#CovidUnknowns: New variants

welcome everybody i'm dr cameron abbas executive editor of the bmj and i'd like to welcome you on behalf of the bmj and our co-conv the university of bristol and dr allison pollock of the university of newcastle to the latest unknowns webinar and today's session will focus on new variants the new the coveted unknowns concept originated in an article by george davie smith in the bmj in october last year and a swift it was swiftly followed by an all-day event in november which attracted huge global interest from the global uh community it's clear that george allison and fiona godly our edge in chief uh touched on something that there was a real hunger for and we've been holding these events every two weeks ever since and each one of them focused on a specific aspect of the pandemic so so far we've covered schools and children testing vaccines uh and recently we had uh zero covet so uh before i hand over to phil hammond i want to clarify one specific aspect of this which is the naming of variants and this is something that troubles me a little bit um we all got rather upset when donald trump was calling uh the coronavirus the china virus and um i guess that was because he was being confrontational and derogatory we've we tend to call the variants by the region that they may have originated from and you know that may well not be the the right thing to do the alternative is to use uh a numbering nomenclature which we'll hear more about today as well so before we do before we launch into the session if we cause any offense to people uh in different parts of the world brazil south africa uh the people of kent um we apologize in advance we're here to help this is meant to be constructive and no offense intended and without further ado over to phil hammond our master of ceremonies and phil introduce the first session and lead us through today's events over to you phil thanks uh very much cameron good to see you getting your pre-apology in nicely done now this is the same format as all the others um we have nine experts each speaking for about eight to ten minutes we welcome your question so there is a q a button at the bottom please tell us who you are we've got slightly longer for questions at the end but we we don't get through all of them but please uh keep sending them to us because they're really interesting uh that feedback uh you can tweet about this on the hashtag covid unknowns to keep the conversation going and of course it will appear on the bmjs youtube channel so explore youtube we we have about 2 000 people registered for this but then tens of thousands of people uh watch the various seminars later so please share the good news we're going to kick off with new variants what are they where do they come from and what might they do to chair this session i'm going to hand over to george david smith george thanks phil i'm going to go straight in to this session by uh introducing uh any other who's from edinburgh university who's going to uh introduce is going to introduce new variants to us now i've introduced her again thanks very much i'll just share my screen now and hopefully you can all see that yes okay um so it's a pleasure to be here talking today i'm going to start us off by giving a quick overview of sarskoff 2 the new variants and some biases in our ability to track these variants i wanted to firstly rewind back to a snapshot of the world in january 2020. the colors here show us how severe lockdown measures were in each country at the time zero being normal and 100 being complete in total lockdown um if we fast forward to may of that year there were four million cases worldwide and the entire world had come to a standstill looking at the numbers today we have detected 123 million cases of covert worldwide and when we're talking about sarskoff ii it's a fairly middling rna virus it doesn't evolve as quickly as flu but it has a lower substitution rate than mirrors and ebola um but it does have a shorter serial interval and this ends up meaning that for sars curve 2 we see on average one new mutation transmitted every couple of weeks today we've got a day focused on variance and when we talk about variance that term can actually mean a couple of things a variant can just mean a mutation and it's often how we distinguish one virus from the other but more recently and in the context of sarskoff 2 the term variant has come to mean a variant of the virus with a combination of mutations or constellation of mutations that may have biological significance and as you're likely very aware there's been a number of variants of concern that are being tracked at the moment and are in the media we've got b117 b1351 p1 they've all got various names uh in different contexts and another variant that has been uh that has had some attention drawn to it is a recent lineage um with a mutation in the spike protein that's predominantly seen in california and has then been exported to a number of other countries but the thing is variants aren't a new thing there are new variants but sciences have been tracking variants of sars cov2 for every year now and one of the early variants d614g was associated with increased transmissibility and came about just prior to the large northern italian outbreak early last year and shortly after that we saw this huge global sweep across the world of this d614g variant so we've been identifying new variants tracking their spread nationally and around the world and to track variance you need to sequence the sars curve ii genome which is a lot of people have been doing uh so far as of this morning there's 871 thousand sarskov ii genomes on this aid we recently had a bit of a milestone ourselves pangolin is a tool we wrote about a year ago now to assign lineages to a sequenced sarah's kov2 genome and as of a few days we hit the million unique sequences assigned with the web app mark and these genomes themselves are coming from around the world and at the moment there's now 166 countries with sarskof2 genome sequences publicly available so because we have this resource available we are able to track lineages of concern and variants of concern like for instance b117 now i'm sure no one has escaped hearing about b117 over the last couple of months but to recap it's a lineage that arose in september last year in kent likely and it's characterized by a burst of mutations at the base of the lineage some of which are mutations that are very likely to have biological significance and there is evidence of this lineage having increased transmissibility we've been tracking this lineage since december and as of this morning there are over 200 000 b117 sequences on jisaid from 102 different countries the vast majority of those sequences on gsaid are from the uk because the uk does an incredible amount of sequencing and counts of sequences are very much a product of the genomic surveillance in that country um but there are quite a few countries that are now reporting local transmission of b117 we can talk about the frequency of the variants within a country as well but for most countries it is very difficult to say anything meaningful about frequency many countries are specifically targeting cases that have travel history say from the uk or from and from other countries that have lineages of concern so this doesn't mean that they've got 100 frequency in the community for instance also we're aware of many biases that are associated with the data whether it's genomic surveillance intensities whether it's targeting specifically travelers whether it's targeting um for b117 in particular um the spike gene target failure so this pcr signature that can indicate whether or not a sample is b117 and there also may be a bias in what genomes actually get uploaded to gsaid perhaps variants of concern are more likely to be uploaded also we know that the uk has approximately a two week lag in community sequencing and this may well be true for other countries too so we're very aware of all of these biases and we can try to account for some and we got our hands on some flight data and which gives us actual passenger volumes from london airports in october of last year and this gives us an idea of the actual people movement and to try and account for differences in genomic surveillance efforts across different countries and this is what the data looked like when we first got it back on the 27th of december so here we're showing any country with more than 5000 passengers leaving london in october and you can see the top three destinations are spain italy greece and at the time there wasn't a huge correlation between number of sequences and number of people but since then slowly but surely we've actually seen these countries report cases of b117 we know that the virus moves with people and this has only confirmed that even more and almost every country on this list now has reported cases of b117 what i would like next is to show movements from later from november december from this year because i think the dynamics of why people were moving will shift a lot since october where people maybe were going on holidays still which spain italy greece and december where people are maybe returning home from the christmas break um we're also tracking media reports um of either uh imports into a country or local transmission to try and counteract the lag of sequences getting on to just aid and as of today there's been 114 different countries reporting b117 and local transmission is being reported in various parts of the world however there's still a lot of countries only sequencing travellers so we may not actually have any idea of what's already circulating in the community for a lot of countries in general what we have seen though which is very promising is that most countries are closely following reports with upload to just aid which is really great because it means we can continue to track these lineages and variants by sequence count i won't go into a huge amount of detail on um other lineages of concern there'll be people talking more about them later on today but um we've also been tracking b1351 or 501y v2 and to date there's 67 different countries reporting cases of this lineage that was first detected in south africa we also have flight data for that too one thing that i'd say about this particular flight data is that it may not quite be as representative of ultimate movement of people and we've got access to destination on a single booking which may not be final destination for instance the top two here are ethiopia and the uae which are hubs for connecting onward flights also south africa is not an island so there could be significant overland movement that's not necessarily being accounted for in this case in fact we know that botswana has about 70 prevalence of b1351 and that isn't represented in this figure we're also looking at lineage p1 first attack in brazil and we now see 36 different countries around the world reporting cases of this lineage we can track um passenger flight volumes as well from manas and we don't have particularly more recent data on this but it gives an idea of where people have been traveling to from manas so i thought i'd finish up with a known unknown summary so we know that the virus will continue to mutate and evolve and spread and generate new variants but we don't know necessarily what those variants will be or we don't know where they'll occur or when they'll occur um we need to sequence to detect these variants and places with little genomic surveillance may have sars club 2 diversity that's largely unaccounted for we know that there are biases in the data set but we don't know the extent of this or how much we can infer from the biosamplings of the populations we also know that the virus moves are people so we can try to account for some of these biases so i'll leave you there thank you that was great thanks thanks very much just uh to remind people that you can put questions uh in the q a and they will be picked up uh and asked either the end of the session or right at the end of today so uh i'd next like to introduce ravindra gupta who's from cambridge university and uh he's going to talk to us about uh things which influence the emergence of variants my dexter hi hello i'm just putting slides up thank you very much for the invitation to speak at this um uh webinar and very much looking forward to the discussion that's going to come so i'm going to share screen one second okay is this working can somebody perfect that's perfect right okay all right so we're going to be discussing what we know about why um uh new variants have emerged and some of the factors that may be responsible um and just at the outset of course we don't know the answer for sure and that's why there's a question mark there but um just uh uh just briefly to start with sarskov two transmission um uh luckily um uh ernie has covered some of this already uh the globalist mutation rates estimated at two mutations per month and uh this is driven partly by its short incubation period and the fact that you get asymptomatic transmission so what what essentially is happening is that the viruses um uh is delivered to your nose it infects cells within six hours it's making uh rna and it uh shortly after that it will be budding away and infecting new cells so you can imagine a very short incubation period lots of virus produced very quickly prior to the interferon response and then um a transmission uh largely during the asymptomatic period so you get you get the opportunity for this virus to jump from one host to the other with very little need to encounter or overcome uh immune pressure and that's probably what's kept substitution rates or or mutation observed mutation rates um uh uh relatively low but the sheer number of them of course compensates this and it was described as the serial interval earlier so you get it when you discuss certain mutation rates you really have to understand the biology uh underlying uh the the estimations in order to understand the limitations and what it really means so all of these factors as i mentioned can have contributed what we think to relatively low selection pressure this is going back to the beginning of the epidemic and of course things have changed since so soon after the the the the pandemic was declared it was recognized that some people were shedding virus for uh considerable numbers of days and weeks and was coined uh long-term shedding uh and increasingly