Hundreds of thousands of people are being tested for SARS-CoV-2 (COVID-19) around the world and the tests themselves are a crucial starting point in efforts to flatten the spread of the virus.
In a recent episode of Thermo Fisher Scientific’s Science with a Twist podcast, we talk with Senior Director of Genetic Sciences, Joshua Tratta. Tratta explains how tests are created, how they’re performed and how they’re analyzed.
QUESTION: Prior to the CDC announcement that they have developed a test, there was this kind of race to develop a test. Talk about what it is that encompasses the time that it takes to develop a test. Why does it take so long? What is the process you have to go through to develop a test?
ANSWER: Most outbreaks, and this one is no different, you start with asking the question, ‘Why are people getting sick? We don’t understand.’ So something happens. In this case, someone was getting sick and people didn’t understand why. The first thing that most outbreaks do is use Next Generation Sequencing technology to globally look at the sample and say, ‘What is it? Is it bacterial? Is it viral? Is it fungal? Is it something we have no idea what it is?’ And it allows you to start narrowing down what it is.
Once we’ve done that, as is the case of the Chinese government, they actually published their NextGen sequencing results, which allowed people to actually see that it was in the coronavirus family.
Not only was it a coronavirus, compared to like SARS, which is also a coronavirus – it’s a novel coronavirus. So they kind of knew what it was. They’ve narrowed it down, but there’s something new about this particular virus.
So what you have to do once you identify that, instead of running that big broad screen over and over again, which can be cost prohibitive for trying to ramp up to millions or hundreds of thousands of people. People move into designing specific qPCR assays to detect just that area which is now identified.
So to do that we use something called bioinformatics tools. It’s designing specific genetic sequences and assays that cover that region. Then, unlike some of the other technologies, qPCR can give you the results in a two-to-2.5 hour timeframe at a reduced cost. And because it’s been a technology that has been around over 25 years, a lot of people are using it for other things. For other viruses. For other bacteria. For research applications. So the prevalence of instrumentation is out there. It’s something you can now push more broadly to people that already have these instruments and can use them to specifically look at something like the coronavirus.
QUESTION: What exactly is an assay?
ANSWER: An assay is a genetic tool that actually helps you identify a virus or a bacteria. Being able to match up that genetic material in one piece versus another that says you actually have this genetic information. ‘I have compared you against this test.’ Or you don’t.
QUESTION: You provided me with an example using Lucky Charms cereal. Can you, using the Lucky Charms analogy talk about how the coronavirus may or may not be present in my Lucky Charms.
Hopefully you don’t have coronavirus in your Lucky Charms. But using it as the analogy… within Lucky Charms you have multiple types of marshmallows. So imagine that the coronavirus is the rainbow marshmallow. Your cereal bowl is your genetic makeup. An assay would try to detect the rainbow. So when you apply that, you would take rainbow – which is the qPCR assay – and you would look over your bowl and say, ‘Are there any rainbows? Are there any rainbows in my cereal?’. And if there was a match, you’d have coronavirus. If not, you’d say ‘I don’t have the coronavirus.’ Doesn’t mean there’s not something else that makes you sick but you can say you don’t have the coronavirus.
QUESTION: I know that qPCR has been around for a long time. What is qPCR? How would I describe qPCR or explain it to someone who doesn’t know what it was?
qPCR has been around for 25 years. It’s a detection methodology that is akin to identifying using genetic information to identify bacteria, viruses, fungal targets. Historically people have used culture for many years. Which is still a robustly and routinely used technology. But in this particular case it allows you to do something faster. It’s being able to use genetic information to identify a pathogen in a very short amount of time.
QUESTION: How does the virus’s genetic code factor into the development of the test?
The genetic code plays into the test. You want to be very specific about identifying just that virus.
So when you develop the test you need to really understand the genetic makeup of the virus. So that you can create the most specific test. Just like for you or for me, I don’t want you to tell me a test that says, ‘Oh I’m only 30% sure you’ve got the coronavirus.’ I want you to give me an answer so I can feel confident behind it.
That’s what we call specificity.
So you want something to be specifically targeting something you’re concerned I may have. And if you’re not sure, and you have a lower specificity, it means that you could be entering a quarantine with somebody when you shouldn’t have been in quarantine.
I want to make sure we are understating the virus, knowing the details will help you design very specific assays for detection.
QUESTION: Everyone that has a respiratory infection doesn’t necessarily have the coronavirus. Talk about how technology enables you to differentiate between the different sources of infection using, for example, a multi-panel test.
So if you look at the symptoms associated with the coronavirus, how do you differentiate between someone who has the flu – that’s pretty prevalent and spreading across the globe right now – or if you have the coronavirus? Or if you have some bacteria? The beauty of technologies like qPCR, we have something that’s called a syndromic approach. Which is having multiple targets that you can design for and test for them all at once. So in a case like for what Thermo Fisher has designed, we have a test that can test for 42 different targets. And they will look at the most common bacterial, viral and fungal targets that cause respiratory illness. Imagine you walk in and somebody suspects that you have the coronavirus, on that test is something that will say ‘yes or no – you have the coronavirus.’ If the answer is no, wouldn’t it be great to say that you actually have a bacteria, and there’s a specific antibiotic that will be used to treat you. Or that you have flu, and it’s just flu A. They’re going to tell you to go home, get some rest and drink some fluids.
QUESTION: Talk about the importance of detection and the importance of testing.
I’d say detection is the beginning of ensuring that we don’t spread the thing. Detection is the starting point. Knowing how to make good decisions from that point.
QUESTION: We talked about a little bit about the sample – it’s a nasal swab/throat swab. If you could just help us understand how you get from swab to test result, that would be really helpful.
Imagine a swab – for any of our listeners who don’t know what that is – imagine a big Q-Tip that is long. For any of you who have had the unpleasant experience of going to your doctor because you thought you had strep throat. Same approach. They’re going to take a sample from your throat or nose and pull out what seems like nothing. But it will have material on that which that material is what we want to extract to get your DNA. So being able to get that DNA extracted, Thermo Fisher Scientific also makes kits to be able to do that. We have one that is focused on viral pathogens and extracts the DNA. That’s kind of the starting point. We got to get the DNA so we can go back to our Lucky Charms example and do some comparisons of understanding if the rainbow is there. DNA extraction works in that way. And then you will actually move over into using multiplex assay. So you combine those pieces together with a single tube sample that they’ve taken and some other reagents and you put it on to an instrument. Let it run for a few hours and a result will come back that will say the lucky charm rainbow is there or it is not.