NEW YORK (360Dx) – The promise held by microfluidic, cell-based molecular analysis for clinical applications has motivated extensive research and development, but fully-integrated systems that consist of modules developed for the various parts of the diagnostic process are at least a few years away from commercial launch and will require more development, some experts believe.
Only a few examples exist of fully-integrated microfluidic devices that have been translated from development to the point-of-care, Dong Jun Lee, a researcher in the department of biomedical engineering at Duke University in Durham, North Carolina, said in an interview. Future research, he added, will need to focus on integrating modular components, which have been developed to address each of the main steps in the cell-based MDx process.
Lee along with coauthors Tony Jun Huang, a researcher in the department of mechanical engineering at Duke, and John Mai, a researcher in the Alfred E. Mann Institute for Biomedical Engineering at the University of California in Los Angeles, recently published a summary of recent advances in modular approaches toward the commercial launch of fully-integrated, cell-based molecular diagnostics for clinical point-of-care applications.
They noted in their paper, published in the journal Biomicrofluidics, that although researchers still have their work cut out, microfluidic technology is capable of efficiently performing all steps of the cell-based molecular diagnostic process — cell isolation, lysis, and DNA extraction, amplification, and detection.
But to move the field of microfluidic cell-based molecular diagnostics forward, "each integrated component must be a master in its own right, [and] the whole needs to be greater than the sum of the parts," said Cesar Castro, director of the cancer program within the Massachusetts General Hospital Center for Systems Biology. "Like any good invention, the key ingredient needs to be planning. All stakeholders from various disciplines need to be equally involved."
He said that planning "not only means carrying out actual experiments but also significant modeling and other in silico approaches."
Castro, not an author of the paper, is working with his colleagues at MGH and other researchers to develop a disposable cartridge-based microfluidic assay for the detection of lymphoma.
"The motivation for cell-based MDx is to achieve all-in-one systems — a specific niche within a larger field that leverages microfluidic approaches for each aspect of the cell-based diagnostics process," Castro said. "Innovations in each aspect does not guarantee integrated success, and synergies between disciplines are needed to achieve true integration and improved efficiencies."
Advances in microfluidic technologies raise the possibility that clinical assays can one day be performed in an automated, sensitive, precise, cost-effective, and high-throughput manner that's only possible now with expensive and complicated benchtop instruments. Researchers noted that the microfluidic-based modular approach to genomic analysis from sample preparation to the extraction, amplification, and detection of nucleic acids has matured significantly over the past two decades.
Overall, traditional analysis of cells as carriers of disease — using technologies that are not microfluidic — is well established as are several cell-based diagnostic methods, ranging from culturing infectious disease organisms to using flow cytometry to count and analyze organisms for life science research and clinical applications.
Traditional cell-based diagnostics also has large total addressable market segments. On Thursday, Luminex announced that it has agreed to acquire MilliporeSigma's flow cytometry portfolio for $75 million, entering a life sciences research market worth about $1.6 billion and considering a future entry into the use of flow cytometry for clinical applications worth an additional $2 billion.
Fluorescent-activated cell sorting (FACS) is among the most prominent traditional technologies for cell separation and has some of the highest throughput and sensitivity, Lee said, adding that microfluidic technologies in development can't be compared to FACS because of the latter technology's superior level of performance and commercial availability for clinical applications.
"Leveraging recent advances in microfluidics, many biochemical assays have been translated onto microfluidic platforms, and various types of microfluidic technologies are being developed that could be good alternate options to FACS in the future, but they have to first overcome many challenges," Lee said.
To understand the challenges ahead for researchers, he said, it's necessary to study the many technologies being developed for cell isolation, lysis, DNA extraction, amplification, and detection.
For cell separation, Lee said, inertial flow microfluidics — which leverages the interaction between the flow of sample fluids and suspended particles — is well defined and easy to implement, but the method currently lacks the required level of purity, the ratio of target cells to the total number of cells that have been separated.
Dielectrophoresis — also used for microfluidic cell sorting and involving the movement of particles in a non-uniform electric field — has high purity but lacks enough throughput for practical applications. Acoustofluidics uses an acoustic field for cell separation and shows throughput and purity between that of inertial flow and dielectrophoresis, said Lee, who is developing cell-based diagnostic devices with colleagues using an acoustofluidic approach.
Although many of the challenges have been worked out for use of microfluidics to perform lysis, DNA extraction, and amplification, there's room for improvement not only with cell separation but also with detection, he said.
In microfluidics detection, researchers are wrestling with a problem of non-specific binding. "When we are trying to use a complex sample such as whole blood which consists of many different components, current microfluidic technology generates many non-specific signals, and we don't have good passivation technology to prevent that from happening," Lee said.
Most detection technologies have a sensor surface onto which the target molecules need to bind. A signal is generated because of a binding event, but developers don't want molecules other than the target molecules binding to the surface.
Researchers are looking for surface coatings to prevent binding. However, current efforts to mitigate the effect have been unsuccessful, preventing the required levels of specificity and sensitivity needed from being met, as well as commercialization and adoption, Lee said.
For detection, PCR-based technologies work well and are highly sensitive, but they are not yet integrated well with other upstream processes, he said. Electrochemical detection has potential to achieve high sensitivity, but it encounters challenges associated with non-specific binding that must be addressed.
In a general challenge faced by developers of microfluidic devices, most cell-based MDx systems have "only been demonstrated with simplified samples containing known cells or DNA fragments spiked into controlled buffers, and it's unclear whether their performance would be downgraded if more clinically relevant samples were tested," the researchers said in their paper.
