Testing for COVID-19 Infectivity

Testing for COVID-19 Infectivity

how can we determine whether someone who has covet 19 can transmit the virus to other people tests in routine clinical use such as reverse transcription polymerase chain reaction and antigen tests are designed to determine whether sars covu2 is present or not but many people have proposed that these tests be used to determine whether a patient is infectious furthermore tests for sars cov2 that are not routinely used in clinical laboratories such as viral culture and detection of sub-genomic viral rnas have also been discussed as indicators of infectivity but how accurate are any of these tests for determining whether someone is infectious welcome to editors in conversation this episode is brought to you by the journal of clinical microbiology available at jcm.asm.org and on twitter at jay clinton micro i’m your host jcm editor-in-chief alex mcadam this podcast is supported by the american society for microbiology which publishes jcm i’m joined by an expert on laboratory testing for sars kobe 2 dr matthew binnicker dr binnicker is the director of clinical virology and professor of laboratory medicine and pathology at mayo clinic you can find him on twitter at dr matt binnicker b i n n b-i-n-n-i-c-k-e-r he has an article in press at jcm titled can testing predict sars co-v2 infectivity the potential for certain methods to be a surrogate for replication-competent virus the article is open access like all of our articles about covet 19 and we’ll post a link to it with the show notes matt welcome to editors in conversation alex thanks for the invitation let’s start with something basic why is it important to determine whether someone with covid19 is infectious can’t we just assume that everyone with a positive test is infectious and leave it at that i wish we could i wish it was that straightforward unfortunately most tests that are performed for coca-19 use a technology called molecular testing so we’re actually looking for the viruses rna in clinical samples and we’ve learned now over 18 months of experience that those tests especially pcr based test can be positive in copit19 patients for prolonged periods of time even after their symptoms resolve and most likely after they become no longer infectious so we can’t just use a positive test to determine whether someone is infectious or not it’s important though to know because right now we require individuals who test positive and have symptoms to isolate or quarantine so that removes them from their normal social routine going to work going to school and so there’s a lot of interest on determining whether there can be a test that’s used to determine that someone is no longer a risk for transmission of the virus to other people so again a lot of focus on that good so let’s get into the potential for tests to do that i like the way that you arranged your article because you went through four separate tests and then after each discussion of the test you gave your own opinion and we’re going to follow that basic structure as we go through these as well so let’s start with the first test you talked about which was viral culture how accurate is viral culture for indicating whether someone is infectious yeah viral culture unfortunately is not performed in many clinical laboratories anymore it’s really become more of a niche test and is performed in some research laboratories and public health settings that being said when a viral culture is performed on a patient with cobit 19 and we’re actually able to recover the virus that’s a pretty strong indicator that they would be a source of transmission uh to other people so in other words a positive viral culture should be interpreted as that individual is a source of transmission or is infectious uh for starters kobe 2. in contrast though a negative culture really doesn’t rule out the possibility that a person could be infectious to others and that’s because viral culture is insensitive meaning you could get a negative culture but there still could be infectious virions in a patient and that’s why viral culture has largely been replaced by molecular testing over the years is because it’s just not that sensitive and we’ll talk about some of the other limitations of viral culture i’m sure yeah you mentioned you mentioned one which is that many laboratories don’t don’t do it why not and what other disadvantages are there to spiral culture as a test yeah for decades viral culture was considered the gold standard for detection of viral infections it is a labor-intensive method it requires expertise it’s can be subjective meaning it’s part science part art form and really requires laboratory technologist to gain that expertise over time to determine what is cytopathic effect and cells and what is artifact so that labor intensiveness that expertise i think has somewhat been lost as fewer and fewer labs have performed it it requires special infrastructure in laboratories and also cultivation of certain viruses including starscope e2 can represent the biosafety risk to laboratory personnel so if there’s a virus like sars cov2 or mers coronavirus in a sample and we grow that up in eukaryotic cell lines it can actually put our lab staff at risk of acquiring the infection and we tend to talk about viral culture as if it were one monolithic thing and i mean for for cyrus kobe 2 specifically are there differences in the methods that can be used does it vary in sensitivity depending on on how you perform it yeah there are different approaches within viral cultures so we can inoculate samples into eukaryotic cell lines in tubes and then incubate those tubes and look for cytopathic effect that is a relatively sensitive for viruses that will actually grow in those cell lines but it can take quite a while it can take days and some viruses it can take weeks to actually see that cytopathic effect there are other viral culture methods like shell vial where we’re