Episode 57- PCR Testing Remains the Gold Standard

If you are able and willing to help me with these translations, please PM me on Reddit ASAP. My schedule is getting busier, so any help is appreciated.

DISCLAIMER: I am not a doctor or a healthcare practitioner, nor do I have a master's or a PhD. I am not a native German speaker either- I only know German as a foreign language, one I learned using whatever resources I could get a hand on. If my translation conflicts with that of anyone belonging to any of the groups mentioned above, please refer to theirs instead of mine. Needless to say, if an official translation, done by a scientific translator comes out, then refer to that instead of this one. I am simply a medical student who admires the work of Dr. Drosten and colleagues, and I would like to provide unofficial translations of the podcasts so that they may be helpful to those who do not speak German.For corrections, please reach out to me on Reddit and send advice in. I apologize in advance for any errors I have made. There are some footnotes at the bottom of this translation.

Original release date: September 22, 2020
Original audio: https://www.ndr.de/nachrichten/info/57-Goldstandard-bleibt-der-PCR-Test,audio750776.html
Original transcript: https://www.ndr.de/nachrichten/info/coronaskript228.pdf

--------

HENNIG: Patients shouldn't infect each other with the Coronavirus in the waiting rooms, says the Minister of Health. That's why he's relying on fever ambulatory clinics during the cold part of the year—a place to go for anyone with SARS-CoV-2 or flu symptoms. Jens Spahn as therefore announced a new testing strategy and given another figure. One third of all tests of the virus in Germany, supposedly, have been done in the past four weeks. Thus, we would like to talk primarily about diagnostics in the podcast today. What tests are there for the coronavirus? What can they do? What can't they do? Over the last week, the incidence rate has become more dynamic. Germany is still far from being in the same situation as countries such as Great Britain and Spain—this must be said. However, the Robert Koch Institute reported the highest number of new infections since April last weekend. Did this surprise you at this point? Is this worse than expected?

CIESEK: I think that it's always easy to say, "I expected that" in retrospect. I think that what we're seeing now isn't completely unexpected. The autumn is coming, it's going to be colder. This means that other viruses are going to become more common, for instance the Rhinovirus. Nevertheless, nobody knows 100% how things are going to go from ehre. All that is pure speculation.

HENNIG: Today, we want to take a look at the possibilities available to the laboratories and to medicine to detect infections. We want to talk about tests today. Of course, test centers also play a role. When the Minister of Health talks about fever ambulatory clinics, it sounds good at first. We've seen similar facilities as early as in the spring. But there is also the issue of capacity. You work quite a lot with the health authorities in Frankfurt am Main and can look closely at the structural conditions there. Is this,a t first glance, a good plan? Or would you say, "Well, this isn't as simple as it sounds on paper at all"?

CIESEK: In Hesse, we have these centers for smear test. In general,. I find them a good idea for the autumn and winter. I don't find the name [fever ambulatory clinic] very appropriate or suitable, because not everyone has to have a fever [to go there]. That suggests that only people with fever can go to these ambulatory clinics. That's not supposed to be the case, of course. I don't think that works very well. But I think that it makes sense for there to be special facilities where patients with respiratory infections can go in order to take the burden off general practitioners and, on the other hand, patients who do need to visit the GP for other reasons don't need to worry about getting infected in the waiting room. Thus I find the idea itself very good. You have to see how to regulate this in every federal state. Yesterday, I read that they were thinking about whether an antigen test can be done, and then for instance after 15 minutes, patients can be told if they are SARS-CoV-2 positive and then have a PCR test done. I'm not in favor of this, I must say, because with symptomatic patients, i would like to know the exact diagnosis at once. We'll discuss later what weaknesses antigen tests have. I imagine that it's difficult—imagine someone with symptoms. They're shown to be negative by the antigen tests, and then they're thought to have rhinoviruses or whatever other viruses, and then they go back home and then after a few days their symptoms get worse and they go to the hospital. And then there's this confusion, they were already tested negative, should they be tested again? Often, during the more advanced stages of the infection, it's not easier to diagnose because the viral load would then not be so high or does not have to be so high. Thus, the immune system as well as its reaction play a role.

Thus, I think that in symptomatic patients, the correct diagnosis must be made with a sensitive test, i.e. PCR, in order to avoid making the whole thing more difficult. Here in the University Clinic, we have also had patients who had been sick for two or three weeks and then came here, were then consistently tested negative by PCR and then it was only determined afterwards, for instance through radiological examinations or, if they had died, through autopsies, that they really were infected.


-----PCR TESTS, ANTIBODY TESTS, ANTIGEN TESTS-----


HENNIG: How this can be, that a test comes up negative even though a patient is infected, is something we will come back to in a bit. We've now started with questions of terminology. Knowledge about the virus and virology grows with the pandemic. Some of the terms we already used are self-explanatory. PCR tests, antibody tests, antigen tests, the last two sound similar, but mean different things. Let us sort them out a little. There are direct methods of detecting the virus and indirect ones. You mentioned that with direct tests, PCR is the gold standard, as the virologists say. What other possibilities are there?

CIESEK: "Direct" means that you want to detect the virus directly. "Indirect" means that you want to show that the person had contact with the virus. An example of a direct procedure is the famous PCR test, in other words an antigen test. If you detect proteins from the virus, that would be direct viral detection. There are still other procedures such as electron microscopy. Viruses are very, very small and you can't see them through a light microscope. Therefore, you'd need an electron microscope to really see them. Of course, not many laboratories do have them. University clinics have them, but access to them is very limited, you must have a lot of experience with electron microscopy. And there is the issue of cultivation. In other words, isolating the virus. With viruses, you always need a cell. Viruses don't have their own metabolism and can't propagate themselves alone. That makes them so-called parasites. In other words, they need us and use our cells and parts of our cells to multiply.

HENNIG: This is different from bacteria, for instance.