it was realized that uh shedding was enriched in in individuals who had sub-optimal immunity in other words they they either had a biological reason for immune suppression or they were they were treated with um immune suppressing drugs and the the kinds of things people were seeing were uh deletions especially in the end terminal domain of the spike uh and and uh and uh mutations in parts of the genome including including the receptor binding domain and then the sort of the this begs the question as to as to whether these individuals are the source of variance of concern so i'm going to tell you about a little bit about the some of the cases that have been described of long-term long-term shedding and and and to tell you about one that we uh studied in in great in in significant detail uh in order to uh uh arrive at some some kind of understanding as to why these things arise and and what they may mean so this this phylogenetic tree um basically shows you this is a sort of family tree of of different viruses and they represent different individuals if you can see my point here there's the case patient that i'm going to describe to you and then there are a number of these other patients uh two from previous studies one from colleagues of ours in boston uh uh that just they described an individual who was positive for a hundred days and who had an underlying disease um sorry an underlying disease process uh called antiphospholipid syndrome that had been treated with a lot of steroids um and therefore that individuals immune suppressed um we have a an individual down here uh described by other colleagues in the u.s and this individual very interestingly um shed fur for for three months um but had relatively modest immune suppression and was completely asymptomatic for most of that time in contrast to this the boston patient who became quite severely unwell and died and including the patient i'll tell you about in our case who also unfortunately died so you've got to range from asymptomatic to a life-threatening disease in these chronically infected individuals the three there are there are two or three other control patients in this group that have that shed for more like that between 30 and 60 days and what you can see from the tree the reason that i've put them in this tree is because um uh it's just to give an impression of how much the viruses have changed um uh during the disease process uh you can see these these individuals who are infected for shorter periods have have relatively short branch lengths this means a relatively a small number of mutations that have happened in those individuals and you can contrast that with the the the boston patient for example here that has uh that generated a significant amount of diversity over the 100 days of sampling and then the the case that i'm going to tell you about which had a similar level of diversity and and um and evolution over time again around 100 days so so this represents a few of the cases in the literature but they have some common uh com things in common that we'll discuss just a little bit of a quick lesson in in solskov 2 and and and how it interacts with cells it is fairly complicated but i'm going to keep it as simple as possible there's the virus there's the cell the virus has um on its surface this spike protein uh sitting on on the surface and that is responsible for engaging a receptor in the target cell that it wants to infect in other words a cell in your nose uh or respiratory tract and it it needs this receptor and ace2 engine tension converting enzyme too and following engagement with that you can see this is a blown up sort of picture of that interaction you can see here that a number of things have to happen you need cleavage of the the spike protein into two pieces s1 and s2 and that exposes this thing called the fusion peptide in the virus that basically inserts itself into our our cells and then causes the membranes to come together and then you get this thing called fusion where the the membrane around the the virus becomes uh one with the uh our own cells and allows the um the virus genetic material to come into our cell and that is basically um entry and infection and that rna is then able to make new copies of itself so that's what we're trying to inhibit we don't want this because if we can block it we can of course stop infection um but the virus is obviously trying to optimize this in order to get into ourselves um if we zoom into that interactive site here a bit more you can see the cell up here and ace2 and how it's interacting with with certain residues key residues in in the virus spike receptor binding domain or the rbd which you may have heard of the rbm is the motif it's a specific region which actually does interacting and you can see some of the key mutations here that you've heard in the new variants and they're all sitting at this interface the reason you can see these funny colored antibody things is because by a quirk you make you may say not only is are these residues important for interaction between the two proteins to enable entry but these um this region also is highly immunostimulatory in that when we make immune responses a lot of them are directed against this region so you get a situation where the virus is trying to infect a target cell but also trying to escape from these antibodies that are going to block the interaction so uh and what's been observed is that you get certain mutations in some of these residues that not only enable the virus to stop binding the antibodies in other words the binding the antibodies become uh uh uh become uh ineffective but the virus is still able to hang on to the um to the res to the target cell protein and and do its entry so that's why uh that's the fascinating sort of scenario that's occurring where the virus is able to simultaneously avoid antibodies and maintain infection of cells and that's the balance of course of fitness virus fitness uh that that we'll discuss a bit later so um this this diagram just shows you some of the the three main variants of concern um this is the spike protein it's uh it's formed of three units that are pretty much identical and then joined together um to form this club-like structure and this is the the piece of the virus that all a lot of our antibodies are directed against and it's the the piece of the protein that enables the virus to do it some uh to it to enter cells so the reason i've um highlighted an area in yellow is that all of these these three variants have this n501y mutation uh and um it's been shown by um us and others that this increases uh affinity for the um uh between the um between the the spike protein and and ace2 so it enables the virus to bind more tightly but also does some contributes to virus expensive escape but in the um in the b351 and p1 variants from south african brazil respectively uh you can see that it's uh accompanied by two other mutations e484k and and changes at four one seven in both cases and and and you you most of you will have heard of all these mutations because they're the ones that we worry most about in terms of vaccine efficacy so there are common pathways being taken by all of these variants which tells it to you know speaks to a concept of convergent evolution whether uh viruses from different parts of the world are choosing common pathways because they enable the virus to simultaneously escape from our immune systems but they also allow the virus to to efficiently infect our cells and therefore propagate so just briefly i'm going to talk to you about our individual who had a lymphoma he was immune suppressed um uh and down here you can see the ct values this is the amount of virus in nose and throat swabs and you can see there are some ups and downs uh that could be due to virus some uh sampling techniques or or or or actually the viral load the true viral load in the respiratory tract you can see it never this is this is the limit of detection and as you go up there's less virus so you um you can see that there's virus detectable throughout that hundred day period there are a number of treatments um used during this period that include rem deserve and in in purple here and you can see that dexamethasone was used to reduce inflammation and of course um convalescent plasma was used in this individual because of failure to clear the the infection over the period between may and july so two but two units here and then a third unit uh in august just before unfortunately the patient passed away um and so the story here really is about uh the trajectory of the virus uh in the face of uh pressure from convalescent plasma and the antibodies contained in that plasma so here you can see what's going on uh what happens in over time in the individual there are the treatments first of course around us the second course convalescent plasma here at day 65 and what i'm showing you here is what's happening the viral population in those nose and throat swabs um this is done by next generation sequencing here you can in dotted line you see the the viral load can you uh close up yep and you can see here that um there are a number of different uh populations emerging we can see here there's a there's a a mutation that a double mutation that emerges at day six uh day 82. uh this fades away um as the antibodies are fading away or or diminishing from the blood and then when the third course of converse and plasmas is administered here you see the resurgence of that group of viruses and and that's shown in green here in the phylogenetic tree um just this is a darker machine where those mutations are one of them's out out here in s2 and one is in the n-terminal domain and they're both hanging out on on edges when we make those those mutations in an artificial virus we see that uh one of the mutations increases the infectivity on the infectiousness of the virus and one of them causes a defect and that's interesting because the one that causes the defect is actually one that enables the virus to become a little bit resistant to the antibodies in the convalescent serum in each of the three units given here so you've got a very nice balance between one mutation conferring escape from the antibodies in the convoluted plasma and the other mutation coming along to restore the defect in infection that that escape mutation brought classical virology we see it with drug drugs all the time and so that's how we think that new variants are are emerging within hosts they're learning to adapt the immune system they're acquiring multiple mutations some are designed to escape some are designed to restore infectivity uh and um uh uh to sort of show that this is this is really happening the uh there is a lineage circulating and transmitting in a number of countries with the two mutations we found in this individual we don't think that those infections are connected to our patient but it shows that this pathway that's described within and within a host is out there as a variant and it is transmitting thank you very much and i'd just like to acknowledge everyone on the slide that's fascinating thanks very much i will move straight on to our third speaker in this session who is wendy barclay who's from imperial college and is going to tell us about lab studies of viral variants ravi might need to stop sharing sorry thank you and i hope hoping everyone can see my slides that's perfect yeah thank you very much okay so um i'm here really speaking on behalf of a consortium of virologists in the united kingdom called g2p which is genotype to phenotype so as anya pointed out there's a huge amount of sequencing going on and these amazing constellations and evidence of what we call convergent evolution of different variants of the virus happening around the world and as ravi's shown there can be a huge number of mutations accumulating in for example immunocompromised patients infected over a long time with these viruses and what it's important to to be able now to to use a combination of laboratory-based studies and epidemiology to understand which mutations really matter which ones are making a difference and in particular which ones are allowing the virus to transmit and spread through the community um so anya of course already mentioned the b117 and most of the talk i'm going to give today is going to use b117 as an exampler virus and here is just another way of showing that epidemiology as the b117 or the uk variant emerged in the autumn and spread and now of course predominates hugely uh in the united kingdom and has spread around the world and there's clearly epidemiological data that i think lugo will cover later which illustrates that this virus is very transmissible and so what we're interested in doing is understanding what are the genetic features of that virus which enable it to transmit so well and in order to to think about that we need to break down i think transmission into the various processes that underlie it so of course transmission happens when an infected person sheds some virus into the environment and this emitted virus must pass and survive in a viable infectious form perhaps in the air perhaps on surfaces and then find its way to an exposed person and at a dose which is reached to that exposed person then initiate infection and so if we think about what factors of a virus infection might impact the efficiency of that process it could be for example that a variant allows increased shedding uh due to higher replication in the person who's infected or perhaps a longer period of virus shedding meaning meaning that there's more chance of this person meeting the susceptible individual it could be that the virus causes different disease uh spreads to different parts of the body making emissions more likely those emitted viruses might survive better in the environment uh and or it could be at the other end in the in the recipient that a lower dose of virus is required to initiate productive infection and that could be for a combination of many reasons it could be that the virus for example is good at evading the innate immune response and then finally it could be that the virus actually has got an increased number of susceptible individuals perhaps because it's now capable of evading acquired immunity or perhaps spreading in a cohort of individuals who who previously have not been as susceptible as before so these are all hypothetical explanations um and i think what we're interested in doing here is to understand how the