SARS-CoV-2 Vaccines and Variants

[Music] hello everyone i'm rick lifton president of the rockefeller university and i'm delighted to welcome you to another of our sessions virtual discussions with genuine experts this is our first session in 2021. today's speaker is the remarkable virologist paul binash whose research has transformed our understanding of viruses and their interactions with human hosts over the last 11 months our nation and the world have experienced the tragedy and continuing challenge of a global pandemic of staggering proportions more than 100 million people have been infected with the sars cov2 virus the cause of covid19 and more than 2 million people have died from complications of this infection the us has been particularly hard hit with more than 27 million cases and over 460 thousand deaths the global response to the pandemic has had many dimensions public health efforts clinical efforts to figure out what works to keep people alive along with politicization of the prevention effort but without doubt the world has benefited mightily from the foundation of basic science that has fueled the rapid identification and characterization of the virus that causes this disease the development of tests for the virus the identification of the viruses achilles heel affording efforts to develop new therapeutics and the unprecedented development of highly effective vaccines in less than a year a truly remarkable accomplishment these vaccines can prevent up to 95 percent of symptomatic disease and nearly all severe disease including cases in individuals at high risk of severe disease and death which increases dramatically with age beyond age 50. these highly effective vaccines are now being distributed and provide a crystalline example of the enormous impact of fundamental basic science on human health in response to an extraordinary public health emergency and yet just as these vaccinations are scaling up basic science enabling the rapid sequencing of the viral genome to monitor them for emergence of variant viruses has identified a growing list of variant viruses that threaten the efficacy of vaccines and therapeutic antibodies rockefeller scientists have played critical roles in the global response today's speaker paul benosh has been one of the leaders in this effort along with many other rockefeller investigators he was able to hit the ground running early last year thanks to years of foundational basic science conducted in our laboratories in fact paul his wife theodora hatziono charlie rice marina caskey and michelle nussenzweig have led research that has identified key targets on the viral spike protein for antibodies that prevent viral infection and have identified mutations that have predicted how the virus will evolve to escape these antibodies and in recent months as we've begun to see the emergence of variant viruses that have acquired mutations that may reduce the efficacy of these vaccines paul and his colleagues have tested how well these vaccines are likely to fare when they encounter these variant viruses the current uncertainties demonstrate that we will likely need a range of therapeutic and preventive agents to finally finally bring this pandemic under control we are grateful to all of our benefactors past present and future whose support has been critical to advancing the research that will get us out of this pandemic your support is vital to our efforts to to battle covet 19 at an accelerated pace before turning the podium over to paul i will give you an update on the current state of uh the covet 19 pandemic this is the current state of how many uh cases there are every day and as you can see starting back last february until uh now cases have had quite a variable uh rise a few peaks occurring along the way and then a dramatic acceleration in the fall with the bump after thanksgiving another bump after christmas hitting a peak of more than 800 000 uh per day worldwide and now we're in the midst of uh a risk reduction in cases uh over the last uh month uh to date there have been more than 106 million cases worldwide uh and more than 2.3 million deaths in the united states similarly there has been a lot of activity in a lot of cases again in three distinct waves and compared to our last webinar in december where we were concerned that we were headed toward 200 000 cases per day we accelerated right through that by christmas hitting over three hundred thousand on a single day with a peak of about a seven day average of about 250 000 cases per day but over the last uh month there has been a steady decline in the total number of cases with a daily average now of about 125 000 cases per day nonetheless the number of deaths per day has continued to rise and while in december we were lamenting having a peak of nearly 2500 cases approaching 3 000 cases per day we actually hit over 4 000 cases per day by the time we hit the beginning of the new year and we're still continuing with a very high death rate across the country globally the pattern is complicated it's quite variable and heterogeneous across the globe the united states continues to be among the uh most active areas with the prevalence of virus worldwide although it's been improving as we discussed still parts of europe with very high levels israel despite the vigorous vaccine campaign still has very high levels but they're beginning to come down uh and it continues to show this pattern that we've commented on in the past where southeast asia and very large parts of africa appear to be relatively spared of cases and deaths from the virus this is what the distribution looked like across the country last december it rose to a much worse uh uh pattern than this uh by the middle of january but uh has now been coming down substantially uh to a pattern that uh uh looks much improved lately and this uh obviously has been welcome news this is what the pattern looks like in new york city where you can see that we hit extraordinary highs in the beginning of january and have been slowly starting to come down uh this uh apparent uh drop is attributable largely to the snowstorm last uh week that uh dropped the number of cases that were a number of tests that were being done on a daily basis and we're still clocking along at over 4 000 cases per day in recent days we still have a significant number of new hospitalizations that's just barely beginning to tail off and the number of deaths per day uh is still relatively steady at about 75 deaths per day in the city as we've discussed previously the vaccines target the entry step of the virus getting into the cell this occurs through the viral spike protein which interacts with the ace2 receptor on the surface of some respiratory epithelial cells that allows it to gain entrance into the cell and go through its life cycle and then gets budded uh and released into the circulation to infect other cells so this has been the target for the vaccine all of the vaccines that are in clinical trials or beyond a wide variety of approaches to the vaccine have been taken these have included um inactivated or attenuated uh viruses that have been pursued by several chinese companies peptide vaccines uh and novavax released early results from its phase three clinical trial engineered viruses predominantly adenoviruses and we now have uh uh reports from johnson and johnson uh the russian company gamalia uh and oxford astrazeneca uh and we now have two approved uh rna vaccines which had never been done before uh and these are the rna vaccines from pfizer and bioand tech which we talked about last time and moderna's rna virus which is very similar which reported out and has been approved by the fda for emergency use uh the this is the data on the moderna rna vaccine which is newly approved since our last meeting and it looks very similar to the pfizer vaccine the main one of the main differences is they spaced their doses of vaccine by 28 days rather than 21. so after the first dose you can see that uh starting about day of 14 or 15 uh the after the first dose before the second dose the curves start to uh part with the number of cases in placebo rising but the number of cases uh in the vaccinated group staying relatively flat and by the end of the trial after the second dose they ultimately had 185 cases in