CIESEK: Yes, that's the big difference. Viruses always exist in cells and if you want to breed them—in Germany, only large virological projects have them, you'd need a so-called S3 laboratory. That is, a special laboratory with a certain standard of safety. There are S1, S2, S3, S4 laboratories. S4 is the highest level. They exist, for instance, in Marburg or in Hamburg, if you want to work with Ebola. S3 is directly underneath. If you want to work in one, you must change your clothes and wear special protective clothing. And the pressure there is very low, you wouldn't want to work there the whole day, it isn't comfortable to do so. And if, for example, you want to take a probe from a patient and cultivate it, you'd need cells first. For this, most of the time, you take tumor cell lines that constantly divide, which make them good to work on in the laboratory. Then you have to plate them. Then you can take material from the patient and then place it on top of this layer of cells. You need to wait a long time for this. And there you see the problem of cultivation. That, unfortunately, lasts days or weeks. It depends very much on the pathogen. You must wait up to a week to really say that a sample is negative or cannot be cultured. And there are different ways to show that the cells are infected. The simplest is if the virus simply kills the cells, in other words, if the cells are destroyed, allowing the virus then to be detected via PCR or other methods.

And then there is sequencing. We sequence many of the viruses which we have isolated. This is what this is called, seeing if the genetic information of the virus has changed in some way. This is done in virological labs as well, mostly in Berlin of course, in Düsseldorf, in Frankfurt. You can see, however, how much effort is needed. You can do this with individual tests. But you can't test a million people per week like this. First of all, it'd simply take a very, very long time. Secondly, you need specially trained staff. Not everybody can go inside an S3 laboratory. And if you need to wait a week or so for a definite result, this would be far too late for diagnosing SARS-CoV-2.


-----DIRECT AND INDIRECT VIRAL DETECTION-----


HENNIG: As you have already said, direct virus detection is immediate, when infection occurs. And if someone has contact with the virus, it can be detected later by antibody tests. The difference is that with antibody tests, you don't need a throat swab or saliva, but rather, a drop of blood or blood serum.

CIESEK: Exactly, that's the big difference. If you're not sure what the doctor is doing, if the virus or perhaps antibodies were detected, you could tell by looking at which sample the doctor took. With indirect detection of antibodies, we need blood or serum, in other words, the liquid component. And then we could do different tests, for instance, ELISA tests. They are now automated and can detect antibodies. And the gold standard here is the so-called neutralization test. But this is another test which you need an S4 laboratory, because you perform it in a similar way to the way I just described, cultivation or isolation of viruses, for which you need cell culture again, which you need to infect with a virus you have isolated before. You then add the serum from the patients. Then you can see whether the cells have been infected. If the cells were infected even in the presence of serum from the patient, this would mean that no neutralizing antibodies were present. However, if the serum causes the cells to no longer be infected, that would mean that the patient has neutralizing antibodies.

HENNIG: Which prevent the cells from taking up the virus.

CIESEK: Exactly.

HENNIG: If we speak about rapid antibody testing, which is always a topic of conversations, where would neutralization tests come in? They are an extremely important factor in detecting a sort of immunity.

CIESEK: Exactly, these rapid antibody tests are unfortunately not accurate enough to be suitable, for instance, for proof of immunity. I believe that this was asked about in April, if an "immunity passport" could be implemented. Anyone with antibodies would get such a pass. We've just now learned that this isn't sensible. I have also just seen that the Ethics Council has just given a statement about this.

HENNIG: Today, in fact.

CIESEK: It rejects this, which is a sensible thing to do, because we can see that this isn't well correlated. And these tests, these automatized test aren't accurate enough to be able to be used to make statements about individuals. This would be more so if you do big studies, for instance testing 10,000 or 100,000 people for antibodies, in order to say how far the virus has permeated into the population. There, it is not so relevant if the test is not 100% accurate. But for individuals, it is important because they want to know if they are protected or not. We can't say that with a commercial antibody tests. In Frankfurt, for instance, we always carry out two tests. We do a combination of two antibody tests in order to increase the specificity.

HENNIG: Specificity being the question of whether a test really responds to bodies that have certainly originated from the Coronavirus and not from elsewhere.

CIESEK: Yes, we can briefly explain the terms again. They are very often confused. Sensitivity means that the test correctly detects positives. In other words, that it can detect if someone really is infected.

HENNIG: Or was, in the case of antibody tests.

CIESEK: Correct. Specificity means that the test correctly marks the uninfected as negative.


-----IMPORTANT: THOROUGH NASOPHARYNGEAL SWABS-----


HENNIG: This leads us to the question of the role swabs play in direct detection, e.g. via a PCR test. Many people who have already had such a test know that to do such a smear test in the nasopharynx or in the throat is not all that simple. You must have enough material on the swab. About sensitivity of PCR tests, are possible errors in taking swabs factored into calculations?

CIESEK: No, as a rule, no. First of all, it must be said that there is a lot of difficulty with this disease. Nasopharyngeal smears depend on many, many factors. For instance, which swab brush you use—there are different levels qualities. I remember when in April, when all the materials were in short supply, we took every swab and swab tube we could get our hands on. They were however of different qualities. There were simple cotton swaps, then there were the so-called PhloxSwaps. They can extract a noticeably larger amount of material are are better suited for this. There are different transport media in which the swabs are transported. This influences the results. The technique of the swabber or researcher, how good they do it, also plays a factor. How many cells they took. How the sample was then stored, for example. Maybe the doctor left it in the sun for a few hours or cooled it. And then it is certainly possible for the pathogen to be unevenly distributed within the sample collector. That is somewhat different to when you look at blood or serum. If a doctor draws blood and then serum, it is relatively defined. You can say exactly, "I am drawing 10ml of blood and I need 2ml of serum". With swabs, it is not so homogenous. And this also plays a role in terms of testing.

HENNIG: But in the blood, a direct test cannot be carried out with only antibodies?

CIESEK: It's impossible for SARS-CoV-2. With the extremely ill, those in ICUs, there are individual reports that the virus was detected. But as a rule, it is not the case in the mildly symptomatic, the outpatients.

HOME SALIVA TESTS
HENNIG: However, there are probably alternatives to these throat and nasopharyngeal swabs, namely saliva tests. We have already mentioned them in our podcast too. There have been encouraging results concerning the significance of saliva tests. And there's also the fact that people can likely take a sample themselves and thus unburden the health service a little. According to research, what other serious alternatives are there to swabs?