mutations that have been identified in various variants might impact on these processes now ravi's given a very nice sort of basic virology introduction so i won't labor this here's the virus it's basically a genome inside a coat and it's a 30 kilobase genome encoding various important proteins including that spike protein and i think anya already flushed up this list of lineage defining mutations of b117 variant and you can see there are 22 of them which perhaps does support the concept that they may have accumulated uh within a sort of long-term infected host and again these mutations that we find in the variants of concern both b117 b1351 and the p1 variant spread across the genome but there are important constellations of mutations in the spine so let's try and break some of that down again ravi's introduced the concept that the virus uses spike to bind to ace to an enter and there's this important molecular interaction here between the receptor binding domain of spike and the ace2 protein and when we focus in on that interface we can see these mutations here in our variants of concern enhancing that interaction what does that mean in the laboratory what does that mean in terms of what we can actually measure changing for the way the virus behaves in the laboratory so here's a a very nice experiment i've taken from a posting from scott weaver and the mini cherry bio archives this is a hamster transmission model hamsters are a good model for sars cov2 infection and do transmit the virus between them and in this experiment the hamsters were infected with a mixture of two viruses co-housed with sentinel animals and then four days later the transmitted genotype was was quantified to see whether one virus transmitted more efficiently and replicated more efficiently in the animals and this is concerning that these results concern the b117 virus and and you can see here that the one line here means that the mixture 5050 didn't change but for two of these viruses there was a significant increase in the variants that were transmitted and that shows us that the uk variant has a spike which enables efficient transmission in this model so we really can see that that genotypic change that's been picked up does actually translate into something we can measure biologically in the laboratory um ravi also mentioned this important process of cleavage of the spike protein to enable that fusion event where the virus delivers its genome into the infected cell and the cleavage happens in two places there's a first cleavage at this junction which is carried out by furin and that enables all primes for the second cleavage here at the s2 and that fury cleavage enables this efficient entry at the cell surface of cells which also express a protease like tempras ii which can activate that entry and here's some beautiful um data that um tom bergown at ucl has allowed me to show showing actually cov2 viruses fusing uh their membranes at the surface of human airway epithelial cells and enabling entry and we've previously shown how important fearing cleavage is for transmission if you if you have a virus which doesn't get cleaved by urine then it transmits very very poorly in relation to those who do due to a genetic change at the furan cleavage site in the opposite direction the the b117 spike has a point mutation amongst the the seven or eight that it harbors at the furing cleavage site which makes the cleavage more efficient you can see here the cleaved product is more prevalent uh in the b117 than it is in the wild type and you can measure that by densitometry and that enables more efficient entry into cells as measured here by these artificial viruses that carry those mutations and we can also ask um whether or not that innate that enhances the replication in cultures of human airway cells in the laboratory and the answer is these effects are very very subtle and if you do a straightforward infection experiment in the laboratory where you infect different variants and this one here in blue is the b117 you see you don't really see any differences but again if you set up these competition assays where you ask one virus to compete with another in in this preprint here just released this week you can see an enrichment here in the final composition and what this is showing you in these higher bars here is that in each of these competitions the uk variant the b117 variant won the race it replicated more efficiently than the other virus that was inoculated into the culture again showing that there is a real biological difference that's measurable in the laboratory for this variant um and finally again another release preprint just this week suggesting that um interferon pre-treatment of cultures which we know inhibits sars cov2 replication is less efficient at inhibiting that replication uh against the b117 variant than against the other viruses now we don't know the full story of that here but that one possible explanation is that they um that efficient entry that i described to you previously uh through cell surface fusion uh enabled by that extrafuran cleavage site enables the virus to um escape from some very potent interferon inhibitors so hopefully what i've shown you is that those hypothetical reasons for why a particular variant might spread more efficiently can be backed up with laboratory-based biology in these in vitro and in vivo systems and i think the final one to just touch on because i know akiko will we'll talk about it in the end is is the one that everybody wants to know do these variants evade our acquired antibody immune response and how will that impact the chances of vaccines affecting them flash up these before your eyes everyone is familiar with this wonderful data that coveted vaccines work and we think that at least part of the correlative immunity are antibodies which neutralize the interaction between the virus and the ac2 at the receptor binding domain and the good news here for the b117 variant at least is that in the laboratory one doesn't measure any difference in the efficiency of antibodies to neutralize b117 whether or not those are from sera from from convalescent individuals from the first wave or from vaccinees on the other hand of course here in this pre-print you can see that the south african variant the b1351 does have a significant decrease in the ability of antibodies um to inhibit its neutralization so i'm sure that's going to be the subject of later talks i won't touch on that i'll summarize here the the known unknowns here is that we know there are variants we know they have genetic differences what we need to understand now is what do those genetic differences mean in terms of our ability to control the virus going forwards and i'll finish there thank you very much that was great thanks very much wendy that was a brilliant first session well to me at least i learned a lot from that first session that's going to that provides the basis for what we'll go on to and i'm going to hand over to uh cameron abbasi who is going to try the second session george thank you very much um i too learned a lot in that session we appreciate what all the speakers did there thank you um we're now going to look at what do we know about how particular variants influence transmission severity and immunity very important and crucial topics we're going to kick off with mu civic from the university of saint andrews on the b117 variant many things i'll just share my screen so um yeah i'll be covering b117 um and i'm a clinical lecturer in infectious diseases and medical virology so i'm i'm a researcher and clinician at the same time these are my disclosures i'm an independent researcher so i will be basically covering um the how we detected b117 in the uk to provide some epidemiological insights cover what we know about the severity and mortality associated with b117 and i will briefly touch base on biological mechanisms but when they covered it nicely in the last session sorry so in terms of first detection i mean what happened is that early in december there was a surge of cases in kent area and there was an analysis done based on this epidemiological investigations and there was a large phylogenetic cluster that was really genetic genomically similar cases over the week of 10 to 18 november and this was a cluster that was phytogenic genetically very distinct from the rest of the uk data set identified by cog uk so these cases were very concentrated in kant area london london area and which was limited spread into the rest of london or uh the rest of england so this basically um uh raised an interest in you know trying to understand what was happening because the rest of the rest of england was under uh restrictions at the time and the number of cases were coming down almost in other other regions as well so this was actually uh followed up with some of uh detailed investigation it really helped us to um one thing that really helped us was the fact that s drop out in the pcr test was nicely correlated with the number of uh sequenced um cases with one b117 as you can see the black line on the left is very much aligned with uh sequence cases b117 and this basically helped phe and ons you know data sets to be very much aligned with what's happening so this rns data shows that you know there's an increase uh proportion of positive tests with s drop outs in london southeast and south west of england and basically around mid-december as dropout cases were dominant in these regions and obviously this was very much correlated with the number of cases you can see this these are the tier two areas that one went into tier four after we identified these cases um but you know the posit proportion of test positivity and weekly cases were very aligned with what was happening with as dropout cases obviously like um this basically tells us that there's a correlation but at the time we were trying to understand whether there's this is just an association or whether there's some causal relationship and actually modeling helped the team to understand this a little bit better so looking at some modeling data with the logistic growth model we could have a look average relative growth rate of this variant compared to the other variants and this early data showed that this variant was uh associated with estimated increase in r of around 50 to 70 percent and this is a a kind of an uh analysis showing a measure of relative fitness of the variant so this is i think when the rest of the world heard about this variant and further epi insights what what we've learned is that v117 continued to growth growth and this was consistent in across all regions this was in a way like confirmation that this was really more transmissible and further analysis also looking at um data up to december january also shown that this variant had the highest average relative growth and this estimated increase was about 40 and um nick davies and a group in london school of tropical medicine also had a look whether this could be associated with some mobility data or you know people's um like google mobility map and whether it could be related to weather or um you know and they basically found it was not associated with that and it was consistent in all regions and further further work uh was done by phe looking at contact tracing data this was again confirmation confirmatory that secondary attack rates were approximately 30 to 50 higher with b117 in this graph you could in this table you can see the shaded area uh is representing b117 cases and the attack rates are significantly higher which is consistent in all regions i think one of the most consistent debated subject was whether this variant is specifically and particularly more infectious in in younger people children and younger adults and the the graph on the left shows the data from november to december this basically showed that the ratio age share of the variant was higher in um 10 to 19 year olds um and but this was a time when england was in in lockdown but the rest you know but the schools were uh back to full-time teaching at the time and when we include the rest of the data from uh december to january that basically difference has uh disappeared so we concluded that there was no evidence to suggest it favors certain age groups more than others this was mainly to do with the national uh lockdown as i said like in second national lockdown we kept the schools open and there was a rapid decline in all all age groups except school-age children and towards the end of november you can see that children basically secondary school children had the highest uh prevalence um in all age groups this could be related to you know in school transmission around the time when b117 was circulating but also at this at the same time this was around the time when national lockdown was in place so adults were significantly less you know connecting with each other so in a way children had the most mobility data at the time and this was also shown in ons data um at the end of november early december you know um secondary school eight children had the highest infection rates and this is the this is the um data from east of england southeastern london region that shows uh increasing cases around uh you know towards the end of november that was then you know uh increasing cases were seen in all age groups and actually uh we we can now conclude looking at several set of data sets uh bb b117 is more transmissible across all age groups but no more so in children than adults so this increase in secondary attack rates around 30 to 50 is consistently in across all age groups so there's no difference in uh children than adults but that means it's more transmissible in all age groups and further data looking at ons assist data this is randomly tested households regardless of symptoms also had a look at whether there's a different growth rate of b117 by age group and this data set also showed no evidence of difference in growth rates by age group so looking at the severity what we know so far is that this is again the ons data looking at households um there was no difference in terms of positive cases with b117 those reporting symptoms versus no symptoms so at the moment we have we don't have any evidence to suggest cases with b117 are more likely to present with symptoms there was a recent data published in the bmj um this is based on uh pillar two data this is the community testing data and this group did a match cohort analysis which shows