placebo and only 11 cases in people who had been vaccinated 30 severe cases in placebo and none in the vaccinated group uh so the efficacy for prevention of symptomatic uh covet 19 uh was 94 extremely effective very similar to the performance of the pfizer vaccine in uh more recently the uh gamalia adenoviral vaccine reported out results of its uh phase three clinical trial and this has now been published uh last week in the lancet 22 000 people a two-dose vaccine they had i think the best approach at an adenoviral vaccine the concern with adenoviral vaccines are you given adenovirus once the immune system is stimulated learns to recognize not just the viral proteins from cyrus cov2 that you want it to but it also recognizes the adenoviral proteins and makes it harder to come in with a second dose and not have the immune system uh react and diminish the efficacy because it gets cleared before it can infect any cells and give you any added boost so uh gamalia had the good idea of using two different adenoviruses adeno 26 which is rare it rarely ex people are rarely exposed to this virus in the human population endovirus 5 in contrast many people have been exposed to the virus about 40 percent of americans uh some parts of the world up to 70 or 80 percent of people have uh circulating antibodies but the thought was that if you come in with this virus second if you haven't seen the virus for a long time the immunity may be waning and you can get a good boost from that i'll just mention that uh the other adenoviral approaches have used the same adenobot adenovirus for both dose 1 and dose 2 which is an interesting strategy that i don't think was likely an optimal choice and people have been rethinking whether you want to be doing two doses with the same adenovirus and this may have in part motivated johnson and johnson's goal of seeing what they can do with a single dose of vaccine in any case the gamalia phase 3 clinical trial was all done in moscow similar uh outcomes of uh scoring symptomatic infections and this shows the departure of the curves of placebo and vaccinated individuals and this looks like a very successful uh vaccine as well uh with about 92 percent uh efficacy uh consistent uh uh across all the ages studied although this was a somewhat younger population than in the moderna vaccine trial where about a third of the population was over age 60. but they also had no moderate or severe disease among the vaccine group this is a very good result so we now have uh two vaccines that are registered in the united states uh the pfizer and the moderna vaccine uh we believe that per the administration's recent efforts they now have secured commitment to 300 million doses of both the pfizer and moderna vaccines by the end of summer there's not i haven't seen an official statement that that's been solidified but since the end of january that was what the expectation was and of course when you do the math for a two-dose vaccine together you could vaccinate 300 million americans which obviously would cover virtually the entire adult population there have not been clinical trials to date on people under age 16 and most trials have started at age 18 so there's very little expo experience in the pediatric population to date but now companies are beginning to pursue trials in that population as well johnson johnson has filed for registration i'll come back to their phase three trial uh in a moment and novavax as reported out in the interim as well so this shows how the vaccinations have been rolling out in the united states this is a slide from bloomberg uh and uh obviously we had a kind of a rocky rollout with uh very uh uneven distribution and delivery of vaccines this has been smoothing out over the last uh several weeks and recently hit a high of 2 million doses delivered in a single day and if you do the math you're going to need pace uh at least that much uh and likely a bit more to get to complete delivery of two doses to 300 million people uh by the end of the summer but that's a laudable and i believe worthy goal uh to be shooting for at this point obviously the manufacturer and distribution and delivery remain logistical challenges at every point along the pipeline but these vaccines are starting to roll out worldwide and israel has been the first to really get going at a high level they now have 62 doses per 100 people in the population and have 23 percent of their population that have been fully vaccinated almost all of these are people over the age of 60 and this is a in fact a high fraction of that population and as a consequence investigators have been able to look at the infection rate in that population before and after vaccination and also comparing infection rates in that population compared to the change in infection rates of younger people and have shown a clear difference demonstrating clinical benefit in the general elderly population of getting vaccinated and this is obviously uh extremely good news the united states has delivered more vaccines than anybody else and is doing quite well at the international level in terms of getting people vaccinated as a fraction of the population so the fly in the ointment has been since we last met in december several variant viruses have started to emerge this is not altogether a surprise but it obviously is worth noting and is of concern why should we be concerned about variant viruses viruses as they replicate acquire mutations all the time and most of these are going to be random occurrences that are going to have no functional consequences but if any of these mutations confer an advantage to the virus either in its ability to survive to get out of the cell and propagate or to increase its ability to infect cells those changes will be selected for and if they have these advantages they could take over the population of viruses that are circulating in a population relatively rapidly and could make it tougher to fight off the virus on you with your own immune system but also in particular could be more resistant to the existing uh vaccines or for people who have previously been infected uh to be able to reinfect people because the virus may now be immune to the immune system response that a person made when they were originally infected so these are obviously matters of great concern to be followed as as you may know some viruses are very successful at evolving in this way uh such as the influenza virus and that explains why we need to get an influenza booster every year because the virus has changed substantially from year to year but other uh viruses do not have that facile ability uh to evolve rapidly the measles viru virus for example uh we give that once and we uh typically have lifelong uh immunity uh to uh to the virus so these uh mutant viruses uh uh are rare in occurrence but uh if multiple changes that collectively may be beneficial to the virus can occur particularly if the virus is allowed to replicate in a host for a prolonged period so imagine for example somebody who has a compromised immune system uh who can make antibodies some antibodies to the virus but can't make enough to clear the virus that allows in a single infection the virus to continue to evolve in a patient and acquire many mutations some of which may be beneficial to the virus and some of which in combination may be particularly beneficial to the virus and this is how these viruses can go from having no useful mutations potentially to having several mutations that benefit the viruses ability to avoid the immune system and or to gain access to cells through increased infectivity so there are three of these three of these variant viruses that have been emerging over the last couple of months one of these is the b 1.1.7 virus which has gained a toehold that has increased in frequency uh dramatically in the united kingdom so this is the dis the genome the rna genome of uh the coronavirus and as you can see the mutations are not distributed so the mutations are shown as these uh spokes of the wheel and as you can see they are not distributed randomly across the viral genome in fact they're highly clustered predominantly in the spike protein which is the protein that gains access uh through it uh binding to the receptor to get entry into cells and a couple of other proteins that