CIESEK: I think there are a lot of people, ourselves included, doing research on this. I think it's very important to evaluate how swabbing oneself will be. For now, this has enormous advantages. You need far fewer specialists and less protective clothing. Theoretically, you can do more tests because a major roadblock would have been eliminated. And, this is not to be underestimated, the risk of infection is much smaller if the patients perform the swabs on themselves and can stay at home and then leave their samples at the door and have a transport service pick them up. There are enormous advantages, it also helps the public health service. Personnel are also kept separate [from the patients]. Thus, there are many efforts to look for other biological matter that can be used for testing. For example, in JAMA Open, a study appeared in July, from the USA. They researched nasal swabs. With these, you don't go through the to the throat, but rather, you swab the middle of the nasal area. They had a sample size of 185 people, albeit symptomatic people, there were no asymptomatics there. Thus, the results may not be applicable to everyone, but they are for the symptomatic. They did a nasopharyngeal swab and compared it with a self swab from the middle nasal cavity and said that if the viral load his high, the results obtained by PCR were comparable. However, if the viral load was not high, if the Ct value was over 33 in this case, false negatives were often to be seen. You can thus see the different again, this depends very much on how high the viral load is. And what should really be said about the study is that a lot of the participants were from the health service. From the 185 participants, 158 were in hospitals or in the health service—there were therefore no laypeople there. And this must really be considered. Another difference was that the samples were then collected and then tested the next day. So not on the same day as the samples taken by the doctor. Thus, with the study, you must say that this is an interesting approach, this can work. The only question is, how will this work by laypeople? And what is it like if someone is really asymptomatic or presymptomatic, as opposed to those researched here, who were symptomatic—does it work with them as well?

HENNIG: What is not yet clear, as far as I know, is whether or not asymptomatic or presymptomatic cases have a comparable viral load. The question is still up for debate, isn't it?

CIESEK: Exactly, in any case. But they also want to do self-tests on people without symptoms. That's why it has to be shown that you have no symptoms. There's a paper in the New England Journal from August, from the USA as well. Salivary tests were compared with nasopharyngeal or oropharyngeal swabs, i.e. swabbing the throat through the nose or the mouth. Interestingly, among them, there were people without symptoms among them. However, most of them were also hospital staff, so the same limitation was present here, that there weren't any real laypeople. They compared patients who were in the hospital, in other words, those with symptoms. They also included people with no symptoms, but almost 500 hospital workers had themselves tested every three days. And in the saliva of 13 people without symptoms at the time the swabs were taken, they found SARS-CoV-2.

HENNIG: So a relatively good result for saliva tests.

CIESEK: Exactly. They also wrote about how the tests were done. The subjects didn't eat or drink anything beforehand, nor did they brush their teeth. And that's my problem with this. You know a little bit about it, when you go to give a urine sample to the doctors—the containers aren't so small at all, around 125 to 150ml. They had to fill the beakers to about a third, around 40-50ml. That's no small amount at all, producing 40-50ml of saliva you wake up and haven't eaten or drank anything,


-----A URINE CONTAINER FULL OF SALIVA-----


HENNIG: You need a lot of saliva.

CIESEK: Exactly, I imagine that this is pretty time-consuming and not at all so trivial. That's one problem. The other problem is in the laboratory. If I think of my laboratory, my assistants would kill me if we only used urine beakers, because they take up a lot of space. In the laboratory, we have stands in which swab tubes fit, but urine beakers are far bigger. And if you think about having a thousand of these beakers per day, you would have to fit them all into a safety cabinet—in the laboratory, we work with safety cabinets in these laboratories, you wouldn't be able to get all of them down. And this is logistically more difficult, and would require more effort, than these swab tubes. Thus you often have this problem, especially when you haven't eaten or drunken anything in the morning, where the saliva is often viscous and isn't very easy to pipette. And sometimes it even causes foaming. And this can also cause problems with the tests. And another limiting factor is that in the New England paper, they required a third of the beaker [to be filled], which is something I can't at all imagine with children, being able to produce such a large amount of saliva without drinking.

HENNIG: Is the exact time, the morning, relevant? Because you said no food, no drink, no brushing your teeth.. You would also have to wait even if I were to take a throat swab sometime else during the day.

CIESEK: That's correct. First of all, the PCR results can be altered by brushing your teeth or by food. That can lead to inhibition, to the PCR not functioning.

HENNIG: So the PCR doesn't work.

CIESEK: Exactly, the enzymes inside don't work properly anymore and then you wouldn't be able to get any results at all. And if you haven't eaten or drunken anything in the morning, it'd be a bit like giving a urine sample to the doctor. They also tell you to hand in the urine you produce in the morning. In other words, that you have material that is as representative as possible and hasn't been diluted and hope that this ends up giving better results. And the good thing about this study is the sensitivity of saliva, as was done in the study, was as good as the probes. So it is working, but it wouldn't be easy to do everything with saliva at the moment. You'd have to discuss this with the lab too. An advantage of the study is that it was shown that saliva shows less variation in PCR than nasopharyngeal swabs. And that is a problem that we see if we, for instance, take many nasopharyngeal swabs from patients at a time, in other words two or three a day for several days—you can see that the  C­t value (1) can vary a lot.
This again depends on things such as which tube, which swab was used and who did the swab. With saliva, it's a bit more standardized, if you think about how it's done.

HENNIG: We'll come back later to what the Ct value actually is. But first, let's try to understand something people see in TV crime thrillers, namely taking a cotton swab in the cheek as well as some saliva—this would definitely not suffice for a saliva test.

CIESEK: No, we tried that, cheek swabs, and they weren't sensitive enough. So what else is there? I have found an old study not at all related to SARS-CoV-2, from 2017, so not that old, but pre-SARS-CoV-2. And they asked: what about gargling? Is gargling possible, would it make sense as an alternative to a throat swab? And they investigated over 16,000 swabs and from these 16,000 had 79 patients who gave gargle samples at the same time, and some more over the next three days. These79 patients were then investigated and then compared. And there were pathogens, most commonly influenza A and B and RSV, as well as other coronaviruses. And they noticed that of these 79, eight of them were only positive in the throat and negative in the gargle samples, and of them half had a C­t value of over 35, meaning that the viral load was very small. And 18 of these 79 were only positive in gargle samples and negative in swabs. The C­t value was over 29, in other words relatively high too. And this shows that this is an alternative in any case. Both methods have advantages and disadvantages, one must say.

