Pools are opening around the world – and I’m not talking about the kind you swim in! Instead, I’m talking about “pool testing” for the novel coronavirus, SARS-Cov-2, which causes the disease COVID-19 (COVID stands for COronaVIrus Disease discovered in 2019 in case certain people, who I very much doubt are reading this post, need reminding…). Diagnostic coronavirus tests check if people are currently infected (and potentially infectious), typically by looking for copies of the virus’ genetic information in a person’s nose fluids (which you can collect with a swab), and pool testing is basically where you take several (e.g. 5 or 10) people’s nose fluids, combine them, and test them as a single sample. If the test comes up negative, you’re done & saved a bunch of time not running each test individually. But if the test comes up positive, you then have to test each sample individually (which could mean doing more tests and taking even longer than if you’d tested each individually to begin with). So it’s a cost-benefit/risk-reward type of situation that works well for mostly-negative populations with low consequences if you miss a true positive (think a large warehouse or college campus where no one is showing symptoms), but works badly for populations where there are many cases and/or situations where letting a positive slip through can be catastrophic (think a nursing home).
Pool testing isn’t a “new” idea, but I heard saw it talked about in the news recently because the head of the US National Institutes of Health (NIH) said the White House task force is “seriously considering” using it. https://cnn.it/31nvGqo My dad heard this too and, when I was on the phone with him & my mom, he asked me my thoughts about pool testing. My mom heard the word “pool” and got excited because she thought he was talking about swimming pools, so I’m sure my response was a bit of a let down, but once I got to explaining it, she got interested (I think) and even got engaged in the conversation. And, especially after what I’m going to tell you about next, I decided I definitely should make a post on it.
After going through the basics of what pool testing is, and how it’s good for mostly-negative populations, my mom said something like, “so it’d be good for places like the convalescent home where your aunt works.” My response: “NOOOOOOOOOOO!” Okay, I definitely didn’t say it like that, because her thought process makes sense from the testing cost standpoint, but it misses the cost (financial and potentially even life) if a positive case goes undetected. And a positive case is more likely to go undetected with a pool test than with a traditional test because when you pool samples, you dilute each individual sample, making it harder to detect, especially if you start with a low amount of viral information. (in her defense, I hadn’t gotten to this part of the explanation yet, and I really don’t “judge” her at all for her confusion, and love her very much).
The viral information these tests are looking for is copies of the virus’ genome (genetic blueprint). Basically, SARS-CoV-2 is a little fat sack containing a single strand of RNA that serves as the virus’ genome. Embedded in the fat sack (lipid membrane) are proteins like the Spike protein that juts out and lets the virus dock onto human cells (by binding to the cells’ ACE2 receptors) and (after some help from cellular proteases (protein cutters)) fuse with the cell’s membrane to release its RNA into the cell. Once inside, the virus hijacks the cell’s machinery to make viral proteins based off of the viral genome’s instructions and uses these proteins to make more copies of the virus & ship them out to infect more cells inside the person (and more people outside the person).
The SARS-CoV-2 genome is big if you compare it to other RNA viruses (it’s ~30,000 RNA letters (nucleotides) long), but it’s tiny if you compare it to our genome (which is about 3 billion letters long). So if we want to find the viral genome, we need to amplify its signal, which we can do by making lots of copies of part of it in a tube or plate in a lab through a process called PCR (Polymerase Chain Reaction) – don’t worry, they’re not copying the entire virus, just a little segment of the viral genome, so this test isn’t “dangerous.”
PCR tests use a set of short pieces of DNA called primers to tell a protein enzyme (reaction mediator/speed-upper) called DNA Polymerase which region of the viral genome to copy. These primers target a region that is specific to this virus’ genome (i.e. sequences that aren’t found in our own genome or that of any other virus or bacteria), so it will only make copies if the virus is present. And then fluorescently-labeled DNA probes can bind to those copies to show that copies are being made, indicating that the virus was present.
Problem is, this DNA Polymerase makes DNA copies of DNA. And the virus has an RNA genome, so first you have to use a “Reverse Transcriptase” to make a DNA copy of the RNA which the DNA Polymerase can then make DNA copies of. This Reverse Transcription step adds a couple of letters onto the technique name (i.e. PCR becomes RT-PCR) but it doesn’t add that much time to the testing process. The part that *does* add a lot of time is the “RNA extraction” step, where you isolate all the RNA from the nose fluid before you do the RT-PCRing. This part is harder to automate than the other steps (which can easily be done in a plate format where you can test hundreds of samples at once) and it takes chemical mixes that are in high demand, so pooling the samples before this step can *potentially* save a lot of time and resources.