are derived from uh the viral rna so these mutations in the spike protein are particularly noteworthy and uh concerning and uh paul benosh will talk uh about the impact of some of these individual variants uh in these uh in these viruses so this is uh the virus that is now predominant in the united kingdom there's a second virus that is of concern and that's a virus uh called [Music] b .1.351 and this is a virus that is has virtually completely taken off over infection in south africa and again you note that there are very few mutations anywhere in the genome except in the spike protein and these are clustered in the spike protein in this case and it is noteworthy that there is one mutation that is shared between the mutation found in uh the uk uh strain of the virus and the south african strain of the virus this is quite unlikely to occur by chance and suggests that this variant may be playing a role in uh giving this virus an advantage over other viruses circulating in the population and then the third virus that has uh emerged to date is prevalent now in brazil and this is the p uh p1 uh coronavirus and again you can see this uh clustering in the spike protein again not likely a chance event and in particular uh you will note that there are in this case three variants uh this uh mutation that alters uh lysine at position 417 to threonine uh a glutamate to a lysine mutation and an asparagine to tyrosine mutation these three variants are also seen in the variant that is circulating in south africa uh and yet there does not appear to be a common ancestor between these two viruses suggesting that these have emerged independently on these two lineages and are extremely unlikely to have occurred by chance and are likely playing an important role in development of the disease in these individuals so these emergent viruses are clearly of concern and we're starting to gain some clinical experience which augments the work that paul and his team have done in evaluating the impact of these biochemically in the laboratory but these three variants cluster now we're looking from the top down on this top surface of the spike protein and the spike protein is a trimer it has three subunits of the same protein and you can see for example the e48k variant is present at three locations as are the uh other two uh mutations in the protein so these are right at the surface of the protein where the virus is binding to its cellular receptor to make contact and there is strong suspicion that these are altering that interaction and maybe masking the ability of antibodies to recognize uh these particular sites particularly in people who have previously been exposed to the virus that does not have these mutations so this has been looked at already in the trial released just in the last uh two weeks from novavax so this is the peptide uh vaccine that tested vaccine versus placebo we haven't seen anything other than the press release from the company they studied 15 000 subjects in the united states many of these were old over age 65. it was a two-dose regimen at the first interim analysis they showed a 90 reduction of symptomatic infection in the vaccine group a fantastic result particularly when you consider that half of their patients had the uk variant of the virus and in fact in the uh people who had the uh uk variant they had about 86 percent uh efficacy for reduction of symptomatic infection uh whereas the people who had the uh the classic version if you will of the virus had about 96 percent protection from infection so this is clearly very effective on these viruses however they also did a study in south africa where they studied 4 400 subjects and they found a reduced efficacy of their vaccine it only provided a 60 reduction in symptomatic infection in their vaccinated group that was not infected at the same time with the hiv virus and in this population as expected the vast majority of infections were with the south african variant virus that i've described to you so this is a strong indication that this vaccine is highly efficacious quite similar to the moderna and pfizer vaccines on the older versions of the virus but on this south african virus it had markedly reduced ability to prevent symptomatic infection we don't know with certainty how well it worked to prevent severe disease or death because most of these subjects in south africa were young and were not predisposed to developing severe disease johnson and johnson similarly reported out results of their trial of vaccine versus placebo as we discussed this was an adenoviral vaccine where they used a single dose of the adenovirus 26 viral vaccine again carrying this vaccine carries the spike protein of uh the sars kobe 2 virus inserted into uh this viral capsid again about a third of the patients were over age 60 and 28 days post-vaccine they found protection against moderates to severe infection of 72 percent in the united states uh but they also did a south african uh cohort and they found uh 57 protection in uh south africa and they found 66 percent protection uh in another cohort in latin america so they had about 20 000 in the u.s uh 18 000 in south america and 6 500 in south africa they reported on an overall 85 reduction in hospitalization they did not fractionate that so we don't know uh how that looked in south africa uh but again likely a younger population in south africa but i have not seen that data reported but once again uh because you see that the single dose of this adenoviral vaccine was quite effective not as effective as the moderna and pfizer uh and uh uh other vaccines uh but only 57 uh reduction in south africa and we don't uh really know with certainty what effect it had on reducing hospitalization and severe morbidity or death so this is where we are with the vaccines so the great news is there are a number of efficacious vaccines the serious questions and concerns are that we have these variant viruses that are starting to spread around the world the uk variant now is the predominant variant uh in the united kingdom uh less so elsewhere there are now cases across most of the us uh the south african virus uh is all over that uh country has spread less through the rest of the world thus far this appears to be a more worrisome virus but we don't have enough data to know what impact it has on whether people will get infected will they get severe disease uh with this virus much more than they would have with prior iterations of the virus what will happen with further uh dissemination of this virus will it overtake other countries as well or will it not be able to get a toehold and we don't know much about what the course of this will look like thus far particularly when you have these different viruses competing with one another in the same population there have only been a handful of cases reported in the united states thus far but we haven't been sequencing viral genomes that much in the united states so we really don't know how extensive the spread of that virus or the brazilian variant of the virus has spread into the population thus far so a lot of questions that we'll get to momentarily when we turn to paul but i do want to give a couple of updates of other research from rockefeller laboratories since our last connection in particular i've told you previously about monoclonal antibodies that have been developed uh by michelle nussenswag and many colleagues shown here on this slide this has progressed to first in human trials they're a combination of two monoclonal antibodies uh derived from uh the genes that encode these antibodies uh from that were taken from patients uh who had recovered from sars cov2 uh infection the two that have been put into clinical trials are by far the most potent of these monoclonal antibodies and one of the possibilities that is being looked at is whether this uh these monoclonal antibodies might be efficacious when given uh as subcutaneous injections rather than intravenous infusion uh the uptake of the existing monoclonal antibodies has been limited uh because of this need for intravenous infusion of the antibodies for either preventive or therapeutic uh purposes and this has been a limitation and we will see whether these new uh monoclonal antibodies uh have efficacy uh as subcuta in a subcutaneous formulation