By Deborah Borfitz
October 29, 2019 | The value of liquid biopsies in precision medicine, particularly for managing cancers where tissue samples from tumors and metastatic sites are hard if not impossible to obtain, was a major topic of conversation at the 2019 Next Generation Dx Summit. Among the contenders was nine companion diagnostics (CDx) approved by the U.S. Food and Drug Administration (FDA) that look for cancer cells, or pieces of DNA from those cells, in the circulating blood.
Tests that look for circulating tumor cells (CTCs) and circulating tumor DNA (ctDNA) are enjoying a “resurgence” in clinical trials because they’re convenient surrogates for tissue, and samples are easy to obtain with an approved device, according to conference speaker Jonathan Beer, director of disruptive technologies in oncology precision medicine at Novartis. Key challenges with liquid biopsies in general are that they’re mass-limited and, in the absence of detectable ctDNA, require reflex testing with a standard biopsy sample.
As first reported by Stanford University researchers back in 2011, the sensitivity of liquid biopsies would need to improve 10,000-fold to detect the release of cancer antigen 125 early enough to positively impact clinical outcomes, says Yong Zeng, associate professor of chemistry at the KU Cancer Center at the UnivDiaersity of Kansas. Zeng presented at the summit on the 3D-nanoporous herringbone (nano-HB) microfluidic chip that is significantly improving the early-detection odds by scouting for circulating exosomes.
BioFluidica’s approach is to program microfluidic chips to isolate any kind of diagnostically relevant oncological markers, including cell-free DNA (cfDNA), exosomes and CTCs, according to keynote presenter Steven A. Soper, professor in chemistry and mechanical engineering at the University of Kansas, and director of the NIH Biotechnology Resource Center of BioModular Multi-scale Systems for Precision Medicine. Soper is inventor of the technology and chief technology officer at BioFluidica, which is now clinically validating its platform in blood-born cancers.
With liquid biopsies, isolation of molecular information is critical because of the complexity of the samples and the minimum number of targets, says Soper. “That’s why microfluidics is an attractive platform.” With BioFluidica, 2 milliliter (ml) or 10 ml of blood can be processed in less than 40 minutes, with no clogging. The enrichment process allows cells to be cultured for up to 72 hours.
Soper focused his talk on exosomes, the best characterized extracellular vesicles (EVs). They’re shed from almost all cells in bodily fluids and contain lipids, proteins, and nucleic acids. They’re also quite small—ranging from 50 to 2,000 nanometers—so they don’t carry much genetic material, he adds. Using a polymerase chain reaction assay for CTC detection masks epigenetic modifications and with cfDNA the number of tumor cells carrying mutations is too low.
Better Capture Rates
When it comes to teasing out clinical information about targeted subpopulations, “microfluidic enrichment is the most important part,” says Soper, noting that the amount of generated material in a sample is usually limited to between one and 100 CTCs per milliliter (ml) of blood. Key attributes of BioFluidica’s process are that it gets high CTC recovery, can isolate any target based on antibody and leaves no white blood cells behind.
The platform was developed in plastics, so it has good optical properties, Soper adds. Blue rather than ultraviolet light is used to release captured cells and minimize oxidation damage.
BioFluidica employs a pair of selection biomarkers—one for fibroblast activation protein alpha (FAPa) and the other for epithelial cell adhesion molecule (EpCAM)—for immunophenotyping CTC subpopulations and the clinical yields are nearly 100-fold higher than those of the FDA-approved CellSearch system, Soper says. The dual selection approach offers notably better sampling statistics for metastatic pancreatic cancer, capturing about 51 EpCAM CTCs per ml of whole blood versus 1 CTC per 7.5 ml for the CellSearch system, which is not approved for this indication.
The FAPa and EpCAM chips are fabricated via injection molding, “the way a plastic bottle is made,” and cost $2 per chip to produce, Soper says. The BioFluidica platform is fully automated and can run eight chip assays simultaneously and up to 32 per day, with fluid handling robots to speed things up. That’s more than five times the throughput of the manual method. No pre-processing of blood is required.
Most EVs can be isolated and the BioFluidica system maximizes their traditionally low capture rates with a bed design using diamond-shaped pillars that enhance biomarker interaction with the capture channel surface, programmable for different biomarker types, Soper says. But the molecular signature of EVs can differ from their cell of origin. Research will determine if diseases associated with EVs express differently, possibly using polydiacetylenes (PDAs) for their selective detection.
Novartis’ Beer says he likes liquid biopsies because they provide strong evidence on early treatment response, making them the basis for determining disease progression and changing the course of therapy—as well as patient candidacy for a clinical trial. They’re an important tool in precision medicine and could help shorten drug development timelines.
One concern with liquid biopsy tests is the associated risk of giving therapy to people who may not benefit from it. The first liquid biopsy test approved for breast cancer looks at ctDNA levels in patients’ blood and has high NPA (negative percent agreement) and low PPA (positive percent agreement) with the gold standard tissue sampling method, says Beer.