HENNIG: The higher the C­t value, the lower the viral load—people should remember this exactly.

CIESEK: I think that in patients for whom a swab test isn't so easy for anatomical reasons, gargling or spit samples are good ideas or good alternatives. There are simply people that can't swab themselves well, you can't see their soft palate when they open their mouths. Anesthesiologists know this well. There's the so-called Mallampati classification. Anesthesiologists always look inside someone's throat and neck and see how far they can see when they are anesthetized and need to be intubated. They therefore know this well. Of course, this is not the same now. But if you can see the throat when taking swabs and not the soft palate, but rather the hard palate, so you can't look deeply inside—these are patients who cannot be swabbed very well. Or if the nose is very narrow, so you can't get to the back of the throat with the larger swab. For those people, it certainly makes sense to have saliva or gargling samples in order to make a certain diagnosis.


-----STOOL TESTS TO DETECT THE VIRUS-----


HENNIG: In the podcast, we have already spoken about stool tests. This would be difficult, especially with small children. Can they also be significant or do stool tests only contain inactive, dead viral material?

CIESEK: That's a good question, that depends on when you take the sample. The virus is detectable in the stool for a long time, longer than the upper limit for the throat, for many weeks. We have seen this with the first patients of Dr. Drosten from Bavaria, in that study. There are individual cases in which this can be infectious. But as a general screening, this wouldn't be useful. So it always depends somewhat in the goal you have in mind. If you want to confirm a diagnosis for example, if you have a patient in the second or third week of their illness, and they're doing worse, they may come to the clinic but then you wouldn't find the virus in the nasopharyngeal area, in the throat. As we have done before, they would give us a stool sample or a swab, and then we would confirm a diagnosis, because it is important to know, for the patients and for their care, which virus they have or even if they have a virus. In general, stool samples aren't very easy to work with in the labs. You must think a little about the laboratory staff, this is another limiting factor which hinders PCR testing and is not so easy to work with—these need far more manual labor than, for instance, a swab.

HENNIG: When we talk about stool samples, we don't just look at individual diagnostics, but rather, at what people want to know about the infection, in other words, how widespread the virus really is within the population, as well as looking forward to the autumn and winter. You and your team have also researched the question of whether it is possible to draw conclusions from wastewater. In other words, look for places with a lot of infected people, the virus can also be found in the sewage system and we can then see, what exactly is the situation in this county? How far has research come there? Is this really helpful as an indicator or not?


-----DETECTION VIA WASTEWATER-----


CIESEK: Exactly, we have done this along with colleagues from Aachen. In April, when North-Rhine-Westphalia was very affected and there were many infections, they examined nine sewage stations throughout North-Rhine-Westphalia and brought sewage samples to us in Frankfurt. And then we saw correlations. The amount of viruses that we find in the wastewater, detected by the amount of PCR or genetic material, was found to correlate with the number of infections in each district. But this should be made clear—we did not find out whether this was infectious or not. That is very important. I believe that it is very unlikely for one to get infected in this way. We have tried to cultivate the virus but did not succeed. There are probably some small genetic segments present in the wastewater, but fortunately they were not infectious.

HENNIG: So to summarize: enough for epidemiological surveillance, but beyond that, no reason to worry.

CIESEK: Exactly.

HENNIG: We have just now looked at different possibilities for swabs. The bottom line for me is that PCR and nasal and nasopharyngeal swabs remain the gold standard. That's a great word. Let's take a closer look at PCR testing, about which a lot is being discussed, in particular sensitivity, specificity. What providers do we currently have in the moment in the market? What are diagnosticians working with? Commercial, non-commercial tests? And what is the difference between them?

CIESEK: Almost every diagnostic supplier now has their own PCR, their own test kit, and they are very different. Every firm designs their own test. And as a rule, two genes are identified, i.e. two different sections of the genome of SARS-CoV-2, sometimes even three. They can of course differ. Then there's a package insert for each test. They have of course been tested and often have a CE label. And if you introduce a new PCR in your laboratory, i.e. if I say, "I want to try this test from Company X, it is my obligation to validate it in the laboratory before I use it as a diagnostic." That means that I must test them with samples which I already know the results of, positive and negative tests, and then show that I would get the correct results. I must document and present my findings accurately. And then I would be allowed to use this PCR as a diagnostic tool. And what I generally want to say about PCR tests is that this is nothing new for us, but that it is rather a diagnostic test that we have been using in virology and microbiology for many, many years, which now serves as a gold standard. They simply have the advantage of being very quick, very, very sensitive and makes culturing the virus unnecessary, which is simply very complicated, sometimes impossible and would take far too long. And there often comes the argument of a detected case being colonization and not infection . I believe that we must first explain what the microbiome really is.


-----THE BODY'S MICROBIOME-----


CIESEK: (cont.) We know this term primarily from microbiology. It means that we are not alone—the entire bodily surface that makes contact with the outside world primarily contains bacteria. But there are also fungi, or nonpathogenic parasites, i.e. parasites that don't make you sick. It is assumed that 1014 to 1015 bacteria are present over the entire outer surface of the body without causing sickness, but rather, the opposite. These bacteria, i.e. the microbiome, depends on genetics, age, gender and where we live, as well as our nutrition. And that starts right after birth, in the birth canal of the mother. When the baby is born, it receives its first microbiome. And in the throat, for instance, we see very different bacteria, the streptococci or staphylococci, though fungi can also appear without causing illness. All of this is relevant because these microorganisms protect us from sickness-causing bacteria by serving as placeholders. They make it harder for other sickness-causing pathogens to settle. This is why our microbiome is usually nothing to worry about. However—this should be made clear—coronaviruses are not a part of this. They are not part of our microbiome and they do not belong there. It is the same for influenza viruses. And this is, I believe, something that gets confused a little. Coronaviruses and other viruses must proliferate within cells, as we have said at the start, and are not part of the microbiome along with bacteria or fungi.

HENNIG: So if you detect them, that means that a person is more or less infected.