uh and also uh will be critical for these monoclonal antibody therapies to know how efficacious they are in uh confronting uh these variant viruses that are now circulating in the population uh other work that i'll mention uh briefly uh there's it's been recognized that uh there are there's a need for other therapeutics uh for these viruses uh oral uh orally available drugs would be very advantageous if you get infected you give it a couple of pills and you start on treatment immediately and there are efforts to pursue novel targets for this tarun kapoor has uh worked on an essential enzyme required by the virus for its uh function uh and this is a an rna methylase that caps the rna and protects it from degradation tom tushall has worked on sorry he's worked on the helicase which unwinds the viral rna after it's been replicated and tom tushall has worked on the methylase and both have identified promising leads that have gone into medicinal chemistry at the tri-institutional therapeutic discovery institute that's a shared resource of rockefeller sloan kettering and weil cornell and bob darnell i should mention has developed a spectacular uh very inexpensive very accurate rapid test for the virus that we've had in use here at rockefeller for the last seven months and we currently are testing everybody on campus uh once a week to make sure that we do not have any viral spread occurring on our campus which i'm happy to say we do not have so it's been extremely useful to us uh it has not been lost on bob and others uh that this could be a very useful test in other settings as well uh and in particular as we've debated in new york city how do we keep the schools open when we have children who aren't going to be vaccinated anytime soon one possibility would be to have more aggressive testing programs of course having a test that is very inexpensive and very easy to administer bob's test is a saliva based test doesn't require a healthcare worker to take a swab up your nasal pharynx and it's a very attractive option and this is being explored now i will now uh come back to paul benosh so i've already told you a little bit about paul and his work he leads rockefeller's laboratory of retrovirology and we're proud that paul is also one of the nation's howard hughes medical institute investigators paul has dedicated his career to reducing the human toll that's exacted by infectious diseases educated in england paul came to the united states in 1980 1996 to do postdoctoral research at duke university and three years later he was recruited to new york to join the faculty at rockefeller along with the aaron diamond aids research center and he was promoted to rockefeller professor in 2010. paul has been a leader in hiv research identifying mechanisms by which hiv overcomes cellular defenses revealing potential targets for therapeutic intervention he and his wife theodora who i've mentioned already expanded the possibilities of hiv research by creating altered forms of hiv that can be used in animal models to correctly to accurately mirror infection in humans paul's team has since developed increasingly sophisticated techniques to manipulate viral genomes so when the pandemic hit paul and theodora quickly pivoted to apply their expertise to the study of covit 19 and have done crucial research that has made key contributions to both their own work and collaborative research with other investigators at rockefeller and at the new york blood center and across the city paul's lab has made many contributions to our understanding of how the novel coronavirus infects human cells and how the human immune system responds to infection this has made paul a go-to expert for journalists reporting on the evolution of the virus and on the biology underlying vaccine-induced immunity his research has been vital in identifying the steps we can take to prevent and treat new infections in the ongoing pandemic so i'm now pleased to turn the podium over to paul at the end of his presentation we will answer your questions you're welcome to submit questions at any time using the q a function at the bottom of your screen and now i'll turn it over to paul it's uh my pleasure to join you this afternoon um thank you very much for the introduction rick and what i'd like to do is is give you a sort of an update of what we've been studying in terms of antibodies and sars kovi ii in the context of what we've just heard about the vaccines that have been deployed and the the new variants of the virus that we're beginning to see circulate in various countries around the world so rick already covered some of this but let's just we're going to drill down a little bit in some details as we go through the the biology of the spike protein and the antibodies that recognize it so as you heard the spike protein decorates the exterior of the virus particle it's the key part of the virus that touches the surface of the cell and allows the virus to enter into the cell and it's also a key target the key target really of the antibodies that provide protection either from re-infection or after vaccination and so as rick mentioned it's a trimeric protein so one of these three subunits is highlighted here in color there are really two parts of it that are extremely important for the recognition by antibodies the the ntd or n-terminal domain is a target for some quite potent antibodies but really the most important target for the the very really the most potent neutralizing antibodies is this rbd which is the receptor binding domain this is the part of the protein that flips up as you can see in this this blue blob here that interacts with the ace2 receptor and many of the antibodies that are neutralizing that are most effective at interfering with virus replication bind directly to the rbd and block its interaction with the ace2 receptor okay so before we get into the details we need to have a little bit of background on b cells and antibodies how they are made um a little bit of terminology so circulating in our blood and lymphoid tissue are b cells there are tens perhaps hundreds of millions of different b cells and they're different because they have different what we call b cell receptors on their surface now when we're infected by a virus for example sars cov2 a small subset of those b cells perhaps tens or hundreds of them will have the ability to recognize the proteins on the outside of the virus particle in this case the spike protein is what we're particularly interested in now that recognition event triggers something in the b cells they grow and divide they grow and divide and after a while they stop growing and dividing and start to make antibodies now during this process of growth and division the ant the b cells they are essentially selected for the ones that are best able to interact with the viral viral proteins and as part of that selection process the antibody genes themselves become mutated and and they are selected for higher higher tighter binding to the to the viral proteins so the antibodies that are eventually made are a little different to the antibod the b cell receptor genes that are initially displayed on the b cell surface they have been mutated matured as we call it um into forms that bind tighter to the to the to the viral protein and this is a really quite important uh feature of antibodies now once these antibodies that are as i say directly derived from these b cell receptors float around in the circulation if there are viral particles around they can just as they did on the surface of the b-cell interact with the viral particle and a subset of them what we call neutralizing antibodies will then prevent that virus particle from being able to infect cells antibodies work in a second way if the cells manage to get past this first line of de defense and the the cells do in fact become infected the antibodies can recognize the very same protein the viral spike protein on the surface of those infected cells and a cell that's coated with antibodies like this serves as a flag to other cells of the immune system to come and kill and eat that cell so antibodies really work in in two major ways one is by neutralizing virus particles the other is by targeting virus infected cells for death and both both mechanisms are very important the neutralizing