The second approved liquid biopsy test, for different indications, likewise has low sensitivity. ctDNA detection rates also vary by cancer type since they shed tumor cells into the bloodstream at different rates, he adds, noting lung cancer’s high shedding rate.
Negative results from a ctDNA test may be prognostic if a treatment has been identified, says Beer, citing its ability to predict survival in patients with melanoma. ctDNA can’t be detected in some patients, limiting its utility for monitoring treatment response.
Single-gene assays that look for specific diseases are particularly useful, at least if detected variants are associated with targeted therapies, Beer says. Large gene panels offer a higher probability of identifying clinically relevant mutations and other potential courses of therapy. A next-generation sequencing CDx test using plasma specimens is “on the horizon,” he adds.
Just last year, Foundation Medicine announced that the FDA had granted a Breakthrough Device designation for its new liquid biopsy assay that includes more than 70 cancer biomarkers. If approved, it will be the first liquid biopsy test approved for U.S. marketing that incorporates multiple CDx.
Earlier this year, the FDA granted breakthrough designation to ArcherDX for its sequencing-based companion diagnostic assay (for both liquid biopsy and tissue specimens) designed to diagnose patients with advanced non-small cell lung cancer. It was more recently reported that the FDA designated Natura’s surveillance tool Signatera, which analyzes tumor mutations in patients with certain cancers, as a Breakthrough Device.
GRAIL shared that its blood test reliably detects 12 deadly cancer types at early stages—including where it originated in the body—with a single blood test. And last month, Foundation Medicine and Natura issued a statement indicating they would jointly develop and commercialize personalized ctDNA monitoring assays.
Novartis also now has FDA approval to market its breast cancer drug Piqray, the first-ever treatment specifically for HR+/HER2- advanced breast cancer with a PIK3CA mutation, Beer notes. It was concurrently approved with the therascreen companion diagnostic test of QIAGEN.
Working With Exosomes
Liquid biopsies will not completely replace tissue biopsies, says Zeng, but will instead act as a surrogate of tissue for early detection and intervention, periodic screening and to guide personalized treatment. His goal is to improve methodologies for early detection, currently the most effective treatment for breast cancer. While the five-year survival rate can be quite good, when diagnosed at stage 4 the odds of survival is less than 25%.
EVs are an “emerging dimension” of liquid biopsies, Zeng says, and they have analytical query advantages that include an enriched source of the original cell content—for improved sensitivity—and reside in cells rather than shed into the blood—for enhanced specificity. Overall, they’re “stable and protective” carriers of molecular cargo in biofluid.
In addition to their low abundance, challenges of working with EV-based tests are complex biofluid matrices, interfering vesicles and subtype heterogeneity, he continues.
For isolation and enrichment of circulating biomarkers, ultracentrifugation is the best approach, says Zeng, although admittedly it can be tedious and time-consuming. Sample preparation and assay times can also get lengthy.
Precision measurements with sample-to-answer microfluidic biochips have been developed to isolate tumor EVs from biofluids, which was the focus of Zeng’s talk at the summit. Detecting cancer with the nano-HB chip Zeng and his team developed at the University of Kansas addresses the main hurdles with surface-based biosensing—mass transfers, surface reaction, and the boundary effect.
A “micromixer” helps overcome the mass transfer problem, a technical challenge of placing tiny samples into microfluidic channels, Zeng says. Exosomes could easily pass through undetected were it not for the 3D herringbone structures creating a whirlpool-like effect that sweeps them into contact with biosensors on the surface of the chip.
The herringbone structures also increase the surface area within the nanochip, he adds. The fluid drains out, further encouraging exosomes to reach the biosensors.
Hydrodynamic resistance at the liquid-solid interface creates the boundary effect, which the chip reduces to promote object-to-object contact or a “lubrication effect” between particles and surface, says Zeng.
Inspiration for the nano-HB chip came from a nanostructure fabricated by researchers at Harvard-MIT, Zeng says. “3D nanoengineering is simple, fast, low cost, designable and scalable.” Pore size is also adjustable.
As reported in Nature Biomedical Engineering earlier this year, the nano-HB chip has a better-than-80% recovery rate of exosomes of various cancer cell lines—outperforming standard isolation methods. It also correlates well with six biomarkers for ovarian cancer, making it potentially useful for exosome profiling to improve stratification and staging, Zeng says.
To scale up, an ExoProfile chip has been nanoengineered for detection of a panel of surface protein markers on exosomes to “get at stage of disease with 100% accuracy,” Zeng adds. The chip is made using ink-jet printing and features a trio of peristaltic pumps allowing the interrogation of eight biomarkers at a time.
Zeng says he’s excited by the technology’s potential in cancer diagnostics and precision medicine. The possibilities include assessing the activity and expression level of the MMP 14 enzyme, an indicator of the invasiveness of tumor cells, as well as tracking lung metastases in breast cancer.