increased mortality with b117 but they've shown no hospitalized difference in hospitalization this was also supported by another analysis published in nature this is again based on community testing data shows increase immortality um but as you can see on the table on the right means cfr increases seen in over the age of 17. okay sorry just one minute please okay thank you um so i guess um i mean there's definitely a correlation between rise and mortality and b117 but the question is whether there's a causal link and there's several biases in these data sets um and especially compounding with uh by setting confounding by comorbidities could be important um biases and most of the data looking at these severity is looking at pillar two testing data which includes community testing so that's also important that we need to have a look at hospitalized cases as well but probably this is the least biased data set um based on gp registry data this also shows sharing b117 was associated with higher mortality but as you can see the risk of debt was very low in for people less than 65 without any comorbidities so the main risk is again over the age of 85. so i think i'll move a little bit and in terms of children this is a very small data set but there was no evidence of more severe disease in children and young people admitted to hospital as you can see the critical disease and severe disease was kind of similar between wave one and wave two um i won't go in detail about that yeah sorry moog it's really fascinating um i think i'll just come to the conclusion slide um yeah b117 is approximately 1.5 uh times more transmissible across all age group but there's no that's no more so in children than adults and fitness advantage has been observed in different countries there's evidence to suggest increased mortality hospitalization but this is mainly seen in over the age of 65 with co-morbidities and and i think we need to make sure that this amplification of transmission and severity will be where the burden and gaps in our mitigation measures already exist so such as in nursing home prisons shelters and workplaces and we still need more data to understand the biological mechanisms for transmission advantage thank you okay thank you that was incredibly comprehensive and addressing precisely the question that i think many people are asking uh well i think we need a whole lecture on that at some point we're going to have to move on we'll we'll save muge for questions afterwards we're going to move on to richard lasalles richard is from the university of kwazulu-natal in south africa and richard welcome thank you for joining us you're going to tell us about the b1 351 variant that's correct yes thank you very much and and thanks to the organizers for asking us to to be part of this and i'm just going to give you a very brief snapshot of what we know about the b1351 or what as we call it the 501y v2 and really actually more what we still don't understand about this variant and like many of the others i'm speaking on behalf of many colleagues in south africa in what we call the network for genomic surveillance south africa and what is subsequently kind of expanded to include like wendy mentioned kind of genotyping to phenotyping consortium across the country so i just want to ground you to start with for for those of you who have not been following the south african epidemic closely or who have been reading too many of the articles that tell you uh how africa has been spared by this epidemic unfortunately those uh that couldn't be further from the truth in terms of south africa we've had a very severe epidemic and here's just the epidemic curve showing these two quite distinct waves of the epidemic the first wave in kind of june july august last year and then the second wave around the end of the year december january and uh the detection of the variant in south africa the 501y v2 we det detected it or became aware of it around the end of november so a very similar time to the b117 that muke has described but the first time that this variant was sampled was around the beginning of october and our analysis actually suggests that it probably emerged a month or two before that so essentially as we were coming out of this first wave and what we know about that is that it emerged in a part of the country that was particularly severely affected in the first wave and probably had a substantial proportion of the population infected in the first wave although we have found it difficult to quantify that exactly it may be quite similar to what you hear in the next talk about brazil so here i'm just showing you the distribution of the different virus lineages in all our sampling across south africa over time and what you're seeing here with the different colors and and the white uh over the first period of the epidemic is lots of different lineages that were circulating in the first wave of the epidemic so the white covers a number of mixtures of lineages and there's about 40 or 50 different virus lineages co-circulating in different parts of the country and none of them has a particular advantage over the others but what you see in the yellow is once this variant the 501y v2 emerges in october it very rapidly becomes the dominant lineage that we're picking up in our genomic surveillance and it displaces the other lineages so within a few weeks it's accounting for over 80 90 of all the viruses that were sampling and that was the case in multiple different regions in the country and now it's essentially very close to a hundred percent of the genomes that we sample in south africa and we don't have time to go into it but what we've seen over the past couple of months is that an anya kind of showed it is that this variant has then spread through africa through the land borders as she mentioned so not through uh air traffic and has been kind of at least partly responsible for the kind of severe resurgences that we've seen in many countries in in southern africa at least you've seen uh these kind of structural diagrams of the different variants both ravi and and wendy showed this uh so here on the right we're showing you the structure of the spike protein and here's just the two-dimensional kind of uh map of of of the protein showing where these mutations fall and really the striking thing here when this variant was detected uh as both ravian and wendy mentioned is these mutations are in the exposed areas of the spike protein both in the receptor binding domain so that's the area that hooks on to the the cells hooks onto the receptors on our human cells and then in this n-terminal domain and what we've subsequently learned about this n-terminal domain is is many of the mutations and the deletions here are the site that's particularly targeted by some of the neutralizing antibodies that are not targeting the receptor binding domain so in terms of the transmissibility and the transmission advantage i i show you quite clearly how this variant very rapidly dominated and displaced the other viruses that were circulating around the country so it was clear there was an evolutionary advantage that this variant had over other variants the question really comes back to what wendy articulated is where where was that advantage coming from was it that this was more inherently more transmissible virus or was it that this virus was able to evade parts of the natural immune response and particularly the neutralizing antibody response so in terms of quantifying the increased transmissibility we have some modeling from the london school group in collaboration with the south african modeling group where they use the same kind of mathematical model that they've used to understand the b117 in the in the uk and their early estimates suggested that the 501y v2 was around 50 percent more transmissible than the previously circulating lineages and this seemed to make sense to us because it was consistent with some of the initial estimates around the b117 and it seemed to be consistent because we knew that these two variants had this shared mutation the n501y and and that was plausible that that may have been something that was responsible for the increased transmissibility but this model is also able to say actually this variant could not be any more transmissible than the other lineages but it could be evading some of the previously acquired immunity so essentially as wendy mentioned this could be able to infect some people who had who had acquired immunity from a prior infection and so it was able to spread more efficiently through the population because it could still infect people who would otherwise be protected and and immune at least for some period of time just i i just to show you results from a different modeling exercise here just looking at the genomic data so a simpler model and looking at uh the the genomic data to understand um how this variant displaced the other lineages and here you see quite similar estimates of around 50 to 60 percent increase in transmissibility of the 501 y v2 depending on what your assumption around the the generation time is but again this analysis does not consider uh immune evasion so this just uh supposes that this virus has increased transmissibility so one of the things we've really been struggling with in south africa is answering the question around re-infections and how common re-infections have been with this variant and how that has contributed to this efficient spread of the variant through the population we know now and i think others will talk in more detail we know well from the genotype to phenotype work from the laboratory that there's very good evidence that this variant is is less well neutralized by specific antibodies and by what's called polyclonal serum from people who've had previous infection so we can say in the lab that this virus is not well neutralized the question is does that translate directly to uh to humans and to the population and we still don't know we we are picking up reinfections we have a number of reinfections that we've confirmed with g genome sequencing and we can look at our laboratory data and look at the the number of of reinfections that may be happening but although we've picked up quite a few these are still a very low proportion of all the infections that are happening and one of the challenges we have here is that in south africa even though we're relatively well resourced compared to the less the rest of africa we still detect a very low fraction of infections through our routine testing which is predominantly symptomatic testing and so really to understand this question of reinfection we need good perspective data where people are followed up over a period of time yes i'm really sorry could could you wrap up recently yes one of the sources of data we have is from some of the vaccine trials and not from the people that have been vaccinated but the people in the placebo arm so these are people that are followed up in a trial and haven't received the vaccine and what you can do is compare those who have evidence of prior infection to those who have no evidence of prior infection and these are results from one of the trials the novavax trial that was conducted in south africa and these results showed that the incidence of infection was the same regardless of whether you had evidence of prior infection or not now we need to be very cautious of these results these were reported as saying that this meant that the the prior infection has no protection against reinfection but there's obvious risks of bias in that data and the company have recently just updated those results and and showed that there's actually a different uh finding now and we wait to see the full data from that so i just finished with showing you just on the severity question again that's been very difficult for us to to look at because this second wave was very overwhelming and it's very difficult to understand the true effect of the variant but what i'm showing you here is just that the in-hospital case fatality rate ratio uh in this second wave uh was considerably higher than in the first wave and when even after you adjust for all the other factors affecting mortality there seems to be increased mortality in this second wave what we don't know is how much of that is due to the variant and how much is still due to other factors that can't be corrected for so just to summarize uh i just leave that slide just to finish richard thank you again very comprehensive really fascinating i'm sure there will be questions so we'll be speaking to you later i'm sure about that let's turn to esther sabino from the university of sao paulo and esther is going to tell us about the p1 variant or perhaps esther it's called something else in brazil over to you thank you the epidemic i also want to show a little bit about the epidemic in brazil um so it started late in february last year and there were more than 100 introductions in brazil but probably only two lineages took over and this arcade one and played two in this original paper but now we know it's p1128 and b1133 so after the introduction what we saw was a slight delay in increasing a number of phases but then it ramped up but it stayed longer a long period of many cases until you start to decline in august and then back start increase in early november late october that somehow uh linked to the local elections and a lot of movements in brazil but i think if you look that way you don't really understand brazil it's a big country signs of europe many different cities and epidemic was different in different regions and we were pushing to understand the attack creating insurance of the cdc cities using blood donor samples and as you can see this is just a number of cases that some regions like mammals and fertilizers had a very important epidemic early on and other cities like sao paulo it was really a long longer plateau of cases and some regions stays in the middle and more the south of brazil uh uh decay the proportion of cases were much less so this was the the the paper we had with the data from a mouse and sao paulo we are doing the eight cities but those were ready earlier the data and we could do that because the blood centers in brazil save the samples for at least six months so we could go back when the test arrived in brazil back and test all those samples and we reached the conclusion that in manaus the attack rate was high almost 65 percent in june and reached three quarters