activity of the antibodies is is perhaps especially important and it's really especially important to us because we can very easily measure it in the laboratory okay and we can show that that ability to neutralize is is really a very important component of how antibodies work okay so over the last almost a year now we've been working very closely with a team rockefeller uh in particular charlie rice and michelle nelson swag to really understand the nature of that antibody response that i've just told you about in the context of sars kovi 2. and this rather intriguing diagram that you can see on your screen now is sort of a representation of a large part of that work so each one of these slices of this circle or pie as we as we call it represents one individual and those individuals labeled in green those are people who have had sars cov2 infection everyone else represented here these individuals around here they are people who have been vaccinated either with the pfizer or the moderna mrna vaccines and these slices represent antibodies okay they're within each slice is represented several tens up to a hundred or two hundred or so individual different antibodies that have been cloned from those individuals the antibody cloning process itself involves taking individual b cells sequencing the antibody genes from those individual b cells after they have been through this process of of of cell division and mutation um sequencing those those antibodies and those sequences are represented in these slices here now the black parts of these slices represent antibodies that are found multiple times within that individual so this cov 107 this black slice represents a number of antibodies that have been found multiple times the white part of the slice represents antibodies that were found once now the interesting and important part is that is these lines that connect the slices these represent antibodies that were found in multiple different people essentially the same antibody sequence of is found in people who have had infection and individuals who have been vaccinated and every line here represents a common or shared antibody from convalescent individuals or vaccine recipients so what this diagram tells us in total is that there are significant differences between individuals in terms of the types of antibodies they make but there are also quite a lot of similarities indeed to the extent that there are some nearly identical antibodies that are found in many many different people that's really quite important for understanding the environment in which the the virus is uh replicating and learning to adapt to to the human immune system it also tells us that what we get when we vaccinate people the type of immunity we get is really rather similar in qualitative terms to the immunity that one gets from from natural infection and that that's quite important to know as well so part of our job over the last year has been to sift through these hundreds even thousands of antibody sequences to identify the ones that are most potent and those are the antibodies that rick just told you about that are now in clinical trial those uh antibodies that really very low concentration nanograms per ml can neutralize infections so as we add more antibody we get less infection and really tiny amounts of these antibodies uh nanograms of these antibodies are sufficient to to stop infection and that that's that's very important from the term point of view of of therapy now um from the very earliest times that we were identifying and developing these antibodies as potential therapeutics we began to think immediately about the issue of antibody resistance and really posed this question how likely is it that antibody resistance will emerge in the context of science cov2 now understanding that the antibodies that we are cloning from human individuals to use as therapeutics the antibodies that are generated in response to infection and the antibodies that are generated in response to vaccination are all subsets of each other there's overlapping types of antibodies it's as you i'm sure you can imagine quite important to understand if the virus becomes resistant to certain types of antibodies how will that affect antibody therapy immunity to reinfection and the efficacy of vaccines and so as i say we've been thinking about this for some months now so in thinking about how likely it is that antibody resistance would emerge in sars kovi 2 we do i guess a couple of thought experiments so let's imagine a transmission chain where sars cov2 is being transmitted from one individual to another to another um going through what we call population bottlenecks when a virus is transmitted from one person to another usually involves a very small amount of virus sometimes as little as one virus particle we don't actually know what that number is in the case of sars cov2 but certainly there are a smaller number of viral variants that are transmitted than are present in a individual who has um florid uh virus replication in them so the probability that in this particular individual that's marked with an asterisk that that the virus might mutate change from an antibody susceptible uh population to one that's antibody resistant is dependent on a number of things first how accurately does the virus make copies of itself so we know that in the case of coronaviruses they they do have an error rate they make mistakes at a number we that we can define and we know that that number is quite a bit lower than viruses for example like hiv or influenza or hepatitis c virus so the the what we call the um the error rate of the replicative machinery is is is quite low however the antidote to replication fidelity in terms of exploring genetic diversity is population size and we know that the population sizes in sars kovi 2 are enormous a single swap can contain a billion different viral rna molecules and so by doing a little bit of math we can understand that let's say sars kovi 2 makes a mistake once in every 100 000 one one at each position in the viral genome there's a mistake made one in every hundred thousand times if the population size is a billion it stands to reason that um just about every possible individual mutation that can occur in saskov2 does occur probably in a significant fraction of individuals and certainly in the global population okay so in terms of thinking about how whether and how virus has the capability of acquiring um mutations it its population size is so enormous that every possible mutation that can be made has been made so there there are other considerations in thinking about how likely it is that antibody resistance will emerge to size covington okay so in order for antibody resistance to occur it will occur the question is will it be selected and will it be amplified in human populations and so in thinking about that we began to think about uh things like transmission dynamics so in the in the in the case of sars kovi 2 what we know is that the about a virus in an individual and likely with it the probability that they are infectious to other individuals occurs quite soon after infection so this is a this is a cartoon representation of what we think goes on in an individual but it's probably in broad broad terms reasonably accurate so we have a peak of virus replication within the first sort of five to seven days sort of coincident with the onset of symptoms with it the probability of transmission and then the amount of virus begins to decline over the ensuing days extending into weeks the antibodies that are developed in response to that infection typically come up sort of a little later than the peak of viremia so one can imagine in a transmission chain like this the virus um quite rarely experiences selective pressure by antibodies before it's transmitted to another individual and that likely underpins the the the observation that we haven't seen a lot of antibody resistance in size cov2 until quite recently because the virus has been um spreading largely in an immunologically naive population now of course these things can can change things like once the the population a large fraction of the population has been infected they've made an antibody response and then those antibodies decline we know that that occurs in sars cov2 infection