparation has a "simple workflow." His lab now plans to run additional tumor samples and will also test the detection of mutations in plasma samples from matching cancer patients. Up until now, his group has been using a number of catalog and custom AmpliSeq panels.
RainDance targets two main markets with the ThunderBolts assay, Banerjee said: molecular pathology laboratories that analyze tumor biopsies and labs interested in sequencing FFPE samples stored in biobanks.
Longer term, the company plans to submit the assay to the US Food and Drug Administration for clearance – either directly or through a partner – but in the short term, "we're closely monitoring the FDA's regulatory guidance, which continues to evolve," he said.
While the company may offer other predefined panels in the future – including additional cancer panels and panels for immune monitoring – it has no immediate plans for doing so.
RainDrop and ThunderStorm; RDT 1000 is out
By offering targeted enrichment on the RainDrop platform, RainDance is expanding its reach into the targeted sequencing market.
The company already sells two platforms for DNA target enrichment – the RDT 1000, which processes one sample per run, and the ThunderStorm, which takes up to 96 samples per run and can amplify up to 1,000 genes – but the number of customers has been limited due to the high cost of those instruments.
The RDT 1000 will be phased out by the end of 2016, and the company has already converted many of its RDT customers to the ThunderStorm, which costs "north of" $250,000. RainDance sells a Cancer Hotspot Panel, which targets important parts of 54 cancer genes and requires 250 nanograms of starting DNA, for the ThunderStorm.
Only about 150 to 180 potential customers exist for the ThunderStorm in total – core labs and large clinical laboratories, such as Myriad Genetics, that can make use of the high throughput of the system and can justify its cost – whereas RainDance anticipates that "thousands of sites" will start using the RainDrop for target enrichment, Banerjee said, many of them existing digital PCR users.