inoculating samples into a tube that has a cover slip we spin those samples down and then we actually look the next day under a fluorescent microscope after we’ve stained those cells with a fluorescent tagged antibody for viral proteins that get expressed on the cells that’s more rapid turnaround time and can enhance the sensitivity compared to cell culture but they all have their disadvantages especially just overall sensitivity got it and you you touched on this but if you want to expand on it that’s fine in those rare instances where we have viral culture available or more likely when we’re trying to interpret a paper that’s been published what should we make of positive and negative viral culture results in terms of indicating whether or not someone is infectious yeah well right now viral culture for sars kobe 2 has remained kind of the gold standard against which other tests that are being considered for infectivity tester compared against so it’s really the only method that we have to show that there’s replication competent virus in a sample so if a viral culture is performed on a cova 19 patient and we’re actually able to recover sars cov2 that should be interpreted as that individual has replication competent virus that could be transmitted to other people in other words they could be they should be considered infectious but a negative viral culture really doesn’t mean much it it doesn’t mean that that person is no longer infectious because there have been studies done where an individual may be negative by a viral culture but has been shown through contact tracing to shed virus and transmitted to other people so we again can’t use a negative culture to rule out infectivity got it thank you man so let’s turn to the next test that you talked about in the paper um and there is rapid antigen tests there has been a tremendous amount of discussion about using rapid antigen tests for detecting people who are potentially infectious with covet 19 what did the data tell us about this yeah first of all the rapid engine tests the benefits of those tests is they’re quick they can provide us with 15 results in 15 to 20 minutes and i think that the experience and the data that have been generated so far show that the specificity of the rapid antigen test is probably better than we anticipated uh when those tests first were being developed the the data that i’ve seen for certain rapid antigen tests the specificity is probably in the 95 to 98 range which is which has improved over prior iterations of rapid antigen tests um the the problem with the rapid antigen test has always been sensitivity and that has played out with stars cov2 as well i think that the rapid engine tests are going to be pretty reliable in a symptomatic individual who is in the first day to three days of their symptoms they’re going to do a pretty good job of being positive and and that population as you get out further from their onset of symptoms the sensitivity of the rapid engine test decreases and then the real concern is in the asymptomatic population so those that don’t have any symptoms there’s been interest in broad screening of the population with rapid antigen tests including asymptomatics the data show though that the sensitivity in that group is pretty poor and the 35 to 50 percent range so it’s worse than to flip up a coin so right now rapid antigen tests utilization usefulness of that method i think is still in question and whether it can be used as a measure for infectivity i think is still upper debate one of the things that seems to happen in every paper describing an antigen test is that they will compare that test to pcr and then say well the test is overall the antigen test is overall this sensitive but if you look only at samples that have a high viral load as determined by pcr well then it’s very sensitive how should we interpret that and and i’ll let you go first and then i’ll i’ll give my thoughts about it yeah i know it’s a really important consideration i think that it’s true that in patients that have really high levels of styroscope b2 uh the rapid antigen test is going to be more reliable so when we compare the pcr and we look at pcr cycle threshold values which i’m sure we’re going to talk about here in a few minutes when we look at those samples that have cycle threshold values let’s say below 25 the rapid antigen tests do a better job because there’s more virus present but as we start to get to higher pcr cycle threshold values meaning lower amounts of viral rna we start to get see less reliable performance of rapid antigen tests now that would be okay if we thought that patients who had more weekly positive pcr tests weren’t infectious the problem is as we we know now through experience and data that’s been published that there can be patients who are positive by pcr at very high cycle thresholds that probably wouldn’t be picked up by rapid antigen who have transmitted the virus to other people really nice paper published in journal molecular diagnostics recently showing that there can be individuals who are infectious even at low amounts of viral rna as determined by pcr to me those papers when they compare the sensitivity of the antigen tests for high and low viral loads my concern is that people will misinterpret that and think oh the test works pretty well for this group of patients but for all patients doesn’t work quite as well and to me really what’s happening is they’re saying that the test is not very sensitive and explaining why it’s not very sensitive it’s not very sensitive because it misses those with lower viral loads but it’s sometimes i think that the data interpretation is not as as clear as it as it could be and i i think basically our perspective on this is quite similar yeah that’s right alex and i think everybody has to remember that the virus and following infection is it’s like a bell-shaped curve right from infection to incubation to reaching the highest point and then it