CIESEK: Correct, it's an indication that someone is infected. It is not an indicator as to whether someone is infectious or will become sick because of it. This must be emphasized. It is wrong to ask whether a virus is dead or alive because a virus isn't alive. Rather, what we should ask is if it is only fragments or a complete virus that is infectious—we don't know the answer at the moment, but this is completely different from asking if it is dead or alive.

HENNIG: You mentioned earlier that as a rule, different genes, usually two, are tested in PCR. This is done in case one sequence is broken, in order to yield reliable proof, to put it casually.

CIESEK: Yes, for instance, to increase certainty. If you test for two genes, which is what is done for different sicknesses, for instance with HIV, then you increase the certainty that the test really isn't a false positive, because you have two independent tests or two independent genes. This increases the accuracy of such tests.


-----WHAT DOES "Ct VALUE" MEAN?-----


HENNIG: With PCR, it's about how many consecutive cycles of viral replication are needed for it to be detectable, right?

CIESEK: Yes, this was developed in the 1980s. By a colleague, Kary Mullis, from the USA, which he won the Nobel Prize in Chemistry for in 1993. You need the sample for this, as well as an enzyme, the so-called polymerase. Nucleotides are needed, which are small cutouts or individual fragments of a gene, how should I say it...

HENNIG: Nucleic acid building blocks? Can you say that?

CIESEK: Yes, for instance. And you need so-called primers, small pieces of synthetic DNA which then define the section of the virus that you want  to replicate, that's how you define it. Then you mix it all together. Then you have alternate heating and cooling of these samples, these are the so-called cycles. The Ct value, meaning "cycle threshold", is essentially the number of cycles you need to have a positive result and to detect the virus. Thus, if the Ct value is low, it would mean that we need very few of these cycles to make the virus detectable. That means, there is a lot [in the original sample]. And if the Ct value is high, we would need many PCR cycles in order to find viruses. Therefore, the viral load [in the original sample] is smaller. That's a good explanation. Ct values are however not always comparable, they are only comparable within the same system. I cannot compare my Ct value with another test, with another gene that I am looking for, and certainly not directly with another laboratory using a different test, for example.

HENNIG: So you can't say, for example, that with a Ct value of 30, you've certainly reached the threshold at which you can declare a test positive?

CIESEK: No, unfortunately nobody can say that. You'd need a standard. You would need a defined standard with which you can say how many viruses there are. And then you could do dilutions. Then these laboratories can take these standards and check them against the time value in their test, which isn't relevant. So you must think of a threshold and say, 10,000 viruses or 100,000 viruses—below this value, the virus doesn't play a role. And then the laboratory would then, "calibrate it out", so to say, and see what Ct value this corresponds to.

HENNIG: However, this Ct value plays a big role, including in the question of how high the viral load detected is, in conjunction with the discussion of lowering the isolation time and cooldown times. This was the topic [of the podcast] three weeks ago. Does this also play a role? Can people be tested freely, can you say that here, we have detected a small amount of viral material, but the threshold at which we say that this is infectious has not been reached. Where are we with regards to this? Is there any evidence laboratory doctors can use to say, "This is no false positive, this is someone who can infect many other people if they continue to go about?"

CIESEK: I believe this is a very important question, which I have just discussed this morning with a colleague from the clinic. That depends a lot on how much we know about the patients. My colleague had a patient who was tested positive at the start of August and had an infection and had to go to the hospital and then had a Ct value of 35 after six weeks or so. That is a completely different situation to a patient who had symptoms this morning coming in and then having a Ct value of 35. You can't just go about them the same way. In the first scenario with my colleague I would have no problem at all because we really can't cultivate the virus after seven to ten days.So what we said at the beginning, that there really is a complete virus present that is infective in cell culture, we can culture it after seven to ten days, but afterwards no more, if the patient isn't symptomatic. For this reason, I would have no problem with this asymptomatic patient with no symptoms but had an infection six weeks ago and so has some gene fragments within them left over—I would not consider them infectious.
 If they were newly infected, however, then all these other factors play a role. How good was the swab? And for instance, as was mentioned, the Ct value can vary a lot. When it is 35 and the patient comes in at the very start of their sickness, it can be 20 the next day. It varies a lot. Thus, you must separate these two cases completely. And you must also remember that the first case is not a false positive, it is certainly positive, but not clinically relevant, so I would say.

FALSE POSITIVES ARE VERY RARE
CIESEK: (cont.) We must also ask: how do we get false positives in the first place? I must say, they are very rare. If you look at the reports we issue, you would only see false positives rarely. There may be contamination. That has been shown in the press in the USA. We do a lot in the lab to prevent this. In terms of PCR, the staff are specially trained. And each PCR run contains not only one, but usually 50, sometimes 80 to 90 probes, and in every run, you have controls. There is always a positive control, one which you know is positive, and a negative control. And you can even use these controls for individual steps. For example, you can check to see whether the extraction, the first step of PCR, was successful. This means that false positives are extremely rare. Even if the test gives a positive result, but you have 90 other samples along with it and they are all positive, the medical technologist or doctor would interfere and tell you to repeat the test, this won't be okay. That's the difference between analytical specificity and clinically diagnostic specificity. They are two very different things because there are many steps between doing PCR and the findings.
And this is probably not completely appropriate now, but I would like to thank all of the medical technologists who have been working on these tests since February, often seven days a week, who have simply done an unbelievable amount for us and have really been thanked far less often than they should be. They are doing a great job. And they are trained to make sure that such mistakes don't happen.
And this is the difference between technical and medical validation. The technical assistants do technical validation, for example. They loook at positive results and  see whether the nucleic acid in the sample was identified, and whether no cross-contamination can be detected...

HENNIG: With other viruses.

CIESEK: Correct, and that all necessary controls have been carried out. Only then do they give a positive or negative result for technical validation. There are no negatives when it comes to PCR, but rather, results that are under detection limits. This means that these specific nucleic acid sequences wre not found in the samples, but also ensuring that there wasn't anything inhibiting the reaction, for instance through suitable controls. And also that all other controls, especially positive controls, correctly show positives, because for instance if you have a run where the positive control is also negative, you'd have no choice but to repeat the run. And that will be done as a rule—that's the so-called technical validation, There is also medical validation, that are usually done by the doctors. There is also a plausibility test, as I always call it, namely, seeing whether the findings suit the clinical contex as well as preliminary findings that we often have with patients. The results of the technical validation are of course also included. Finally, an interpretation of the findings, an assessment, is also made.