by october and this is very so oh uh this is much higher than what happened in europe even in sao paulo the first wave probably end with 30 of attack brain so with this in mind it was difficult to understand what happened late in the epidemic in mammals so after the the big epidemic uh the social er distance was easily in late july and for six months the number of cases stayed low and she starts and slightly increasing in november and then we had this big increasing number of cases by the end of december early january and probably as you see i think this has a lot to do with p2 but then we know now that this is related to p1 so this was the first uh try ways to try to understand what was going on and probably one of the things is that maybe 76 was too much to be working hard on that but even if it's less than that it will not be much less so most of the data converge to something within the interval confidence interval of the first measurement so taking this into consideration the the things we could explain was the immune contract declining the immune protection against us shoe especially if it comes from a new variant and that could evade and also could be more transmissible that would be the explanation and that what happened so we could attack early on in january when we first get got some samples from from our now that it was a new friend with 70 mutations the commutations that have already been described we also detect the speech that was present in brazil earlier on in the last uh the second semester of 2020 and that was arriving in in in manaus by by november so the the peak of the epidemic coincides with the increase in p2 and p1 uh proportion uh in males so now we know that this this variant starts in november it's very different the way it starts than p2 we can see that p2 starts traveling may or june and different than than p1 that's really started in november probably the origin is is from and now so we can see some sequence that is near uh p1 and we and this is this is also the same what the group from philip navecca december there was no case attack before before december so in this part we show that it really displays the the other uh variants present there which suggests a higher increase of of trans transmission and the viral load was higher although when you control for time synthesis uh symptoms onset you don't see the difference and this was also happened in the data set from the field group that also i are studying d1 now so to understand what was going on how come you have such a high uh higher increase of a new virus in this population previously highly exposed probably think three things has to happen about the same time you cannot explain by one thing only probably and this is a mathematical model developed by the group for the imperial college it's a collaboration and so the outcome is pride that the p1 is probably more transmissible but the intervals are large yet for the type of data we have we were able to obtain children probably can evade people previous infected so this is an important number but it's just mathematical models saying how what's the proportion of the people who could get infected if they were exposed it's not that we believe 25 to 61 percent got infected in my mouth again but that's just the thinking and also there probably there was an increase in immortality which is hard to know if it's directly to the virus or to other any other um uh to all the problems that we face when we were trying to respond with epidemic in mammals so it's still in an early phase of data collection and we cannot extrapolate to other parks or to other regions but what happened in brazil after that was that um as you can see this is coming from july and this data is from field cruise and it's just showing the proportion of uh uh intensive care units available since july and as you can see it moves over time so the epidemic was not to synchronize until now when after february that we start seeing or everywhere in brazil having a large epidemic about the same time we don't have yet a well collection of data to say what's the proportion of p1 in each one of the signs for the regions i have tested which was in the northeast sao paulo wheel general and most of the time we are finding around 70 percent of the t1 so probably this epidemic is now synchronized all over the country with the p1 bearing so just to summarize and i think it's important to to to realize it's kind of hard to do research uh especially on an epidemic and under under project but we are glad that we have all this collaboration and so what we know it's so t1 has this key important mutations and starting to have really nice detail thank you very much thank you very much that was really marvelous um impressive to see how much work you're doing in brazil but also impressive to see all the work taking place globally but still many questions to answer i'm sure you will have questions for you at the end the session that phil will chair but but immediately we're going to go to alison uh for the next session which is on the future of the pandemic alison thanks very much cameron i'm completely in awe of all the speakers i've learned so much as usual um tremendous work um so we're going to go straight now to jeff barrett who's going to talk about the future for surveillance and genomics um thank you jeff without morrow if i'm wondering whether you're muted at the moment thank you very much uh sorry i'm going to say i'm going to tell you about the grist that goes into the mill of looking for these variants uh in particular in the uk and um since around last summer the sanger institute where i work which is part of the cog uk consortium which in the uk is the network that does the genome sequencing of the virus which anya alluded to in his sequence over 350 000 uh genomes so far so the bit that we do at sanger is we work closely with the so-called lighthouse labs which do um the kind of general population testing uh in the uk and we have these vans these these chilled vans that drive down from all over the country every day they unload these big green boxes um into sort of back of a lorry sized walk-in freezers where we store hundreds of these boxes which in turn contain millions of the waste material from pcr tests after they've been done and inside those boxes are these little plastic plates that are you know the kind of bog standard of molecular biology experiments and they have 96 little wells and in each one is a sample and the idea is to pull those out after the pcr test has been done uh get the material from the virus and read its genome and we have a high throughput uh setup that is a combination of of people and robots and we've repurposed several of the labs at the sanger institute which is a big genomics facility um to basically generate this data at scale and in fact we have scaled up to the point where now we're sequencing between 10 and 20 000 virus samples every week and rapidly putting those data both into the hands of academics and the public health agencies in the uk and indeed global databases and one thing that we focused on over that time is is try to reduce the time between when the swab is up someone's nose and when the sequence is in the hands of public health agencies and if you look back since when we sort of started this in earnest you can see it was about two weeks but actually as we've moved forward to today it is just under one week a couple of kind of amusing things in this this giant uh spike is the great pipette tip shortage of 2020 which um people may have experienced in different circumstances but it meant it was much harder for us to do sequencing and this was a combination probably of supply chains related to the pandemic and to brexit and the other thing i'll note is there's a very gentle slope of this as we slowly work to do this faster and faster you can see in recent weeks there's been a steeper slope and this is in part because we've we've actually developed an even closer relationship with the department of health and social care and test and trace in the uk to try to really streamline this process to get the data out as quickly as possible and the reason for that is exactly what this whole this whole seminar has been about which is the variance of concern and tracking them fast becomes important so what do we do with that um it's been kind of covered a few times but let me just show one of the ways that we are sharing these data now just recently from this online interactive website of the of the sequences as they come off and you can see in particular the prevalence of the b117 lineage that we've been talking about and you can see that it was growing in the southeast in december and if i sort of jump forward it spread throughout the country to the point where now uh 99.9 of new infections in the uk are uh b117 you can also look at the famous r number um by lineage which is basically if it's bigger than one uh the epidemic is growing if it's less than what it's shrinking and just briefly in in late november when b117 was getting started you can see it was pinky colored even though this was locked down light in the uk and it was unfortunately the bad news that meant um it was going to as we jump forward into december sort of take off and spread uh throughout the country by contrast if you looked at other uh lineages in the same time they were growing but sorry red is growing fast and blue is shrinking in these maps at a slower pace and then in fact the good news is if you jump forward to this year you can see the whole country turns blue other lineages are going extinct even b117 which was growing in the lockdown light has been consistently shrinking during the current strict lockdown in the uk and if you look at this chart on this side um these are the proportions of different things so the uh the red was a common variant of the genome that was in the uk and elsewhere in europe last fall blue is everything else and green you can see is just the explosion of b117 which has grown very fast and is now as i said essentially all of the infections in the uk um what does that mean going forward well um what we hope is that other countries can learn from the fact that we discover these things and and i'd like to draw the example of denmark who or another country has done an amazing job of sequencing a huge fraction of all of their positive tests so they really have a very clear picture of what's going on and this is each of these bars is the number of uh sequences they did of positive tests in denmark week by week and the red fractions are the fraction that's b117 and you can see that it they saw it early and to their credit introduced early in 2021 around here when it was still only a couple of percent stricter uh restrictions and what you can see is the gray everything else just like in the uk is basically going extinct they managed to squash the complete pre-b117 epidemic but the red actually is growing sort of week on week not only proportionally but in absolute terms and denmark in recent weeks has been in a kind of you know slowly growing or steady phase of their epidemic and this really happens this b117 just grows so much faster that um even if you can restrict other lineages this one will just sort of blow up um so that's what's going to happen sort of country by country we can see it in europe and elsewhere in the world the b117 is spreading so fast what about in the uk this is now breaking down um all of the sequences we've seen since the beginning of 2021 in our kind of random surveillance um that i'm that i'm alluding to [Music] and you can see that b117 we're seeing tons and tons thousands of sequences every week a couple of people have mentioned the e484k mutation which is the other one that's very important and likely plays a role in evading immunity um it's sometimes new mutations happen and sometimes it appears on the b117 but you can see these are just relatively low counts so there's an interesting puzzle here as to this is a 484k is an essential part of b1351 and p1 but not 117. it's not clear why that is if you look at every other lineage that has the 484e position you can see they're just as i mentioned previously going down to almost zero but what is slightly concerning is that other lineages with the e484k so this includes b1351 p1 uh a few others b1525 is one that hasn't been much in the news but is starting to grow in the uk they are kind of holding their own so they're very rare uh in terms of the overall proportion in the epidemic in the uk right now is almost all b117 but they're not being kind of squashed into non-existence and so that's something that i think needs to be watched very carefully and if i just wrap up you know trying to pull together some of the numbers i've just shown you in some of the comments that have been made um it's very clear that b117 can just sweep through a country extremely quickly and so um my advice to any country where that hasn't happened yet is act as quickly as you can because as we saw in the uk and are starting to see probably uh in continental europe it can create huge problems setting aside vaccination um if you are already in a b117 swept area that's kind of your vanilla wild type virus now then you really need to be watching for the variants that are much more likely to have to be less well neutralized by vaccines so there's as previous speakers have mentioned there's pretty good evidence that b117 will be protected against by the existing vaccines it's not as cert we don't know for certain but there's certain certainly concern that for example b1351 or indeed something we don't know that well yet will not be as well uh stopped by vaccines so then this is the state the uk is in which is you really want to watch as carefully as possible for those and really either way it's just super critical that in as many places as possible in the world we continue to sequence as many cases as possible in 2021 and beyond because we need to know what's happening with these things in order to know whether as we just get ahead of the virus in the foot race with vaccines we can kind of stay ahead of it by adjusting the vaccines or by changing how we we introduce lockdown measures and those kinds of things okay thanks very much and i'll hand over to the next speaker [Music] very much indeed jeff that was superb and i'm sorry you didn't have more time uh the new variants for a key reason as you know for um border closures and also lockdowns this uh