and we know that some individuals are beginning to be re-infected so that changes this dynamic here it can be a situation that an individual that that is that is experience infection already has some neutralizing antibody and that antibody can can in principle impose a selective pressure on that virus and enable the virus to experience that selective pressure and adapt to it other things like monoclonal antibody therapy obviously we and others have combined antibodies in cocktails to try and ameliorate this uh this possibility but applications of antibodies as individual therapies could in in themselves impart selection pressure on the virus also tens of thousands of individuals across the world were given convalescent plasma therapy in instances where the plasma was not well characterized so again there could be sub um sub neutralizing levels of antibodies that could impose selective pressure on the virus but but not be sufficient to prevent onward transmission of the virus and of course as vaccines are rolled out if vaccines are more or less effective that is going to affect how the virus experiences selective pressure um by antibodies and and the crucial point here is that if the virus is is able to encounter antibodies in an individual such as this but still maintain sufficiently high levels that that virus can then be transmitted onto other individuals then one can in principle get transmission chains like this where an antibody susceptible virus encounters an antibody a particular antibody and then that antibody loses its effectiveness because the the selective pressure imposed on that population has favored the propagation of antibody resistant strains so from quite early times in the pandemic we develop methods to try and study this problem this is obviously something that's a little difficult to study in the laboratory people would perhaps get a little upset if we took sars kovi 2 and tried to generate antibody resistant versions of it in the laboratory so what we did was to build a tool to study uh antibody resistance and we did so by using a virus we call vesicular stomatitis virus it's a very distant relative of rabies virus it's it's quite a safe virus to use in the laboratory it's actually used as a vaccine vector in some instances there's a version of this virus where the ebola virus spike protein has been inserted into it as an as an effective vaccine so what we did was to take vsv to remove its own spike protein and replace it with the sars kovi 2 spike protein that gives a virus particle that is vsv this vesicular stomatitis this rather safe virus on the inside but it has a sars cov2 spike protein on the outside so it has the target cell tropism of sars cov2 and most importantly the neutralizing properties neutralization properties of sars cov2 the same antibodies that can neutralize sars kov2 can neutralize this virus we also gave this virus what we call a reporter gene it's a gene from a jellyfish that makes cells glow green when they're infected and that gives us the ability to track virus replication in cell culture and so the experiment that we do is to take this virus we generate large diverse populations of this virus in cell culture the the replicative machinery of this virus has a rather higher error rate than does sars cov2 so we can represent quite a lot of diversity in smaller populations of virus so we generate diverse viral populations we mix those virus populations with antibodies those can be individual monoclonal antibodies from individual b cells or the mixture of antibodies what we call a polyclonal plasma that's generated in individuals where hundreds of different types of antibodies have been made so we take those virus populations mix them with antibodies and then we simply put them onto another cell culture okay so members of this diverse population that are antibody sensitive will be blocked from infecting these cells and lost from the population these escape variants that we call them antibody resistant variants will infect those cells they will replicate be amplified and then after a couple of days just takes a couple of days to do this we can see which viruses which variants have escaped the inhibition by these antibodies by sequencing them we can even isolate the individual viruses and under go back and understand did they really get uh resistance to the antibodies that we applied in this experiment so i'm just going to show you some actual real data so these are little dishes can full of cells um they're about two centimeters in diameter and they've they've got cells uh laying on the bottom and we've infected these cells with about a million variants of that recombinant virus that i just told you about now if we do that without antibody all of the cells in this culture turn bright green and we can see them in the fluorescence microscope this sort of green speckly pattern is infected cells now if we add an antibody then most of the cells do not get infected because this antibody is very good at neutralizing the virus that we have made however if we look closely in this area demarked by this little white square we can find little areas of cells where the virus has infected and began to begin to grow and this signifies the the presence of an escape mutation a variant in the sars kovi 2 spike protein that has learned to resist the effects of that monoclonal antibody and so we've done this many times now but one of the very first mutations that we identified by this approach is this mutation e484k so let me just show you the experimental result so here's the virus that we begin with it's a vsv virus that has a sars cov2 spike it's very efficiently neutralized by these three antibodies that were part of the cloning effort here at rockefeller tiny amounts of antibodies can totally block infection now after having done the experiment that i just showed you growing that virus in the presence of this antibody c121 then what grew out was a virus what that little group of green cells you sequence the virus that comes out of that and it's this virus e484k okay this is one of the very first mutants that we we found during this approach and when we started to publish this work about six months ago it's i think it's fair to say that the field underestimated the importance that um variation in the spike protein would eventually um the importance that it would take on and this is a quote from one of our colleagues in the field when we began to talk about antibody resistance i'm not worried about this at all the dogma in the field was that um antibody resistance in coronaviruses was not not something that really should be particularly worried about um now here we are six months later and uh i it's not that infrequent that i take phone calls from journalists and all they want to talk about is e484k um and of course now we're we're quite uh concerned i think about the the emergence of the variants that that rick told you about and of particular interest is changes like e484k are appearing in multiple places in the world particularly in the brazil and south africa variants that you've heard about the uk variant originally did not have e484k but just in the last week or so it's been reported that there are now derivatives of that original b117 uk strain that in fact do have the e484k and it's very likely in our view that these are mutations that are arising in response to antibody selective pressure we're sort of emboldened in that view by a number of experiments that we have done recently so again using this experimental scheme where we mix the virus with a number of different monoclonal antibodies we've now done this with a rather large number of antibodies this table shows some of the results that we have gotten and so what we're doing in this experiment is to do exactly this experiment mix the virus with individual monoclonal antibodies those monoclonal antibodies are listed down the left side here these are all cloned from people who have received vaccine but as i've shown you they're basically the same types of antibodies that exist in people who have been infected and what this table shows are the percent percentages of viral sequences with the indicated mutations these mutations across the top basically after 24 hours of replication in the presence of each of these individual