A Smart Approach to Testing

Did you know that the average person spends over four hours per day on their mobile phone (1)? And although smartphones are becoming indispensable tools for business and entertainment, they also have the potential to transform the way we interact with the health system. With computers in our pockets, it should be easy to get more immediate, more convenient, and lower-cost access to doctors. Infectious diseases are one application where getting immediate results from a test, rather than waiting several days for samples to be sent to a lab and cultured, is very important. Speed can limit the spread of a virus among animal or human populations. It can identify a dangerous bacterial species in food preparation or in the hospital before it causes an epidemic. Infectious diseases also represent one of the leading causes of death in resource-limited parts of the world. In those places more than anywhere else, an inexpensive point-of-care test can have an enormous impact.

The chip and the cloud

Nucleic acid (DNA and RNA) testing is the key to rapid detection and identification of pathogens. We no longer have to culture each microbe and wait for it to grow to a point at which we can make a morphologic diagnosis; now, we simply take a look at its sequence to deduce both what it is and – in some cases – the best way to treat it. In our research, we showed that the conventional laboratory test for nucleic acid amplification of pathogenic DNA (or RNA) can be performed in a silicon chip – and that a single chip can perform tests for up to eight separate pathogens in a single droplet of fluid without severely compromising the detection limits of the test. Our system uses the phone’s internal camera to gather a fluorescent image of the chip and its internal microprocessor to interpret the image and give a validated result using integrated experimental controls (see Figure 1). It also integrates a local smartphone app with a cloud-based service system that combines the results of the test with the patient’s other medical records. And that carries an added benefit: it enables epidemiological interpretation when results are gathered from a distributed network of users taking the same test.

We began our work by developing an endpoint (yes/no) test for a set of equine respiratory diseases (2). Our goal was to demonstrate how the system could be used for racing animals, food animals, and companion animals before moving to human applications – partly for safety reasons, and partly because veterinary medicine represents a market with fewer regulatory hurdles than human diagnostic testing. We also wanted to contribute to that field because, through collaboration with a practicing equine vet, we learned that there is a substantial unmet need. The test was successful enough that a company has now licensed the patents and pending applications and is planning development of a commercial product for both animal and human applications.

Meanwhile, we developed a subsequent assay for four human viral pathogens (Zika, dengue 1, dengue 3, and chikungunya) using a single droplet of whole blood as the test sample (3), which enabled us to integrate sample handling with our test. We also incorporated kinetic monitoring of the chip so that we could estimate the concentration of the virus, rather than simply provide a yes/no output. One vital member of our team, a molecular biologist, has helped us develop and validate a set of selective primers for all the target pathogens, and is now advising us on the performance criteria that our test must meet to be equivalent to conventional laboratory methods.

Practical purposes

When might a test like ours come in handy? One example might be when an animal appears to be sick, but the cause is not known. Our test could diagnose whether the animal has one of the eight most common respiratory diseases – and, if it did, it could be immediately quarantined to reduce the opportunity for the infection to spread. Consider the case of food animals being raised in a facility with thousands of others; if one falls ill, it’s unfortunate, but if they all fall ill, it’s catastrophic. Imagine the housing of racehorses before a big event – even a single animal’s illness is a costly event, but it’s staggering when multiplied.

Eventually, we can envision tests like this being available in the drugstore, so that when you feel lousy, you can perform a test on yourself and have the results communicated immediately to your doctor. The doctor could then prescribe medication or another course of treatment based on your results, along with a video interview, your medical records, and other symptoms. You could avoid ever going to a clinic, waiting in long lines, or coming into contact with other people’s germs. That’s a nice thought for anyone – but even more of one for people who have difficulty attending clinics in person, such as those with limited mobility or those who are immunocompromised.

And, of course, the test’s benefits to humans aren’t limited to the doctor’s office. Other scenarios could include validation that food preparation surfaces and facilities are free of listeria or salmonella at fast food restaurants, testing for the presence of antibiotic-resistant bacteria on surfaces in the intensive care unit of a hospital, testing water at the beach for E. coli, or making certain that norovirus is not present on a cruise ship. There are even defense-related applications, such as the detection of biological warfare agents. What started as a simple test for respiratory diseases in horses has the potential to offer a whole new world of health and safety applications for everyday life.

The move to the clinic

We expect our chip to become a supplement to current laboratory methods, rather than replacing them completely. For instance, it could be used in rapid-response situations that might still require follow-up validation by a conventional laboratory test. Pathologists will be able to devise tests for common sets of pathogens in specific scenarios, and will be able to continually adapt and deploy modified tests as new strains of pathogens are identified. Our test is basically equivalent to the laboratory methods used now, but implemented in a lower volume format and using a smartphone camera as the detection instrument. I expect that pathologists will want to contribute to developing more sophisticated experimental controls and image processing methods to improve the validity of test results, and they will certainly be involved in interpreting the results that will be delivered by a distributed network of instruments. Overall, by lowering the cost and convenience barriers for performing a test, pathologists will likely become busier as the populations they can serve expand.

At the moment, the chip is not yet ready for clinical deployment. We have a working laboratory prototype developed by a team of professors, graduate students, and a veterinarian. We are licensing our patents and pending patent applications to enable the device to be commercialized and mass-produced. In the meantime, we’re also working on two more aspects of development: ways to integrate more of the sample handling, so that the test can be more fully automated, and image processing approaches and engineering designs that will hopefully enable us to surpass the detection limits of conventional methods.