christmas so it's really great to have alexander in here talking about some of the geopolitics of border closures in association with the variants so thanks um alexandra we'll go straight to you now thank you so much thank you let me pop the slides on wonderful um so i'll be speaking a little bit about the uh the implications of these variants for uh international law and travel and i think it's an important starting point to understand why our international law has uh developed the way it has particularly given the fact that travel restrictions were such a hot issue early on in the pandemic so the law governing the international spread of disease uh is over 170 years of norms that have developed and they really developed when during the industrial revolution uh you had suddenly a movement of people through steamships and steam trains crossing borders and at the very beginnings of germ theory a lack of understanding about what control measures would be legitimate and as a result significant human rights violations followed but for the countries at the time particular concerns about the economic impacts of closing borders and destroying goods and subjecting individuals to quarantine and so much of our current international law which i'll look at specifically at the moment despite being adopted post sars in 2002 2003 is really built on these sort of historical norms and that is the international health regulations adopted following sars in 2005 that seek to prevent protect against control and provide a public health response to the international spread of disease but also in a way that avoids unnecessary interference with international traffic and trade and this unnecessary interference has really become a touch point during this pandemic because it doesn't necessarily mean no interference with international traffic and trade but we haven't had sufficient guidance on this um partly because of a lack of historical evidence to to really work out how we should uh have nuance in this net this idea of necessary in light of covert 19 and in light of the emergence of of new variants that are coming out into a scene where travel has been significantly disrupted so prior to covert the the rationales of of who and these laws advising against international travel restrictions really came out of the ideas that you know travel restrictions impede the flow of medical goods and services they are typically insufficiently tailored to actually prevent transmission given the scale of global travel a big concern and we saw this early in the pandemic was that imposing travel restrictions gives a false sense of security uh and you know and governments may not actually uh implement the necessary local public health measures uh to prevent the spread of a novel disease and as countries impose these measures on on places where new diseases may emerge it does mean that future diseases may not be reported rapidly because countries are worried that they're going to have these measures imposed on them and the economic impacts now in the middle of this pandemic we are in a different place in terms of the impact of travel restrictions given how disrupted international travelers but in many respects these similar concerns still play out with the emergence of novel variants particularly when we see governments potentially responding and implementing restrictions based on the emergence of new variants for which we may not have data or for which we start to have data that they have increased transmissibility or impact the efficacy of existing measures covert 19 has also helped us build somewhat of an evidence base the first really clear lesson and this is particularly relevant for variants is the detection of a novel variant in another country does not necessarily mean that variants are not already transmitting locally in the country and so an effort to impose a travel restriction on originating country um to to you know protect the second country fails to recognize that there may be necessary local measures that are required what we have noticed through this pandemic because of the scale of disruption is restrictions may slow but not necessarily stop the spread of new variants there are some partial exceptions to this and that is countries like australia new zealand taiwan that have implemented very strict domestic border control measures that are not simply travel restrictions but essentially filter all incoming travelers citizens or not through uh through quarantined processes in particular australian new zealand we've seen even in those countries that have approached elimination the quarantine is actually that weak point particularly when we've been seeing these novel variants being detected and potentially spilling out into local community without proper public health measures in those quarantines like we saw in the recent melbourne uh and and queensland outbreaks um and you know fundamentally uh travel restrictions regardless of this this evidence that they may now slow the spread of variance are not replacements to those local surveillance and public health measures this is a piece that we publish in cell that goes into a lot more detail on this but i think a couple of critical points to hit here variants may shift the justification of what is the least restrictive measure necessary for public health it may prompt governments to act in ways that they had not yet because there may be evidence of that increased transmissibility but the reality is is in many places governments have not appropriately acted to wild type size kobe 2. so new variants may prompt overdue action they may also result in overly restrictive measures that sort of go beyond as a result the challenge with this is if governments do that is it may undermine the legitimacy of uh of the measures that should should have been in place prior to the detection of variants where that is a risk is if we then have later data that demonstrates that there's no increased risk of transmission or immune escape that then could justify rolling back measures that really should have been in place in the first place anyway so there's a balance particularly if we see governments using it to justify even more restrictive than necessary measures the challenge with this is the imposition of restrictions uh travel restrictions will still have future implications for variant reporting the concern that the imposition of travel restrictions will prevent countries from rapidly reporting new diseases equally applies to the rapid reporting of new variants and whilst countries have binding legal obligations under the international health regulations to notify who of potential public health emergencies of international concern and to share information such as the emergence of new variants we have seen through covert 19 in the early days and then you know subsequently with variance there has been a an undermining of this rapid notification um partly also because of concerns about the imposition of economic uh restrictions like travel restrictions i just want to you know it would be remiss to not mention how variants relate to the potential use of vaccination passports in this broader discussion if we're looking at reopening borders and reopening travel i think we've seen businesses and we've seen countries seeking to and regional blocks cp seeking to adopt their own vaccination passports but the reality is this is governed by international law there are two ways in which you could have have a vaccine passport recommended and one would be through the fact that we are in a public health emergency of international concern which gives uh who jurisdiction to recommend at this stage who has not recommended the use of vaccination passports and that's because it's quite complex there's a lot of detail to be uh to set out and that's why we have only one disease with that level of defec uh detail and that's the uh yellow fever cat jean international certificate of vaccination so there is scope for countries and an important part for countries to play in actually achieving consensus on this you know if we have this fragmented approach where businesses or airlines or governments are implementing their own vaccine passport scheme we start to really run into issues with inequitable global vaccine distribution which means some parts of the world will be able to travel and others will not many of these decisions are coming along geopolitical lines china has recently announced it will only accept foreigners arriving that have had had a chinese vaccine and we do risk having situations where it's not clear which vaccines will and will not be accepted under a passport scheme that could entrench the potential emergence of future variants and prolong the pandemic we also have clear data gaps in in what variants vaccines will be effective against and you know whether vaccines are fully effective at reducing the transmission and so in in many respects the emergence of novel variants highlights the urgency that already existed for clarity and processes for global consensus on any passport scheme that the countries may choose to adopt you know just to summarize and and end our existing international law is insufficient and vaccines are exacerbating these exist so variants are exacerbating these existing issues we need to decouple the obligation to notify from travel restrictions we need to actually develop guidance that is up to date on what we have learned from covert 19 including from the emergence of new variants um there are going to be new challenges that variants might pose if we introduce things like vaccine passports or travel restrictions that are based on on what vaccines are accepted or not accepted so that we can develop certainty and ensure that we get rapid notification of future pandemic threats how this might occur would be fundamentally through reform of the existing international or the ihr or as some of you may be aware there is uh proposals for a new pandemic treaty these are two critical ways in which we can start to address these gaps the world health assembly is in may and really the emergence of variance just underpins how how urgent it is that we start to address these gaps not just for public health but for to avoid any potential geopolitical risks that arise from this going forward thank you thanks very much alexandra and rather chilling as well so um it's really very uh that we're going to finish now with akiko ibusaki from yale who's going to talk to us about how variants affect vaccine efficacy and the implications for policy thanks akiko thank you alison um so i'm delighted to be on this panel to discuss what variants mean for vaccine efficacy and what we can think about based on our current data so um just to recap vaccines and naturally a natural infection with sars cov2 elicits immune responses and variety of immune responses the correlative protection for um co-stars cov2 is still not clearly understood but it will involve things like the neutralizing antibodies non-neutralizing antibodies that can bind to the fc receptor of various phagocytes and nk cells to kill the infected cells as well as t cell immunity that can directly lyse and eliminate the viral factories and there are some subtle differences between how vaccines and natural infection induces these immune responses many vaccines actually surpass the ability of the natural infection to induce neutralizing antibody titers not all vaccines are you know good at inducing diesel immunity but some are capable of directing the th1 and ctl type of t cell immunity natural infection will also be able to induce local immunity within the respiratory mucosa that may be also quite protective against uh reinfections what is not known however is the relative importance of these effector arms and how much redundancy there is in conferring protection against both the original the non-variant of concern as well as the variant of concern and whether there is enough of an escape from antibody detection as well as t cell recognition of target cells so unknown so what we currently know about the variance of concern which you've already heard extensively discussed today just to recap the there are some mutations like the d614g that um is really dominant worldwide in its um enhancing infectivity and 501y is present in these three strains of three variants of concerns and leading to increased transmissibility um e484k and k417 mutations have been seen in variants uh that are indicated here uh also helping to evade antibody recognition and another variation the mutation in the l542 present in these sort of california variants may also help immune escape however very little is known regarding the impact of the non-spike mutations that are accompanying these variants which also requires further investigation so do variants of concern evade naturally acquired immunity the data are emerging we don't have quite the full picture yet but there are some data that um for instance discussed by esther already of the p1 variant uh being able to uh infect people who have recovered from the original version of the source code b2 in the first wave and so you know these kinds of data are now suggesting that the p1 variant may be capable of reinfection the b1351 variant of concern as you heard from mitchell's talk um the the phase 2 a b south african trial for the novovax vaccine the placebo arm had basically similar you know reinfection cases in the two months follow-up however at the 90-day follow-up there appears to be less rate of infection in the sero-positive compared to the serial negative suggesting that pre-existing anybody may confer some level of protection against the south african variant again the numbers are very small at this point we don't really know exactly how this is going to turn out uh and the note of caution that uh was raised by uh muge was that there are some reports showing that the service cov2 igg elisa depending on what the assays are used may be picking up some false positives especially in populations that were studied in ghana and nigeria indicating that we may have to really look into the specificity of the sierra surveys that are done in different uh parts of the world just to make sure what these reinfection rates are calculated correctly so about the variance of concerns uh evading vaccine-induced immunity um so the clinical phase three trials of various vaccines are now giving us some input insight into this for instance this is the johnson johnson