HENNIG: This means that the statistical value of the pretest possibility, which is always critically observed in relation to the specificity of the PCR test, is for you as a clinical practitioner a purely theoretical, on-paper value, which you cannot however find in your studies and in day-to-day sample testing.

CIESEK: Exactly. It is of course important to know because it is related to the quality of the PCR. If the specificity and sensitivity are bad, I would not choose such a test, but would look for a better manufacturer. You also have to take it into account when interpreting the conclusions. But of course, this is a very small part of our work in the laboratory. And I think there's something else to say—there is a curve when it comes to Ct, which rises and eventually becomes positively inflected. And you can see this curve in the laboratory and then calculate if it is really a positive, or if it is simply a crooked line that at one point crossed the threshold. This can happen, for instance, if you put in material that foams a lot, this can give a rather noisy signal which is then detected as a positive by the machine. And then the TA or the doctor would look at the curve and say, "Well, that's not real." And a lot would be corrected. And this would be used then for these INSTAND round robin tests. A round robin test has a completely different function than to ascertain the specificity of a test, namely, to compare results between laboratories as a sort of quality control.

HENNIG: INSTAND is an association, a non-profit organization that carries out standard quality testing for laboratories.

CIESEK: Exactly. We as well are required to do that if we want to be accredited, as well as virtually all laboratories or at least many of them. As a rule, round robin testing is done at least twice per year, not just for SARS-CoV-2, but rather for all viruses that we detect diagnostically. There are samples that INSTAND prepares, whose contents are known and which are then given to laboratories. And as I said, this isn't to conclude anything about the tests in general but rather an indirect measurement of the tests, used to assess the laboratories.
This means that if I do not pass the round robin test, I would know that we would have to re-evaluate
ourselves and then do another test. Or maybe something's not right with my method. And this very round robin test was done in April for SARS-CoV-2, so relatively early, and did not even factor pre- or post-analytics or medical validation in. They didn't have to prepare and interpret a result or compare which swabs were used. On the contrary, it was only about the technicals in executing PCR.


-----PCR CONTROLS-----


HENNIG: Nevertheless, I have to ask something about the findings because this topic is very important and has been discussed a lot. You have said before that if we look at the findings, there are very few false positives in there. How do we know that exactly? How did you test the findings before you could make such a conclusion? Are there subsequent controls?

CIESEK: Yes, for example, as we have aid, we detect two or three genes. If you have a sample in which one gene is negative and the other has a Ct value of 37 or 38, we would never get a positive result because someone would have look at it and say, "That could be something. Please send us a new sample." And with a patient in the hospital, you would say, "Send us a new sample, let the patient be isolated for this length of time, we must first find out whether this is accurate." And then you find out that many of them aren't actually positive going by the new test and then, for instance, the symptoms do not fit at all. And this is the control that we still have to carry out. So we also speak to our colleagues and then find out if what we've seen is likely or not. And I've always find it quite unfortunate, because PCR is really a very sensitive method. The example isn't perfect now, but I think that it helps people think about it a little. So if I think for instance about the blood donation service or the blood donors, all of them are tested in pools for HIV, HCV, HBV. If a blood bag was positive for HIV via PCR, via a PCR test, nobody would give the bag to a patient. And of course, the donor shouldn't have AIDS. Do you know what I mean by this? That doesn't mean that PCR is wrong, but rather that you can detect these viruses without having a complete picture of the disease in the sense of AIDS or Covid-19 and without having to have been intubated and ventilated. You can still be infected, but may simply have a very mild or even asymptomatic course of disease.
REASONS FOR FALSE NEGATIVE TESTS
HENNIG: If you think about reducing the duration people are supposed to isolate themselves for, the effect of these borderline cases in social and private live may not be so big any more. We have spoken very briefly about sensitivity already. There is one more topic we have to briefly discuss, false negative tests, they exist too. If the sensitivity of a test is high, what factors lead to an infected person actually being correctly identified as such?

CIESEK: There are many causes of false negative tests. For example, you tested too early. Imagine that you had contact with someone with SARS-CoV-2 yesterday and go immediately to the doctor today. That would be too early. The test could be false negative even though you would've just been infected yesterday because of that contact. But after a few days, it would be positive. Then there are technical problems, such as the sample taking too long to be processed or being stored improperly. How the smear was taken is a big problem. We have spoken a lot about it, if you do not take the sample properly, have too little material, have a bad swab brush or if the pathogenic material is very unevenly spread out—all of this could lead to a false negative result.

HENNIG: Do you know anything from your practice about the best possible time during the course of infection to get a PCR? There have been meta-analyses, studies that have evaluated different studies because as you have said, a patient may have only been infected the day before. We know however that the viral load is highest around one to two days before the beginnign of symptoms. Is there any evidence supporting this or is the question still open?

CIESEK: Yes, it is said that testing is best around the start of symptoms. But it is very difficult to know when the start was, because you only ever know this in retrospect. But we know for certain that at the start of symptoms, give or take two days, the viral load is highest. To someone like this you would say, "Come back in two days" and then do another test. That is also the purpose of these antigen tests—even though you have a lower sensitivity, meaning that you won't be able to detect a big range of viral loads, you compensate for this by the frequency and speed of such tests. That's the idea behind the antigen tests and their use. I don't think they're 100% ideal for symptomatic patients because I really want a sensitive method. But for asymptomatic ones for instance, those who want a bit more security before visiting their grandmothers in a home for the aged, or for travelers, I really find it a good idea to use such tests.


-----ANTIGEN TESTS AS A QUICK ALTERNATIVE-----


HENNIG: This is also being discussed with regards to big cultural events, for instance. But first an explanation. Antigen tests do not detect the genetic material of the virus, but rather, the proteins on the viral envelope?