drops off so if you perform a rapid antigen test soon after expo exposure the viral load is not going to be very high and the antigen test is likely going to be negative as might pcr but then that viral load is going to increase so it’s all about when the test is performed as well following exposure and the recommendation has been use the most sensitive tests like pcr because we want to make sure we’re maximizing our chance of detecting patients regardless of where they’re at on that bell shaped curve anything to add to your your thoughts about whether antigen tests should be used to determine whether somebody with covid19 is uh is infectious well i think that a positive rapid antigen test in my opinion probably means that that person has higher amounts of virus and i would uh be reluctant to say they’re anything but infectious if we get a rapid antigen test being positive on the contrast a negative rapid antigen test should not be interpreted as this person is no longer infectious um especially if it’s being used in a diagnostic setting i think where there’s a lot of interest in potential future application is in those who have been diagnosed with covid by a pcr test can we then use rapid antigen following their diagnosis to help get an idea of whether someone is shedding high amounts of the virus or not and so rapid antigen testing may actually play a role there in helping to determine after a diagnosis when an individual stops shedding high amounts of virus the concern there is that again there can be people who have low amounts of the virus in their respiratory tract but can still transmit the virus to other people so i have some concerns about a negative antigen test even in that setting got it um so let’s talk about cycle threshold and this has been uh what cycle threshold of reverse transcription polymerase chain reaction of course which has probably been the most widely discussed and studied test for determining whether someone can transmit sars cov2 so just to begin with can you give us a refresher on what a ct value is and what it means sure so during real-time pcr the fluorescence that is generated each cycle of the pcr is plotted in an amplification curve and the point in the pcr reaction when the fluorescence crosses a threshold or basically pulls away from the baseline or the background noise the instrument says that’s the cycle threshold that’s the point at which the test is becoming positive and so a lower cycle threshold means that there was more viral rna in the sample because it pulled away from the baseline earlier in the pcr reaction in comparison a higher cycle threshold like 35 out of 40 would mean that it took almost the entire pcr reaction before we started to see a fluorescent signal so the cycle threshold is inversely proportional to the starting concentration of the amount of the target that we’re after so a low ct means a higher amount of viral rna in the sample a high ct means a low concentration of viral rna perfect and just to to go a little bit further quantitatively uh what does a change of i don’t know three cts indicate does that mean three times more virus more or less virus or how does the number for the ct relate to the quantity of viral rna yeah most of the real-time pcrs being used for use for cova diagnostics aren’t fully quantitative so we don’t use standard curves and actually get a pure quantitative value but in general when we see changes of let’s say three cycles that typically correlates with about a log difference in the starting concentration of viral rna so if a pcr gives a cycle threshold of 30 and we say that’s like a thousand copies of a virus per milliliter of sample then if you got a cycle threshold of 27 that might be like 10 000 copies of virus per milliliter of sample got it thank you um i had a very intelligent highly educated physician asked me not long ago this question uh we have a patient the sarsko v2 pcr was negative can you provide us with the ct value and again this was somebody who you know was an outstanding doctor who’s well informed but it indicated to me that there is not as good an understanding of ct values as there might be because if the answer is of course there was it did not cross the threshold otherwise we would have ordered it as positive have you had similar types of encounters where you’ve had to educate people about what ct values mean for sure yeah i think there’s a whole range of knowledge and understanding of the nuances of real-time pcr some physicians especially those that infectious disease providers who routinely discuss these types of cases with the laboratory have a higher level of understanding whereas others may not and so it just kind of emphasizes the importance of discussing these cases with the laboratory director ensuring that there’s the right information right understanding of which test was used how the result should be interpreted all right so let’s get to it what do we know about whether ct values indicate whether someone who has covered 19 is likely infectious or not yeah so there have been a few papers showing that there’s a correlation between patients who test positive by pcr with a low cycle threshold so let’s say a cycle threshold of 20 or below 25 there’s a correlation there between low cycle thresholds and recovering the virus and viral culture more often okay that makes sense there’s more viral rna probably means that there’s more virus in the sample that means the culture has a better chance of being positive so not too surprising but there’s some good papers showing that that correlation is there where we get concerned i think it gets a lot muddier is on the opposite end of the spectrum where the pcr of this test the pcr test is positive but the cycle threshold value is high so once you start to get above 30 then we start to see a less strong correlation between the pcr result and the viral culture so we see fewer viral cultures being positive in that setting right now we don’t