single shot antivirus vector vaccine and here is a comparison of the efficacy vaccine efficacy against moderate and severe infection 28 days after the shot and in the u.s where 94 96 were the d614 614 g variant uh the dominant ones that's circulating uh the efficacy was 72 and this uh dropped to 66 in latin american um uh survey uh and in south africa dropped to 57 where 95 percent of the circulating virus was the b1351 variant however it's also important to note that severe disease from all these variants were avoided after 49 days of the first shot indicating that uh it's you know even if there is this moderate and severe moderate infection that's occurring um most of them actually all of them in this cohort was protected from severe or lethal covet another insight comes from the novovax two-shot vaccine trial and here it's basically comparing the symptomatic infection seven days after the second dose in the uk where over half of them were the b117 variant the efficacy was 89 whereas in the south african um survey it was dropped to 60 or 49 in hiv positive and negative population um and so again because the the dominant strain at that time variant was the b1351 in south africa this suggests that there may be evasion from vaccine-induced immunity in this variant and then again the severe infection was a hundred percent uh reduced across all of these uh variants and finally the the more recent data from the character study [Music] in south africa really indicating that so if you were to look at the original efficacy of this particular vaccine in the non-variant of concern was around 66.7 percent and then against the uk variant was um sorry b117 variant was the um 74.6 and yet within the south south african study the these efficacy really dropped down as you can see here um the mild to moderate illness within over 14 days after the second injection as well as really focusing on the b1351 variant dropped down significantly suggesting that you know not all vaccines can provide protection against the variant and there's a degree different degree of efficacy of protection based on which vaccines are used in which regions of the world so there are lots of unknowns um for instance the vaccine of variants of concern with certain mutations such as the e484k and l452r and some of the deletion mutations appear to reduce vaccine efficacy and pre-existing natural immunity and so that this begs the question about the t cell responses whether they are sufficient to protect against these vocs voc can also cause reinfection but how much better can they do this uh we still don't know the answer eoc might also increase host range and resistance during a theorem which you've already heard from other talks that's also a major concern so i'm going to end by imposing some some of what we can do to prevent further spread of the vocs i'm obviously vaccinating as soon as possible in many parts of the world but particularly those parts of the world in which we already know the vocs are dominant is very important um survey we need to survey more uh viral genome surveillance across the globe we need to adapt the vaccine to match the vocs for the target regions we need to treat aggressively immunocompromised patients with monoclonal antibody cocktails to eliminate development of new variants and we need to also develop transmission blocking vaccines and finally if we can develop a universal carnovice vaccine to prevent all of these variants from um becoming a problem would be ideal in the future so i end here by thanking these four brilliant students who put together the data that i was able to share today these are medical students that i taught this year and they became so interested in immunology that they formed a group called spike support so i'm really really grateful to them thank you that's greater kiko and thank you and thank the students as well i'm going to hand straight over to phil because he's going to have questions from nikki nikki's going to ask questions from the panelists so thanks again thanks uh alison wasn't that fabulous that four medical students collected all that data so some some good has come out of this pandemic we've inspired the next generation of immunologists and much else besides so nikki i gather you've got a couple of questions for each of our sections do you want to kick us off yes i do indeed so the for the first panel our first question is is it merely a coincidence that those countries which have been heavily involved in vaccine trials such as the uk brazil and south africa are also the same places where the most virulent mutations we currently face have arisen or could there be a connection oh that's a tricky question who wants to take that is it because is the vaccine mutating most in countries is the virus mutating most in countries so it's doing most vaccination is that of spur variation who would like to answer that i think i could probably have a go you want then well i think it's it it's highly unlikely uh certainly in the uk the um if you look at the the the the the branch lengths of these uh of the b117 it's really quite long which means that it's been hanging around somewhere and it's unlikely that simply vaccine pressure has has caused that i think if you look at the timing of when the vaccine trials were initiated uh in in most of these countries i think i think there are there are more pertinent there are more relevant explanations for for for the emergence of new variants that's not to say that vaccines may not end up applying a similar or or encourage new variants in the future i think that that is certainly a possibility but i think that chronic infection is an entity we've observed numbers of you know in the literature there are now more than 10 cases described of accumulation of mutations over time that mirror perfectly you know what we're seeing in the new variants um so i think that given we we could we have this concept of chronic infection we know that's highly these are highly transmissible variants you don't need to invoke a vaccine sort of theory very quick before i bring in wendy quick supplementary question claire gerard gp said we shouldn't name variance by place um because it leads to all sorts of bigotry and discrimination but can come will we ever absolutely know for certain where these variants originated from i mean can you be fairly certain the kent variant came from kent even though we're not going to call it that or not well there's for the kent variant there is a there's a there's a related sequence in jizzaid uh four out of the eight spike mutations so i think that there there is that's one way of detecting whether it's likely that infection arose in that geographical area so the closest one of the closest viruses to the b117 is another virus isolated in the same region so i think that's what's making people um uh confident this is that the virus emerged there but i'm not on the group that's been investigating and there may be others with more knowledge okay thanks uh wendy what did you want to add just to agree with ravi's basic answer i i don't think that vaccines have driven the emergence of these they haven't been widely used in comparison to the number of people who've been infected for example and i think the talks that we heard from south africa uh from richard and and from esther showed that there was some immune pressure which was certainly independent of vaccines and probably much more numerically relevant in terms of of emergence of that virus but maybe richard wants to come in there richard do you want to comment on that well no i just wanted to add the other the other point to ravi and wendy that of course vaccine trialists and vaccine companies want to test their vaccines in places where there's quite high levels of transmission because then they get an answer to their question of whether the vaccine works or not much more quickly than if they tested somewhere where there's no virus being transmitted so that's that comes back to the the the point ravi and wendy have made that these were places where there was a lot of infection spreading in the population and so it's it's the classical thing of of correlation and not causation so so we need to be very careful about thinking that vaccinations have have uh contributed to the emergence of the variants thank you that was very clear nikki what's our next question yeah i think our next question is for anya because i think she mentioned that there are um some biases in the genomic surveillance and someone in the audience was wondering if she could explain what those biases are all right are you still with us i am i'm here um yeah no problem so i i kind of touched on it in the talk a little bit but that the biases are a lot of very different kinds of biases right so you're talking about how much people sequence in each country and as you know we've heard about today cog uk sequences a huge amount and denmark sequence is a huge amount and other countries don't necessarily sequence quite at the same level particularly in the community so another bias is whether people are specifically targeting cases of travel or cases um in the community as well and even within cog uk a portion of sequencing uh goes towards community surveillance and a portion of that goes towards outbreak investigation so that bias is the representation of the data that we see up on gsaid and and another bias that we might actually see is what data goes on gsa so so certain um labs may be uploading a subset of their data they might be uploading um variants of concern because they're more interesting or they want to share the data or something like that so so there's all sorts of biases and when we're looking at counts and sequence counts across different countries we need to keep those things in mind that was good it's very clear andy thank you very much nikki what's our next question okay so if we move on to panel two and then if we have time we can come back to panel one um but for b117 and i guess also for the other variants as well when you looked at disease severity did you look at disease longevity and see if it was more likely to stay with patients and cause long covert disease longevity i could try to answer at the moment there's some data were following up patients were discharged in the uk but we don't have currently enough follow-up data for those patients who were infected with b117 but over the next couple of months we will have some data about this okay well hopefully we might even in three weeks because our next seminar in three weeks is on long covert so we might have the answer by then so thank you would anyone else like to chip in on that or should we move to the next question just very quickly i think we should make a distinction between long covered and chronic shedding which is i think was that what you were suggesting so we would i think the duration of illness and duration of shedding may be different in b117 there are some there was a report from a u.s basketball survey that suggested that the shedding period was actually longer in b117 infected individuals um but that is not necessarily related to long purpose we need to make that absolutely clear to everybody in the room or the world thank you anyone else want to add to that okay nikki next question of course so can we can anyone on the panel explain the significance of the recently added category by the cdc of variants of high consequence oh who wants that variance of high consequence no takers oh maybe yeah wendy go on wendy there's a lot of debate about what we should call all these different uh variants and a sort of traffic-like system really to try and judge how concerned we we should be and of course when you start calling a virus a variant of concern that does trigger as as we heard a lot of a lot of policy changes uh including border controls perhaps and other things which have huge impact and so at the same time we can't ignore a variant when we see it appear particularly if it has some hallmarks in its sequence which might cause us some worry and predict that it might invade vaccine or behave in a different way so the idea of calling things here in the uk we tend to call them variants under investigation or vui's um but there are a number of different names now around the world for variants that haven't quite reached the pinnacle if you like of a variant of concern but are still things that we need to watch very very carefully and there are various levels of public health response that are appropriate for those and and these things can move up and down on a sort of almost daily basis as you risk assess them and gather more information it's interesting i find risk communication fascinating it was when uh as i will call it the kent variant first kicked off we announced it as a mutation as if there was a mutant on the loose in london and everybody panicked and all these people getting the last train out of saigon and the word mutant was seen as something incredibly frightening and they scaled it down very quickly to variant and now a variant under investigation a variant of high consequence a variant of concern there it's a very nuanced explanation uh richard did you want to come in well i just wanted to say and you see this problem with different terminology being used we saw it yesterday with this report of this double mutant from india which which which then got a lot of alarm and and then we're all scrabbling to try and understand what what it is they're talking about and it it they're just talking about a variant it seems with with two mutations in the in the spike again and the rbd so it's we've got to get away from this kind of way of people wanting a new term so that they're capturing a bit of attention and things and we've got to try and get more consistency in in how we're talking about these things that's very interesting alexandria you wanted to come in yeah so jumping off what has been said i think the one of the challenges for the public in dealing with this and even for scientists who may not be following the variance discussion quite closely is this terminology can communicate severity um and and may not also inadvertently and it may not also match up with what government's actually doing and how governments are responding and the the long-term concern with that is we still have a baseline of control measures that are required and responses that are required and any sort of relief in similar ways we see going up and down the an epi curve um