antibodies and if we just go through the variants that come out of an experiment where we allow antibodies to select the viral variants that can resist those antibodies what do we find well we find mutations at position k417 that's exactly the same place that we find mutations in the south african and brazilian strain n4 n439k that is a variant that's actually achieved quite significant numbers of cases in scotland initially and now in parts of western europe n440 those mutations are achieving quite high prevalence in india l452r that is the so-called california variant if you pay attention to the news in southern california you'll you'll find that people are very worried about this particular variant in california f456 and n460 haven't yet hit the news but um i confidently predict at some point in the not too distant future they will e484 we've already heard about there are a number of antibodies that seem to select for variants at this position another two mutations not yet in the news n501 again is present both in the uk strain the south africa strain and the brazilian strain and so the impression we're beginning to get is that these antibody selection experiments that we do in the laboratory are in a sense predicting the types of variants that are now rising to prominence in nature this is basically the same result it's it's really a confirmatory type of experiment in this case we actually make viral variants with particular mutations and then ask how much antibody do you need to neutralize those viruses and if the starting virus what we call the wild type original strain you need rather small numbers of small amounts of virus this is the number of nanograms you need to neutralize that virus but if those viruses are engineered to have these resistant mutations what's shown in red here here and here are the the the variants where you need much larger and sometimes you can't neutralize the virus with and with whose spike protein has those variants and again k417 e484 and n501 are prominent in these types of experiments but there are other variants that um that are present in nature that do also appear to confer resistance to individual antibodies that have been cloned from human individuals now with with uh pamela bjorkman's group um we've been trying to understand this in in structural biological terms and this is the work of christopher barnes and and pamela and so what you can see here this sort of gray cloud shape is the receptor binding domain the rbd portion of the spike down here is the spike protein is sort of a lower magnification version and here's the rbd and these squiggly lines represent different antibodies that can interact with the rbd and you'll see that each of these mutations that we think are quite important they form part of the binding site on the rbd that's recognized by these antibodies and so mutations at these positions really make it more difficult for those individual antibodies to recognize uh the the rbd now that this all sounds um quite worrying um that the virus is making mutations and it's causing really quite high levels of resistance to these individual antibodies but there is some good news okay uh and some of the things we found lead us to think that this this will be a problem but perhaps not quite as serious a problem um as the impression that i might have given to you thus far in the talk might um might suggest so one of the things that that we've been finding is that this process of maturation the process that the b cells go through once they have recognized the viral protein the viral antigen they go through a process of mutation the antibodies get more and more mutated and they're selected to be higher and tighter tighter binders to the to the to the viral protein this seems to have at least in some cases quite an effect on their ability to cope with resistance mutations that the virus might acquire so in effect in response to the um it's not just the case that the virus is evolving the antibodies can evolve in individuals too and so in these two examples what we have done so michelle's group cloned antibodies from individuals very soon after their infection and then some months later after the antibodies had acquired higher levels of mutation and had learned to bind the the virus tighter and you can see that if you if you plot on the y-axis here antibody potency so the lower you are on the axis the less antibodies required to neutralize the virus so better antibodies are plotted lower down on the on the ax on the y-axis you'll see we have the wild-type virus represented here and then a whole bunch of variants that are found in in circulating populations represented on the x-axis here what we find is that antibodies that are cloned very shortly after infection there are quite a number of variants represented by these green dots here where the antibody has lost its potency conversely if you take antibodies later after infection then they are much more resilient so mutations here that cause resistance to the the 1.3 month antibody they are much less able to um to avoid antibodies that have been generated after a longer period of time very similar example here so the antibodies themselves can get better they can mature with time okay the other thing that's quite important is that while we've focused on a particular particular sets of mutations it's it's not the case that all of the antibodies are focused on one spot on the on the viral spike protein so i've talked primarily about the rbd okay the receptor binding domain this blue part this blue part here that sticks up and recognizes the ace2 viral receptor and we found a large number of mutations in the rbd that can confer mutations to individual antibodies but what you'll notice from these sort of colored patches that mark where antibodies um antibody resistant mutations occur is that they occur at a large number of different spots both within the rbd itself and also on in other places on the spike protein so every position here that's marked in red marks a position that we think is important for recognition by antibodies that can neutralize the virus so while we analyze the effects of antibody resistance one antibody at a time it's almost never the case that one antibody at a time will be asked to neutralize the virus the when we're immunized or infected we make dozens perhaps hundreds of different antibodies that all of which act together basically to cover each other's back to to try to neutralize the virus so really the the the issue in terms of becoming of vaccines becoming less effective and antibodies antibody resistance mutations becoming important really is a numbers game how many different mutations does the virus have that confer resistance to how many different antibodies and so we're beginning to understand that and this is the sort of experiment we do where we analyze the effects of groups of mutations so the the k417 e484 and n501 mutations that you've heard us talk about over and over again what do they do when they're all together in one viral spike protein and they're asked to avoid the whole cohort of antibodies that are generated by an individual that's either had infection or been vaccinated and that that's basically the result you see here so every point that's represented here we've plotted what the plasma neutralization potency the mixture of antibodies in plasma how good they are at neutralizing um the original virus and then this this constellation of mutations that we find both in the south african and the brazilian virus and you can see that for for most individuals there's perhaps a small or no declination in the potency of that plasma versus the wild-type virus as compared to the mutant viruses for some individuals there's a large change you know this is a logarithmic scale so we're talking about a 50-fold drop but that's pretty rare for most individuals there's a rather small change it perhaps gets more exaggerated towards the bottom end if you have low levels of neutralizing potency then it looks like um those plasmas are less good at coping with with the with the mutations and so what we anticipate is that um really giving people that as much immunity as possible um perhaps through booster shots perhaps through changing the vaccine to better match circulating strains is what