CIESEK: Exactly. And yes, I think that'd work for these cultural events. I only think this because of my own gut feeling, even after the first few days in our study, if you do an antigen test, you would need some kind of process or a link to someone who can then do a PCR. These tests can be false positive and if you do the tests at home, without controls, without supervision, some people may be afraid because they do not know how they should behave and who to turn to for help. I believe that if anyone is to do these tests at home, there must clearly be a telephone number or contact address so that if someone is positive, they can go and, as much as possible, get a PCR test done on the same day. That would be the ideal place for me for these tests.

HENNIG: You just alluded to your study, a study about teachers testing themselves with these antigen tests. Is this an experience that you have had, that teachers may feel unsafe because they have a positive test result?

CIESEK: I really am enthusiastic about this, and I would like to take this opportunity to thank our teachers for participating, so many of them are involved in this study. I find it really great. No, we have simply seen to it that they have a hotline which they can use to contact us and would then be treated. If a test is positive, we would make sure that the probe is taken and a PCR done on the same day. It is important because this leads to uncertainty, if there are two stripes on the test [indicating a positive], who should I turn to? That's the experience. I think that with teachers this works very well because there are exact procedures and a process detailing what happens if a test is positive. If there were no such thing, I imagine it'd be difficult.

HENNIG: According to you, that means that under certain circumstances, antigen tests can be carried out by laypeople and not just by trained specialists.

CIESEK: That is something that I think politicans should decide. In terms of implementation, it's possible. It's similar to a pregnancy test. It's a bit more complicated, but it would go very well if you had a video and a detailed description of how to carry out the test from a layperson. That works very well. But the question is political, if such a thing is desired. And as I said, I regard it as very beneficial if there are very clear instructions on what to do, for instance: what do I do if the test is positive? Who should I go to? If possible, this will happen every day of the week. I imagine it has to be organized.

THE LAMP TEST
HENNIG: This entire test scenario is a question of capacity in lieu of the rising number of infections. I would therefore like to briefly talk about another test because in August there were reports that Great Britain's focusing more on the so-called LAMP tests. RTLAMP, in the case of RNA viruses, is not a new procedure, but rather another method comparable to PCR, with which one can amplify nucleic acids from sample material. How does it work and is this a possibility for Germany?

CIESEK: There are different research institutions that have also developed a LAMP test in Germany, for instance in Heidelberg or at the Fraunhofer Institute. The advantage of this is that you don't need a cycler because this is an isothermal amplification, i.e. one that happens at the same temperature throughout. But you need similar components. You need an enzyme, primers, these...how did you describe them before?

HENNIG: Building blocks.

CIESEK: Exactly, these building blocks as well as nucleotides. And you don't use a cycler, in which you cool and heat repeatedly, but rather, so-called primers, which are a bit longer and form loops. In this way, you get a faster duplication of the gene or segment you're looking for.

HENNIG: You only heat once at the beginning.

CIESEK: Yes, but you see, the problem is that you need the exact building blocks. Another problem is that you will certainly have to compete with the PCR providers. I have spoken to a company that synthesizes these nucleotides. Very few companies do this. You need both processes. Then you of course need these plates in which the reaction takes place. You need them with PCR too. And with the LAMP test. You also need swab tubes, of which supply is tight. And this is an elegant method often established and developed by scientists, but has so far not yet been used very much in the medical field. It still hasn't been introduced to the market. I believe this is because this is very theoretical. You must be able to apply the whole process in a medical laboratory. For instance, our cyclers, if they were to become obselete—they are currently connected with our laboratory software and communicate with it. They practically shove their results into our laboratory software, and thereafter, this would no longer be possible. And you still have to ensure that when you do 1,000 LAMP tests, every patient gets a result. This is an insane amount of logistic effort. I believe that in this study, using LAMP would be a success. But until now, the application and the marketability is missing for this to be a part of daily routine in a medical laboratory.

HENNIG: Nor is there any alternative if the reagents, the ingredients for the PCR test, become scarce, as I have understood it.

CIESEK: Yes. And of course you need an accredited medical laboratory as well as trained personal. There are strict guidelines about what they have to be able to do and how they are trained. You must thoroughly train them to use a new method, which would take weeks or months before you could introduce them to a laboratory under these accredited conditions.


-----WHEN A TEST MAKES SENSE-----


HENNIG: Let's go back to the here and now. Many doctors are faced with the question of, "When should I test? When should I take a sample?" This is a question that has been asked of us by listeners who are either patients or doctors themselves. Capacity is not unlimited and the symptoms are not always clear—it is difficult to differentiate between this and the flu. What do you advice, if you wish to give any, to doctors who want to know when they are to take a smear sample and do a PCR test for the virus? With which set of symptoms?

CIESEK: Yes, the Robert Koch Institute recommends that each patient be tested for acute respiratory symptoms, severity and loss of smell or taste. And as I understand it, this is very difficult. If your practice is full of people with cough or colds, that would mean that you'd have to test everyone. Of course, that's the ideal situation. There is also a national testing strategy from the Ministry of Health, which also states that all symptomatic people should be tested, and that this should be given the highest priority. Unfortunately, they have not defined what "with symptoms" means. And I think this is part of the problem for colleagues. And as I have said, I would not want to go against this at all. If we have a large testing capacity, it makes sense to test the symptomatics before everyone else. If I had a practice now or had the problem that there wouldn't be enough tests, I would focus on a number of symptoms. Loss of taste and smell are relatively specific. And if it wasn't possible to test everyone, I would give a questionnaire out in order to assess the risk for myself. I would ask for instance: Have you traveled abroad within the last 14 days? Did you partake in any risky behavior such as family celebrations in closed spaces or certain "Superspreader Events" which are not yet well defined? You'd write down a few situations on the paper as examples.
Whether you have had contact with someone with Covid-19 is an important question, as well as whether someone within your social circle has. And I would also ask patients whether they have risk factors for a severe disease course. For instance, an older person with diabetes or a metabolic syndrome. That is relevant in any case to detect the sickness and perhaps treat it with antivirals. Thus I would like to try to classify the risk with a questionnaire or a checklist and then primarily test those at risk or who have partaken in risky behavior. Of course, everyone with an infection who shows symptoms should isolate themselves and not go to a party.