can be sort of filtered through that and so there is a there's a real risk that we that this terminology can affect public policy in a non-linear way um that can have implications in the future can i ask you a tricky question of course is there good evidence that international law has been broken in this pandemic yes um [Music] so well if you're discussing if you're referring specifically to the reporting early on they're you know what is hard is there is always a fog of outbreak and our laws are structured in a way that we have obligations to report within 24 hours of detection the big question is is when did the national government detect so there is um you know i say yes certainly but i'm not just referring to the initial notification i'm talking about the imposition of travel restrictions beyond what who recommended and a much broader set of of potential violations um it is really hard to work out at what point is an obligation actually violated particularly when it comes to just the normal um delays i mean if we look at the emergence of the the first variant of concern you know that was quite a long period of time it was much longer in terms of reporting obligations than if we look at say the the reporting of the emergence of the outbreak at the beginning so i think pinning the point at which those obligations actually arose is difficult but if we look globally not just at the initial report yeah there are multiple instances of international law not being complied with and is it like to end up in the courts do you think there will be legal cases where you didn't alert us to the variant quick enough good question there is a really live question uh and those of us who are international lawyers are in the process of debating this um whether there could be a claim under international court of justice jurisdiction for it we haven't got it we haven't had a test case there is definitely pathways for disputes to arise whether states would even consent to that which would require their consent is is quite unlikely so we have a lot of paths to go in what a pandemic treaty or international reform could do for improving compliance and accountability thank you very much nikki next question please yes so i know that some of the speakers covered a bit about transmission in children and different age groups but are the new variants more likely to cause symptomatic infection in children ah who would like to take that i mean i can i think i covered it in my presentation but we have very a small data set coming from a single hospital looking at hospitalizations in children comparing what has been happening in way one to wave two and in terms of asymptomatic infection and symptomatic infection and different severity of disease there's they haven't found any difference and there's another study looking at households so there was no difference in symptoms reported but these are early data things may change so we need to keep looking into it do you think uh it's ethically reasonable to vaccinate children primarily to protect adults um so i guess at the moment the priority is to vaccinate high-risk groups um but i think as long as we have clinical trial data that the vaccines are safe and effective in children yes that may come at some point but i think that also relates to vaccine equity internationally so how can we distribute the vaccines available for all countries that's a very good point too uh anyone else want to come in on that are we back to nikki nikki let's have your next question yeah of course so we're sort of moving into the final panel now um someone in the audience has asked by priming our immune system so specifically and strongly did the new vaccines change the frontline killer cell response in a way that might reduce the immune response to other challenges and if so how permanent is this change oh gosh that's a tricky question does anyone know wanna hazard a guess at that the frontline killer cell response and will it affect our ability to fight other infections yeah maybe i can try yeah so i was sort of alluding to the t-cell immunity that's induced by natural infection versus vaccine and typically when you get exposed and infected with a virus in the mucosa you generate what's called the tissue resident memory cells and they reside within this original site of encounter for a very long time and provide protection that's been shown with the non-human primate model as well that the cda t-cells in the nose is very important for protection um within that tissue the vaccine does not necessarily induce these types of responses because it's a systemic immune response so as far as i'm concerned the vaccine-induced immunity or the naturally acquired immunity there's no really competition for a new t-cells to emerge against other pathogens even if you were to like really fully maximize a immune response against uh sarco v2 you can still respond to flu rhinovirus adenovirus or whatever so i don't think it's going to interfere with future pathogens if that's what the question was okay well if i rephrase the question if you could choose between natural virus infection which we knew would be completely harmless obviously you can't but if it was versus vaccination would you choose natural infection or vaccination to give you the best protection so if the natural infection was not at all harmful is that the question yeah yeah if i could say i can guarantee you'll get asymptomatic and it won't harm you won't give you long curvy what would you choose i would probably to choose the natural infection but but you know that that's not the case right so um unless we can make a vaccine that's completely safe to give in intranasally and that would be something that i was talking about like we need to develop new vaccines that would prevent infection and transmission in the future and that could come in the form of a mucosal vaccine okay good uh nikki do we have any other questions you want to pick up on yeah so there's one question about um which of the current vaccines are more or less effective against any of the new variants and then to follow on from that then how valid is the original clinical trial results for the current vaccines if they were conducted on a variant that has now been superseded by new variants okay so which vaccine is best for variants does anyone know that yeah i was just going to say like you know essentially what i presented is what we know so far and um you know what i was trying to uh communicate that not all vaccines are equally effective against all variants but they all seem to prevent severe disease and certainly lethal diseases and the johnson and johnson for example went from 72 percent down to 57 with the b1 351 variant for example but still there's that 57 protection against moderate and severe diseases um and similarly the novovax uh did have this comparison between the uk uh trial which was 89 down to 60 in the hiv negative south african trial which was dominated by the b1351 variant so um even though there is that drop there's still protection against infection uh sorry symptomatic infection as well as complete protection against the variants from severe to lethal covet that may or may not be true for the uh the oxford the the chadox vaccine as i showed that there is a significant reduction in the uh moderate to mild illness provided by the that vaccine and i'm not sure if the severity was um measured uh like severe infection was measured but you know my guess is that those data will be coming out soon in slightly more consumerist america are you allowed to choose what vaccine you have because we don't in the uk you get whatever we happen to have do you yeah we have a menu you just check off no just kidding we don't get to choose either all right okay just sit right so it's a moot point really isn't it you can't choose what vaccine you're getting anyway so did you want to comment on that uh ravindra yes i just wanted to add that um part of the question was alluding to the fact that vaccine trials were done in places where the variants weren't circulating but you can't predict what's going to be circulating when you plan a vaccine trial these things are hugely complex and uh and so i don't think that's really a sort of feasible argument are you are you confident we'll be able to tweak the vaccines because that's what they say in the press conferences all the time yes there will be variants that might include the vaccines but we can tweak yes i mean we can because you know these these these vaccines have been synthesized with building blocks and you just have to change the building blocks the key thing will be to figure out whether which building blocks you're going to change um which mutations you're going to introduce because that's not such an easy um uh thing to to plan um and also uh what it does to the stability of the of of that vaccine and if there are any unintended consequences so so that's the really the tough part is the is the safety and validation and what and how much fast tracking can be done yeah now i can see that uh nikki are long shedders the chief source of the new variants so is it the long shedders that are giving us the new variants probably yeah ravi do you want to say something more i mean i'm slightly brighter because of course you know i mean but it'd be good to have what other people think but um there are many there are many long shedders uh we see this in the clinic all the time we've got people persistently positive um you know even people who are who are not immune suppressed so just just you know healthy people uh we are not necessarily sequencing all of them we don't necessarily know what changes are happening but they're chronic shed shedding chronic shedding over let's say a few weeks is very is actually quite common so i don't think it's all shedders but there will be specific subgroups and especially when your immune system isn't is is lacking in both the b cell and the t cell arms and you give a therapy then that's the area where where the virus basically has big opportunity to colonize different tissues um and replicate to high levels and it makes it harder to control that infection and that's the opportunity for the virus to start mutating but there will be lots of different um permutations of that as we've seen from all the cases i presented one was asymptomatic one had antiphospholipid syndrome with lots of steroid treatment mycha had um lymphoma and was treated with rituximab which is an antibody or b cell depleting agent so it's really we don't understand what the com the common themes are unfortunately yet when you come across a long uh shedder though do do you ask them to isolate for longer i mean they may already be shielding or something but if as if they didn't have enough on their plate are they then told yeah it's very anxiety provoking for everyone involved because you know we we know that they're positive and they're at home it you know if they can shield then that's good but then we have a problems testing them because they then have we don't want them coming back in the hospital so then we have to send kits out for them to be tested so it's a real problem i think we need to this is one of the big research priorities that we need to get on top of quickly uh and and globally as an and work together to understand what this all means because i don't think that aspect of covert 19 is going going away uh jeff you wanted to come in and then move afterwards jeff first yeah just on that topic one thing that several of the speakers alluded to which is clear is that these variants of concerns ten gov concerns tend to arise with several 10 20 mutations uh compared to their sort of relatives in the virus family tree and we that that's unusual especially in places where we sequence a lot like the uk we usually see mutations kind of appear one by one and so that does suggest i think that there is some other process where the virus evolves that leads to the variance of concern that is unlikely to be just the general one person transmitting to the next person kind of thing and that's why i said probably at the beginning because i think for all the reasons ravi just said um long-term shutters are a likely source of that kind of different evolutionary process that can yield these things as opposed to um you know the the bulk of the the new mutations that we see which come as i said sort of one by one uh throughout the population last word with you mugay because we've run out of time but what did you want to say finally i just wanted to add to what uh ravi said i mean i guess like for the lay um audience just to make sure that you know generally when we look at the majority of cases uh pcr which is the test that's often done is positive up to like maybe 17 days in average but most of the people will clear the virus quicker but you know there's a small subset of patients with maybe immunosuppression or severe disease who may have longer shedding so i guess like there was a question i think from one gp whether we need to test everybody uh i i don't think that would be uh like a general general uh approach unless like there's some concern um you know from a clinical perspective or maybe if the patient is immunosuppressed for example so because the pcr could be positive for a long time and we did a systematic review that found some patients may be positive for up to three months but that doesn't mean they're infectious or that doesn't mean they have like this ongoing disease i don't know whether it doesn't mean that if we do mass testing of people with no symptoms we're going to end up making people isolate when they perhaps aren't infectious that may happen anyway this isn't about testing and that's probably all we've got time for because uh it's 6 15. uh thank you to all our fabulous panelists thanks to the bmj uh please continue the conversation with the hashtag covid unknowns you can tell any colleagues who missed this you might be interested that if they go onto the bmj's youtube channel so go to youtube and search bmj british medical journal you'll be able to watch this webinar and all the others in its entirety we have a three-week break but we'll be back better than ever with a session on long covid so please join us again in three weeks time and thank you everyone for joining us enjoy the rest of your day

Channel: The BMJ

Views: 35,889

Rating: 4.6749115 out of 5

Keywords:

Id: vsbOfWLWlU0

Length: 134min 20sec (8060 seconds)

Published: Thu Mar 25 2021