Etvi Juntunen, chief technology officer at Inme and a researcher of point-of-care diagnostic technologies at the University of Turku in Finland said, "In microfluidics, the common finding seems to be that everything works well with water samples and buffers, but real clinical samples with all the matrix effects, variable viscosity, and general inconsistencies" present challenges. Overall, although new innovations are frequently emerging, the number of commercial products based on microfluidics pales in comparison, he said.
Paul Yager, a professor in the department of bioengineering and a leader of microfluidic test development at the University of Washington, said in an interview that many early microfluidic devices that have been developed into proof-of-concept systems fail for many reasons when they are confronted by real biomedical samples due to their complexity and variability.
Whole blood consists of plasma and a mix of cells — including red blood cells, white blood cells, lymphocytes, neutrophils, monocytes, and platelets — as well as circulating tumor cells, circulating DNA, and exosomes, and all are of interest for diagnostic purposes.
Castro noted that microfluidic diagnostic systems are hampered by matrix effects coming from different clinical specimens, biofouling, and varying targets that yield challenging detection limits, as well as other pitfalls. "A case in point is that even if we kept the specimen consistent [and used peripheral blood, for example], the viscosity of blood collected from cancer patients tends to be greater than the viscosity of blood from healthy volunteers. So, the true challenge is to develop systems that can accommodate such diversity," he said.
As a counterpoint to the skepticism surrounding performance degradation in real samples, several commercially available and cleared point-of-care microfluidic devices use whole blood samples for blood chemistry assays, and their performance has been validated using clinically-relevant whole blood samples.
The US Food and Drug Administration has already cleared several point-of-care tests that employ different approaches from cell-based molecular diagnostics. These tests evaluate blood chemistries from whole blood and have microfluidics at the heart of their operations. They include, for example, the Abaxis Piccolo Xpress chemistry analyzer, which is exclusively marketed by Abbott in the US, and composed of three ultrasonically welded plastic parts that contain the diluent, dry reagent beads, and barcode. Clinicians can perform a complete panel of up to 14 chemistry tests on-site in about 12 minutes.
Quidel's FDA-cleared Triage MeterPro point-of-care system, meanwhile, is a capillary-based fluorescent immunoassay in which analyte molecules in the sample react with labelled antibodies to create fluorescent analytes that can be quantified by the amount of light that they emit when they are excited with a laser. The Abbott i-Stat and Siemens Healthineers Epoc are cleared blood analysis systems are biosensors with advanced microfluidics.
In the growing area of capturing circulating tumor cells and circulating tumor DNA by traditional means, the CellSearch CTC Kit marketed by Menarini Silicon Biosystems has been clinically validated and FDA-cleared for identification, isolation, and enumeration of circulating tumor cells uses ferrofluid-based capture technique and immunofluorescent staining of the CTCs from whole blood.
The number of microfluidic approaches for molecular diagnosis of cell markers is expanding, Neelkanth Bardhan, a microfluidics technology researcher at the Massachusetts Institute of Technology's Koch Institute for Integrative Cancer Research, said in an interview.
Along with colleagues at MIT and National Chiao Tung University, he is developing a single-cell capture device that could eventually lead to a point-of-care diagnostic test for infectious diseases or cancers, cost less than $5, and produce results in less than 10 minutes.
The payoff for working out the kinks in microfluidic system design is potentially high not only for researchers but also potentially for patients, such as those with infectious diseases and cancers, that could benefit from cheaper, more accessible, and more broadly spread diagnostic systems.
Some researchers believe that the use of microfluidics in the analysis of exosomes holds particularly promise, and that it could be a way to help clinicians overcome vexing clinical challenges associated with heterogeneity in which different tumor cells can show distinct morphological and phenotypic profiles, making the cells more difficult to diagnose and treat.
Exosomes are 30 to 150 nm-sized vesicles secreted by virtually all types of cells. They contain proteins and DNA/RNA markers that can be correlated to the condition of their originating cells.
Researchers are already developing microfluidic techniques to better analyze the stochastic, or random, behavior of cells within cancer tissues, which is associated with their heterogenous properties. In October 2017, a team of industry players and a Japanese government agency announced a collaboration leveraging exosomes to improve cancer diagnosis and treatment.
Further, researchers in South Korea are developing a lab-on-a-disc that incorporates centrifugal force and nanoporous membranes to identify exosomes that carry potentially important biomarkers of cancer in urine.
Developers at Harvard Medical School and Massachusetts General Hospital, including Castro, are developing an assay that leverages low blood volumes and the clinical potential of exosomes. Their multiplexed nanoplasmonic assay is being developed to fit into clinical workflows in the detection of pancreatic ductal adenocarcinoma — an aggressive and often inoperable form of pancreatic cancer.
"There is high promise for leveraging cell-based MDx for exosome analyses," Castro said. "The jury is still out, however, on whether single exosome analyses can make significant inroads into the therapeutic space. That’s far down the line. But what could be of more practical use is a cell-based MDx approach that can discriminate cancer exosomes from host exosomes."
In the microfluidic center at Duke, Lee and his colleagues are developing a digital acoustic technique that enables trapping a single exosome within a droplet for further analysis.
He noted that the microfluidic system would use RT-PCR for detection. Such a system could be fully developed and ready for launch within about three years, he said.
"While there can certainly be an argument for the use of microfluidics for exosome analysis owing to their relative abundance compared to circulating tumor cells, one must be cognizant of the challenges involved in defining what exosomes are, in particular, and how to identify them," Bardhan said. "Not all types of cells secrete exosomes, and even if they do, the science concerning exosomes is not fully developed or understood."