HENNIG: That means that if I as a patient would like to help out, a contact diary, which has already been mentioned several times, could help. That way, I could at least have prepared myself somewhat and can then tell the doctor, for instance, that I go to boxing training on Tuesdays and to choir on Fridays, and that last week I was at a grill party.

CIESEK: Yes, that'd be good, of course. Of course that'd help. If you ask me what I did last week, I wouldn't be able to say so right away. At the moment, I still do not know how politicians define these events. The most commonly listed superspreader events, for instance. It would be much easier then for individuals to understand what the danger is or how much they should pay attention. What could they avoid? But if you get a cold or symptoms, you would know the important thing, namely, where you for instance could most likely have made contact [with the virus]. It would be nice to have such a list. That would make it easier for doctors in private practice to make such a questionnaire and also to help citizens avoid these exact situations.

HENNIG: Something for politicians as well as ourselves to do. As far as symptoms are concerned, it still remains difficult to differentiate Covid-19 from the flu. Two weeks ago we also talked about this in the podcast. Is it possible to test for both in one test and then have separate results?


-----MULTIPLEX TESTS IDENTIFY SEVERAL PATHOGENS-----



CIESEK: At the moment, suppliers are preparing tests, the so-called multiplex PCRs, which can detect both pathogens. There are test manufacturers that can detect even more viruses at once. They are gradually appearing on the market. Influenza comes mostly earlier in the year, in January, February, so they will be certainly available then, so with one sample and test, we will be able to detect both viruses. That is not at all uninteresting, as patients have been seen with co-infections. That means that both pathogens were detected in them. That is very rare. Nobody knows why that is rare. But it is possible to be infected with both viruses. And of course, it helps to be vaccinated against the flu. Even if the vaccine doesn't work 100%, I have yet to see a serious course of illness in someone who was vaccinated. By "serious cases", I mean someone who had to be taken to the ICU. So even if you do end up getting the flu, you'd be protected against serious courses of illness.

HENNIG: Regarding the multiplex tests, would they be a possibility for the ambulatory clinics that Jens Spahn mentioned yesterday, at least for the start of next year? Or do you believe that the capacity would again not be enough?

CIESEK: That depends on which multiplex PCR you're talking about. There are the so-called point-of-care tests which are also based on PCR, and work with cartridges that you can even put in such ambulatory clinics. Within one hour, you would get a PCR-based result. The problem is only that there aren't enough cartridges for this. So these firms are not able to keep up in terms of production. This system should be reserved by hospitals for critically ill patients, where you need a result very quickly. And that would of course be ideal, but I really see problems of capacity with this. Not every laboratory carries out multiplex PCRs, if you take the longest-lasting variant, which requires four or five hours, you must see that this is frankly very expensive. We have some PCR systems which can detect over 20 respiratory illness-causing pathogens, including rhinoviruses and influenza. That is really expensive and requires a lot of effort. You must ask over and over again: what really is the use of this? Many of these pathogens don't even have clinical consequences. Whether someone has rhinovirus or not but doesn't know it is probably not so essential [to know] for someone who simply has the colds and goes to an ambulatory station, compared to someone in the ICU, where someone needs to know what is going on.

HENNIG: So let's wait and see what the politicians decide about such centers and how they are to be equipped. Dr. Ciesek, we have just been talking about questions from the listeners. To finish off, I would like to take up a concrete question about day-to-day life, which you can answer from a virological perspective. Namely, daily life with masks. That is really not a topic for virologist but rather for hygienists. Nevertheless, you can provide us with an explanation. We often hear about washing as hot as possible, at 60 degrees [Celsius], at 90 degrees if possible. On the other hand, there are statements from American epidemiologists, which say that even 40 degrees would be okay for instance, if washing solutions are used. The coronavirus is an enveloped virus. What does this mean exactly fro the use of soap and temperatures?

-----SOAP DESTROYS THE ENVELOPE OF A VIRUS-----

CIESEK: Yes, this is a classic virological question. Some viruses have an envelope, a lipid envelope, one made of fat. And there are viruses without an envelope. For instance, the famous rhinovirus has no envelope, but SARS-CoV-2 does. That is important for hygienists and for dealing with the viruses—how easily can they be broken, i.e. inactivated? Viruses with an envelope are much more sensitive to damage than viruses without one. Noroviruses or other rotaviruses, for instance, have no envelopes. What the CDC and the Robert Koch Institute say are both correct. I believe that depends a little on what you're looking at. Based on temperature alone, we know that at 60 degrees Celsius, the virus is completely inactivated at 30 minutes, and faster at 90 degrees. However, that is under conditions without washing solution or soap. When you do have them, they destroy the envelope of the virus. At that point, temperature isn't so important. That's why I keep saying that when you wash your hands, please do so with soap because washing your hands without soap doesn't do much, soap is really essential and destroys the envelopes of enveloped viruses and inactivates them. It's a different scenario with rhinoviruses, for example, who don't have envelopes. Here it should be said that they are more stable. They survive longer in the environment and are much harder to inactivate. When you take a look at disinfectants, which can be bought at pharmacies or drugstores, there are different classes. For instance, there are normal disinfectants. Then there are limited virucides, limited virucides plus and then simple virucides. Normally, you have limited virucides, which work very well against SARS-CoV-2, are alcoholic, i.e. contain alcohol, and thus destroy the envelope. Then there are vircuides that work against unenveloped viruses such as rhinoviruses too. But they are for special situations, for when we work with unenveloped viruses in the laboratory. Or they may be used during an outbreak of gastrointestinal diseases, because they aren't usually very good for the skin. We use them in the laboratory if we must, but not usually, because they can be very irritating to the skin. Limited virucides plus means that they are also effective against rotaviruses and noroviruses. But as was said, the normal standard disinfectants, the limited virucides, are enough for SARS-CoV-2, but not for rhinoviruses.

HENNIG: Regular soap does the same thing too.

CIESEK: Yes, soap does that. That's right.
NOTES
1.  A measure of how much of a particular genetic sequence is present in a sample at the start. It measures the number of PCR cycles required for the amount of copies of the sequence to exceed a pre-determined threshold. The lower this number, the larger the amount of sample present at the start.