But there are some serious asterisks around potentially because pooling is *not* always a timesaver, and it’s important to appreciate when it is and isn’t appropriate.
If someone *has* symptoms, they should be getting their own diagnostic test from the start. Say you have 1 person with symptoms and 4 people without symptoms. And you pool their samples & test them together. So that’s one test. And the test comes up positive. So now you have to go back and test each individually. So that’s 5 more tests. So you end up doing 6 tests to find 1 case. And you’ve taken almost twice as long as if you’d tested them all individually to begin with.
If, on the other hand, you test that symptomatic person’s sample by itself and the other 4 pooled together, you only have to run 2 tests to give you results for 5 people in a single testing process timeframe.
Another reason not to test the symptomatic person pool-style is that, as I mentioned above, pooling the samples dilutes them so, even though PCR has the potential to make a bunch of copies of the viral RNA, there isn’t enough there to “get the ball rolling.” If a symptomatic person gets told they’re negative, they might go about their business as normal, which can be catastrophic. And even an asymptomatic person going about their business as normal can catastrophic – especially if that person is working with vulnerable populations, like the elderly.
Speaking of asymptomatic… In the beginning of the pandemic, there was a great shortage of diagnostic tests, and only people with serious symptoms and/or a connection to a known case were able to get one. But, as tests became more widely available, scientists found that people could test positive for viral RNA but not experienced any symptoms – these people could either be presymptomatic (i.e. they just hadn’t experienced symptoms *yet* but they were coming) or they could be asymptomatic carriers (i.e. they never experience symptoms or their symptoms are so so mild that they don’t even realize). There’s still not that much information on the role never-symptom-having people play in spreading the disease, but presymptomatic people definitely can spread the disease – in fact, people are often most most infectious in the very beginning of the infection, before they even know they’re sick.
Part of the reason we don’t have that much info on never-symptomatic-people is that people who don’t have symptoms aren’t very likely to volunteer to have a stick stuck up their nose if they’re sure nothing’s wrong with them, so these people often go undetected. And it often doesn’t make much financial or time sense to test hundreds of people who have a very very low likelihood of carrying the virus. But these people can still be spreading the virus. And pool testing is one way to find them at a lower cost.
With pool testing you still have to get people to agree to testing (easier now that shorter swabs are becoming more commonly used than the original super long ones that go really far back – and I’m really hopeful about spit-based tests making testing even easier) but you don’t have to use as much time & resources (hopefully). And, as long as everyone keeps taking social distancing and cleanliness precautions, cases could be caught in time to prevent other people from catching the virus without having to shut down entire workplaces, etc. This could potentially prevent “superspreading events” where one person can infect dozens or even hundreds of people.
Another use for pool testing is a sort of population surveillance. Normally, “surveillance testing” has been used to refer to antibody testing which looks for evidence that someone has had the virus in the past. But, here we’re talking about testing for whether people have the virus “in the present” – right now. If you regularly tested random groups of people in a community you could potentially find early warning signs that predate the rise in hospitalizations and deaths that could follow if you don’t take action. https://bit.ly/2ZhCrHB
A last caveat: these tests look for viral genetic information, *not* “active virus.” It looks for the RNA without taking into account whether that RNA is surrounded by the lipid coat it needs to actually get into cells (or whether that RNA is even intact). So, basically, it’s possible that people could have “dead” virus in their nose, which seems to be the case with some people who did have the virus, but then recovered, yet still test positive for weeks and weeks because. they’re shedding “viral debris.” For this reason, guidance has gone from “you have to test negative twice before you can go back to work after infection” to “you can come if you’ve been symptom-free for at least 3 days and it’s been at least 10 days since your symptoms began. This debris-shedding is also believed to be responsible (at least in part) for those cases which got a lot of buzz where people tested positive after testing negative, leading some to worry that these people were getting re-infected which would be really bad from an immunity standpoint. Scientists “de-bunked” that by running more elaborate tests that showed that the virus was “un-culturable” – they couldn’t get it to infect other cells in a dish, so it likely didn’t stand a chance of infecting a person https://bit.ly/3g9kVMH
more Covid-19 resources: https://bit.ly/covid19bbresources
more on topics mentioned (& others) #365DaysOfScience All (with topics listed) 👉 http://bit.ly/2OllAB0