feel comfortable i don’t feel comfortable saying that individuals who have a high pcr cycle threshold value so again let’s say 35 that they are not infectious because we have cases where we know individuals are very ill and there’s contact tracing studies that have been done showing that some individuals with high pcr cycle threshold values can transmit the virus to other people so although there is a correlation at the low cycle threshold value that they’re probably more infectious we can’t use a high cycle threshold value as an indicated indicator that a person can’t transmit the virus to others one of the concerns i always have and this i think relates to your comment about samples with high ct values is the variability of ct values and so how variable are they and what are the sources of variation in ct values yeah uh it’s a great question and i think there have been a lot of studies published showing that ct values are highly variable across laboratories across tests there was a nice cap review showing that if we take the same samples and we have them sent to a number of laboratories who are using different molecular testing methods that those cycle threshold values can range by 7-10 cycles again when testing the exact same samples now what’s the source of the variability it can range from the pre-analytic all the way to the analytic stages of the testing so we know that there probably is more virus and the nasopharynx as compared to the anterior nerves so if we’re comparing ct values between those samples we might get lower ct values and the nasopharynx compared to the anterior nerves but also just the way the sample is collected so if you have a provider who does a really good nasopharyngeal collection really makes that patient uncomfortable you’re going to get more samples more sample you’re going to get more virus and you’re probably going to get a lower cycle threshold value in comparison i’ve heard of cases where patients said they got a nasopharyngeal sample but the swab didn’t make it past their mid-turbinate in other words the nurse or the physician didn’t get back to the nasopharynx and that setting you’re not going to get as much virus and you’re probably going to have a higher pcr cycle threshold value extraction can play into how much nucleic acid we recover so not all extraction protocols or equipment generates the same quality or amount of nucleic acid and then the actual test uh some assays break apart the genes and provide cycle threshold values for each gene some assays combine the targets into one cycle threshold and that may actually reduce the cycle threshold value uh because we’re combining the fluorescence signal so a lot of variables uh and i think the take-home message there is um it’s gonna be not consistent across laboratories and there’s a high level level of variability in cycle threshold values how do you think that laboratories and laboratory directors should handle requests to provide ct values for purposes of patient care yeah there’s been a lot of discussion on whether ct value should be included on patient reports i’m not a proponent of that because again the lack of universal understanding of what a ct value means it’s not there and there’s a lot of nuances that need to be discussed and understood when when using ct values i think in certain situations the ct value can be helpful and a phone call from the physician to the lab director to discuss the case and have a verbal of what that cycle threshold value is is appropriate and one example would be a patient was diagnosed with covit 19 in january of 2021 and six weeks later is coming back in for a procedure they do a pcr test for part of a pre-procedural screening and they’re positive they don’t have any symptoms at that time point they completely recovered call the laboratory discuss the case and if the cycle threshold value there is 38 or 40 we can probably put the pieces together and say this is most likely viral rna that’s just sticking around from the infection that that person had in january but if that cycle threshold value is 20 i’d be a lot more concerned about proceeding because that value might tell me there’s a lot more viral rna there and we might need to pursue that with some other studies so it sounds like you you would consider reporting on a case-by-case basis where you have the opportunity to communicate with the provider so you can provide a complete picture and help them understand what that ct value means for that patient yeah on a case-by-case basis i like to have a conversation with the physician and provide them with a verbal we have not taken the step of including cycle threshold values actually in the report itself sometimes the physician will make a note of it in the chart and say after conversation with a lab director the cycle threshold value for this patient is and i think that’s appropriate just for documentation purposes but i’m not a proponent of it actually including that value in the report itself and another reason for that is that these tests are emergency use authorized as qualitative assays and when we go as so far as to include a value on that report it may change the regulatory viewpoint of that assay into more of a semi-quantitative which can put some risk on the reporting laboratory got it um and finally the last category of tests you talked about uh include a couple of tests for viral rna intermediates and these have been studied as possible indicators for the presence of replication competent sars cov2 so they include sub-genomic viral rnas which we heard about very briefly when president trump had covet 19 his personal physician referred to them can you tell us what uh subgenomic viral rnas are and how well they might indicate whether there’s replication competent virus in a sample sure so after cyrus kobe 2 infects a human cell the virus unpackages and the genome is replicated and when that genome is replicated a full length genomic replication event