If you want to see the front lines of personalized medicine, take a look at the treatment of patients with non-small-cell lung cancer (NSCLC).
Oncologists have a veritable arsenal of weapons for treating NSCLC, including several targeted to specific molecular mutations.
Patients with certain “activating mutations” in the EGFR gene, for instance, make excellent candidates for second-generation tyrosine kinase inhibitors, such as Tarceva and Iressa, explains Lin Wu, vice president for development at Roche Sequencing Solutions; patients who become resistant to those meds often contain a different mutation (T790M), making their disease susceptible to third-generation drugs, including Tagrisso.
“The clinical utility of EGFR mutations is very clear,” Wu says.
First, though, oncologists need access to lung-biopsy material with which to assess those mutations. And that isn’t trivial. Lung biopsies are painful, invasive procedures, and in about 25% of patients, Wu says, they cannot be done. Even if they can, there’s no guarantee they will be representative of the entire tumor, especially in the case of metastasis. And oncologists certainly aren’t going to keep doing them to monitor tumor response as treatment progresses.
But there is another option: Liquid biopsies.
Liquid biopsies are exactly what they sound like. Instead of solid tissue, they use biofluids, such as blood plasma, urine and cerebrospinal fluid, as circulating (and easily accessible) sources of biomarkers. Those biomarkers come in three basic flavors, one of which is cell-free, circulating tumor DNA (ctDNA).
According to Dawne Shelton, a staff scientist at Bio-Rad Laboratories’ Digital Biology Center, ctDNA analysis offers oncologists the potential to integrate tumor genetics in a way that isn’t possible with solid biopsies. “Every tumor biologist knows that one small piece of a tumor doesn’t represent the whole tumor, let alone metastases,” she says. But ctDNA pulls DNA from every cell, even those in metastases unknown to the physician. As a result, she explains, “It gives a much more holistic view of how the system is doing: Are you globally responding to the drug? Are the tumors shrinking or dying off with this treatment?”
Of course, to make such determinations, clinicians must actually harvest ctDNA. Here, we review some of your options for doing so.
ctDNA purification
According to Marco Polidori, global product manager, DNA purification, at QIAGEN, researchers interested in ctDNA face two primary challenges. First, circulating DNA is very sparse, in the nanogram range, with levels varying substantially between individuals. Furthermore, he says, ctDNA is invariably fragmented. “Blood is a hostile environment for DNA or RNA, so the DNA mostly consists of short fragments, and you have to be able to isolate those,” says Polidori.
Traditional DNA purification kits tend to filter out short DNA pieces, Polidori explains. But kits intended for ctDNA, such as the QIAamp Circulating Nucleic Acid Kit from QIAGEN, do not. QIAGEN currently offers both research-use-only (RUO) and in vitro diagnostic versions of the kit; a third version, intended for use with the company’s QIAsymphony automated platform, is slated to launch later this year.
According to Polidori, ctDNA purification kits basically come in one of two formats: column or magnetic bead-based. The QIAamp Circulating Nucleic Acid Kit is driven by a vacuum-powered column to accommodate up to 5 ml of sample at a time, thereby increasing sensitivity.
Promega’s Maxwell® RSC ccfDNA Plasma Kit is bead-based. According to Chris Moreland, global product manager, the kit provides “hands-free” purification of ctDNA from up to 16 samples at a time using paramagnetic beads. “The user must only pipette into the first well of the pre-dispensed Maxwell cartridge and initiate the run on the tablet,” Moreland explains. “There are no other loading steps or user interventions needed during the process.”
Labs looking to see how well they do at purifying and detecting these DNAs can use reference standards from a variety of tool providers, including Horizon Discovery. According to Jonathan Frampton, the company’s associate director of diagnostics business development, these standards—comprising fragmented genomic DNA in three different allelic frequencies—can help companies and laboratories in validating and developing their own assays and can help users ensure that assays are working properly in their hands. Among other things, he says, laboratories can use the standards to assess their sensitivity to different mutations.
Horizon Discovery offers two off-the-shelf standards, a “multiplex standard” of eight variants in four genes, and a BRAF V600E standard, which specifically looks for a biomarker commonly found in colorectal cancer and melanoma. A third standard, for identifying larger structural changes in the genome, is in development, but laboratories developing their own assays can also contact Horizon to create custom standards. In that case, he says, the company, which specializes in genome editing, could use CRISPR/Cas9 technology to introduce the desired mutation into a known genetic background and then use those edited cells as the “building block” for a new standard sample.
Using ctDNA
After researchers have their ctDNA in hand, they can quantify it using either next-generation DNA sequencing (NGS) or PCR-based approaches.
Roche Molecular Diagnostics, for instance, recently released the cobas® EGFR Mutation Test v2, a multiplexed qPCR-based assay that scans ctDNA or tissue-derived DNA for any of 42 different EGFR mutants in seven locations across the gene sequence. “The analytic sensitivity of [the] cobas EGFR blood test can be as low as 10 to 50 copies of mutant DNA, depending on the mutation sites,” says Wu. (The test is FDA-approved for tissue only, says Wu, and CE-IVD marked for blood.)
In early April, a research team led by Adrian Sacher at the Dana-Farber Cancer Institute used Bio-Rad’s Droplet Digital PCRTM (ddPCR) technology in “the first prospective study of of the use of ddPCR-based plasma genotyping for the detection of EGFR and KRAS mutations” [1]. The team scanned 180 NSCLC patients’ ctDNA for mutations in EGFR and KRAS, two actionable genes commonly mutated in NSCLC. “This assay exhibited 100% positive predictive value for the detection of these mutations,” the researchers wrote, with sensitivity values of 64% to 82%.
According to Polidori, any researchers interested in working with ctDNA should be aware of two potential complications. First, blood samples should be stabilized—using QIAGEN’s PAXgene ccf (circulating cell-free) DNA collection tubes, for instance—to ensure white blood cells remain intact. “That’s a huge problem,” he says. “If the cells die in the tube, you will mask your circulating DNA.” Second, if you plan to use NGS for detection, be sure to use a library preparation kit capable of handling very low (and potentially variable) DNA inputs.
Logistical difficulties notwithstanding, though, liquid biopsies are emerging as a big business for QIAGEN, he says. “We see a very strong uptick in interest in the market in all kinds of liquid biopsy areas, because it holds so many promises in terms of real-time disease monitoring.”
And ditto for Bio-Rad, says Shelton. “We have sold well over a thousand instruments; at least half have been used for cell-free DNA work, and probably 20% to 30% are doing so consistently.”
“It is a very popular application,” she concludes.
Reference
[1] Sacher, AG, et al., “Prospective validation of rapid plasma genotyping for the detection of EGFR and KRAS mutations in advanced lung cancer,” JAMA Oncology, April 7, 2016.
Image: Shutterstock Images