In Search of the Perfect Test

Fast, effective, and affordable: an ideal triumvirate of characteristics for a diagnostic test. It may seem impossible to achieve all three in a single test – but I don’t believe it is. As a community of technologists, my peers and I have been working on rapid, low-cost biomarker disease detection for many years. Of course, the COVID-19 pandemic has brought the issue into the limelight – but there are many other diseases and use cases that make this technology valuable. For instance, we can glean a great deal of valuable information for diagnosing or monitoring cancer via a blood test to measure the concentration of circulating biomarkers. In many cases, the sooner you detect the presence of cancer biomarkers in blood, the higher the chance of survival and the more conservative the treatment options. And that’s why my colleagues and I decided to develop a technology that can detect biomarkers with single-entity resolution.

Once our test is ready for use, I believe it will change the lives of both patients and practitioners. Our goal is to empower patients to monitor their own health, reducing the number of hospitalizations without placing an additional burden on pathology labs. For instance, nurses who work with cystic fibrosis patients often travel miles to their homes just to take a blood sample for CRP testing (an early marker of inflammation that can signal a need for medical intervention). Why put nurses through hours of driving, laboratory medicine professionals through hours of work, and patients through days of waiting for a single result? I think point-of-care testing has incredible benefits to offer pathologists, patients, and health services as a whole.

How it’s done
Our discovery was a combination of community and serendipity. Scientists all over the world have been trying to use DNA structures as protein carriers because the former is much easier to detect than the latter – we just had the good fortune of finding a way to make it work.

We use nanopore technology to measure one biomarker at a time. Essentially, we run an electrical current across tiny holes (nanopores) and measure changes in the current as the biomarker of interest passes through. The beauty of the technology is that it’s completely electrical – meaning that it’s both portable and quick. We’ve even developed it as a USB-powered device that can be plugged in anywhere for on-the-spot biomarker testing. At the moment, the technology measures nucleic acids accurately – but, in the future, we want to expand the analysis to proteins as well. That will be a game-changer. Take cancer, for instance; you’ll be able to go to your doctor’s office, have your finger pricked, and get a biomarker test result in about 20 minutes. That will then guide the doctor on whether to order more in-depth testing.

Of course, point-of-care protein testing is not easy. Our approach marries nanopore technology with DNA origami (a way of using DNA as a “Lego brick”). We build a structure that looks like a picture frame in the center of which we can capture a protein biomarker. Capturing that protein significantly changes the electrical signal across the DNA origami frame – allowing us to measure the protein level in a sample.

Early detection difficulties
In the research stage, biomarker detection technologies often perform well – but, when it comes to translating them from a controlled laboratory to a real-world environment, complications arise. And we know the same may be true of our technology; it is a new approach, so we are trying to partner with hospitals and companies who can test it in a clinical setting.

Traditionally, validating a medical technology takes years – often a decade or more. Since COVID-19 struck, the world has completely changed and we’ve seen a lot of technology fast-tracked to the end user. Can we speculate that, in the next few years, we’ll see more and more of this speed-validating? Perhaps – but let’s not forget the complexity of the challenge we face. When we provide clinical information, we need to have complete confidence in its accuracy – and that means carefully evaluating, standardizing, and controlling our tests, and ensuring that they are within acceptable margins of error.

The key difference between our proposed test and others is that ours has no optical component – it’s all electrical. Imagine the miniaturization – and the power – of a smartphone applied to a biomarker testing tool. Because we can create a tiny device with integrated wireless connectivity and data analysis, we can turn a time-consuming laboratory procedure into a point-of-care test. This vision is further helped by the fact that our testing requires only a few microliters of blood – so not only is it minimally invasive (a fingerprick suffices), but it also means we don’t need expert assistance from a nurse or phlebotomist. Finally, we measure biomarkers at the single-molecule level – one protein at a time – and build a signal by counting those proteins. Unlike a test that requires millions or trillions of individual proteins, this approach has the advantage of high sensitivity.

From concept to reality
To measure a specific protein, we start with a blood sample and pre-processing – at a minimum, to isolate a plasma sample. Then, we incubate the processed sample in a vial of our DNA origami to allow the biomarker to bind to the “frame.” After incubation, we place the sample into a cartridge with a nanopore filter; then, the cartridge goes into a machine roughly the size of a book, which measures the electrical current. Within one or two minutes, we have the raw data we need. Finally, we analyze the data offline. Total time? About 35 minutes – perfect for rapid or point-of-care diagnostics.

The catch is that each protein biomarker needs its own specific, custom-designed DNA origami – so there will be different vials for the different “frames.” Moreover, our technology currently detects one protein at a time, but for many diseases a single biomarker is not enough for clinical decision-making. We’re working on measuring multiple proteins at once by combining different DNA origami “frames” into a single vial, but I anticipate it will take us another five years to get there. And that’s why we need to ask ourselves if we truly need to enable the detection of multiple markers. Are there any applications for which just one would be useful – and would those applications be commercially viable? After all, we will need funding to support the development of our technology. We’re talking to clinicians to find out what they need most.

Our current prototype is not optimized for routine clinical use – something we realized pretty much immediately. We are now working with industry partners who can help. For instance, we recently joined forces with a company that provides disposable cartridges for digital measurement – an approach that will be much more suitable in the clinic. Single-use equipment means that every new measurement would be in a completely sterile environment. Our initial experiments look promising, but now we need to validate the test for clinical use.

The future of biomarker research
These days, everyone is talking about liquid biopsy. We typically use liquid biopsy to analyze nucleic acids circulating in the blood, helping us detect and diagnose certain diseases. But why focus on nucleic acids when so many technologies already measure DNA and RNA? In my opinion, it’s time to focus on proteins. Our approach to protein analysis is an unusual one; rather than examine the sequence that underlies the protein, we simply look for its presence.

Fundamentally, the presence of an unexpected protein – or the absence of an expected one – tells you that something is wrong. Nucleic acids tend to provide information about risk factors for disease; you might never develop the disease itself. Protein detection, in contrast, allows you to intervene when there is an active problem – and the earlier we can detect a protein biomarker, the better the outcomes (and the lower the stress) for the patient.

Another key is engagement with medical professionals. We need our clinical colleagues to co-design assays and devices with us to ensure they are accessible and easy to use. And that means we need to involve everyone – from nurses to consultants – so that anyone who might be using the technique has the opportunity to inform its development. As engineers, we’re very good at overcomplicating things, so it pays to get as much input as possible from end users. Only by involving everyone in the process can we create a test that is beneficial for both professionals and patients. And, after all, that’s what diagnostics development is all about.