occurs but also there is a discontinuous transcription process where subject genomic or kind of pieces of the genome are replicated now those subgenomic transcripts are only produced following infection so if we’re targeting those and those are detected that means that the virus has infected a cell and produced those so they’ve been considered as a marker for infection but there are some nuances to interpreting those results that we can discuss how good is the evidence that those sub-genomic rnas can be used to indicate whether someone is infectious or not yeah i think it’s it’s mixed data there have been some publications showing that detection of subgenomic transcripts correlates strongly with lower cycle threshold values by pcr and culture positivity but there have been some other studies showing that detection of subgenomic transcripts can be persistent and prolonged and so what’s uncertain is how long those sub genomic transcripts are stable and how long they hang out in our cells there’s concern that they may persist for quite a while because they can be protected inside vesicles from breakdown and if that’s the case if they can be protected from breaking down and might actually hang out in a cell after infection for weeks or months then they might not be a real reliable marker for infectivity so let me make sure i’ve got it right so the sub genomic rnas are only produced during viral replication and so people thought well maybe that’s a good way to measure whether the virus is replicating or not but the concern is that they may hang on a long time and so they may persist even when the the virus is no longer replicating is that is that correct you’ve got it yeah we make it much cleaner to interpret if those subgenomic transcripts were produced following infection but then were rapidly degraded and over the course of days what we’re learning though is that in some cases those sub genomic transcripts might actually be kind of packaged inside these lipid vesicles that protect those transcripts from degradation by nucleases and they might actually hang out for weeks or maybe even months after an infection so may not be a real reliable marker for a period of infectivity got it the other kind of viral rna intermediates that you talked about in your article are the strand specific rna transcripts so what is a strand specific rna transcript and what does its presence indicate in a clinical sample and i think you’re gonna have to take some people you have to take people back to a little bit of basic virology on positive and negative strands so don’t be shy again let us have it yeah i think in essence the principle is very similar to what we just discussed with the sub genomic transcripts there’s the positive and negative strands of the rna and so these strand specific assays have multiple steps where they are targeting a positive strand of the rna and a negative strand of the rna and again looking at both genomic level and subgenomic levels of those strand specific transcripts as a way of determining uh whether a virus is more likely to be infectious or not so again complicated but we’ll leave it there at the high level all right staying at a high level what do you think are the most important questions that we need to answer about laboratory testing to determine whether or not someone can transmit sars kov2 what what should people be doing next yeah i mean there’s still a lot of interest in whether any of the existing tests can be used to categorize people as infectious or not again it’s because of that disruption to their social lives work school right now we’re relying on symptom and time based strategies to determine whether someone should be in isolation or not as more people are infected the longer we go into this pandemic we can’t just have everyone staying at home for 10 days if they test positive by a pcr test so we need to continue to focus on ways to develop tests or use existing tests to begin to say this person is less likely to be a source of infection to others versus this person is still clearly infectious and needs to remain in isolation i think that there’s work that can be done and studies that can be done to continue to look at our existing tests and how they might be used but likely we’re going to need to develop new tests that are different than the ones that we’ve viewed to really give us an accurate and confident answer of this person is a source of infection versus this person is no longer infectious do you think that the subgenomic rnas or the strain specific rna transcripts are likely to be the ones that that we need to work on for those those new tests or quantity better quantitative assays using conventional uh real-time reverse transcription pcr yeah i think that the subgenomic rna tests represent a lot of promise uh the strand specific do and i think that what we’ll need to do is get more quantitation of the transcripts that are being amplified by those tests also looking at maybe a number of different genes some multiplexing by for example droplet digital pcr i think holds a lot of promise when looking at those sub genomic transcripts so i i think again there’s promise there i hope that we’ll we’ll get to the point where we can get those answers um i do think that there’s potential for antigen-based testing to play some role again in people who have tested positive um but we’re going to need to make some progress especially with increasing sensitivity of the antigen test before we’re really able to be comfortable in saying this no this person’s no longer a source of infection thank you we’ll put the link to your article in the show notes matt this has been a really interesting conversation thank you so much for joining me on editors in conversation oh this has been fun thanks again for inviting me and thank you for listening
rn

Testing for COVID-19 Infectivity

rn

Share this post

Leave a Reply

Your email address will not be published. Required fields are marked *