the outlook will will be in the future okay so let me wrap up with a couple of take home messages so the sars cov2 spike protein elicits similar potent neutralizing antibodies in convalescence and vaccine recipients inevitably antibodies exert evolutionary pressure on sars kobe 2 and can select for mutations that can confer resistance to individual antibodies but the effects of those individual antibody mutations are ameliorated first by antibody maturation the antibodies get better over time the polyclonal nature of the antibody response the fact that many antibodies are going after the spike at once at least in most people and of course there are other components of the immune system t cells for example that we've we strongly suspect are important but they're just uh more difficult to study and more difficult to show demonstrate unequivocally uh that that they are important all right let me wrap up by acknowledging again the fact that this is a this is a huge team effort um all of the work in my laboratory is co-led by my partner in science and life theodora hatsuyanu i'd acknowledge in particular three post docs frauke mitch fabian schmidt isca weiss bloom who've been working incredibly hard since the beginning of the pandemic to to bring this work to fruition uh other other members of my lab who've contributed enlisted here as i've mentioned michelle nus and swag's lab have really been key in this effort um in many aspects both of the clinical work and the antibody cloning itself charlie rice and pamela bjorkman have also been key partners in in the work that i've talked about today okay thank you very much for listening thank you paul that was really a tour of force and fascinating i uh i realized that uh the hour is late but uh i do want to take some time for questions for those of you who can stick around uh paul one of the things i'd like to ask you about is you focused on the polyclonal serum how about the different antibody cocktails that are approved for use or are in development for treatment right so so that that is uh individual depending on which particular cocktail that that that has that's been used so regeneron have theirs lily have theirs astrazenecan have theirs and of course rockefeller we have our own cocktail um in general um what we're finding is that because the antibodies for inc that have been selected for inclusion in those cocktails have been selected to target different parts of the rbd what in general we're finding is that the the new variants take out the activity of one of the two components of the cocktail but that leaves the other member of the cocktail uh fully potent and able to do its job this really was the reason why um regenera and ourselves and others used cocktails to to exactly to cope with this situation i anticipate that over time the cocktails will be updated particularly as we get these better antibodies that have gone through the affinity maturation process and are more resilient to these mutations thank you i should uh mention that in congratulations that the rockefeller antibody cocktail just last week was licensed to bristol myers squibb for clinical development and therapeutic and preventive use which is a terrific achievement there are a lot of questions that converge on the issue of how long do vaccines work and how long do you retain immunity after a natural infection right so so that was a point i intended to but forgot to make um so we know that in in natural infection the the magnitude of immunity the levels of antibodies um that are generated in individuals is incredibly variable there are some people that make very low levels of antibodies some may people that make really high levels of antibodies and that correlates at least to some extent with the severity of infection and we suspect it's a reflection of really how much virus replicated in those individual more virus more antibodies more subsequent immunity for vaccination um there is variation but it's over a narrower range than natural infection because the amount of antigen the amount of vaccine that's injected into each person is a lot more consistent how long that's going to last so i think in the context of natural infection weeks to decades right that that's i think i can't be more precise than that it's going to be very variable uh in the context of vaccination i think the answer is going to be a bit more consistent but we i we just have to wait and see i i would anticipate based on the levels of antibodies that are generated in response to vaccination that you know hopefully years okay it really depends a little bit on how the what happens as the variants play out because as what as i'm sure you can imagine protection is going to depend not just on how much antibody you have but how well matched it is to the circulating uh viral variants um and that a combination of effects will will determine how long that that protection lasts given these uh challenges several people have asked about the potential and prospects for a pan coronavirus vaccine interesting so we we just submitted a grant application with precisely that goal in mind um it's it's quite complicated and difficult i don't think it's going to be possible to generate a vaccine that protects against all coronaviruses i think it might be possible to generate a vaccine that protects against all the what we call sarbaco viruses these are the bat coronaviruses the group from which the original sars virus and sars cov2 came from plus we'll we'll already have some immunity to that group of viruses by virtue of being vaccinated against sars kov2 i think it'll be possible to build on that to provide protection against at least that group of viruses how broad we can get i don't be on that i don't know at this point paul a number of people have asked about uh whether the vaccines protect against primary infection or just this the development of symptoms or developing severe disease can you still be infectious after being successfully vaccinated so that is that's a very interesting important question that i'm afraid i'm not sure i can answer with current information now it given the the sort of the rather extreme potency we've seen with the mrna vaccines in terms of protection from disease it seems um almost inescapable that there will be at least some protection from infection reduced viral load reduced infectiousness benefits not just to the vaccinee but also to the people that they they come into contact with really the the best vaccines don't just protect the person who's been vaccinated but protect everybody that they come into contact with by reducing the amount of virus replication in that person to zero okay i don't think that's going to be true with with the vaccines that have perhaps lower diminished efficacy or as the as is happening in the uk when people are given one out of the two doses of the mrna vaccines these sort of partially protective vaccines while they might completely prevent hospitalization and severe disease there is likely to be some level of virus replication perhaps confined to the upper respiratory tract what that's going to mean in terms of terminating or continuing transmission chains we just don't know at this point we're going to have to let the epidemiology tell us the answer to that a number of questions came in about how concerned should we be about adverse events following vaccination and how does that track with age well i have to say i'm a virologist and i'm not a clinician i have not been involved in administering vaccines to people there do seem to be sort of rare uh allergic reactions um nobody has yet died they have all been treatable with things like epipens or benadryl um and i as a as i understand it they are they are extremely rare um given the numbers of people that have been vaccinated these seem like pretty safe vaccines when i say pretty safe i mean pretty safe as compared to other vaccines that are extremely safe so that they're i would say just as safe as as any vaccines that uh you've been used to yes i think the i think it's fair to say the uh uh number of severe uh reactions have been uh quite limited most people who have had them have had a history of anaphylaxis to other vaccines or other agents there has been some interesting data recently on people who have been previously infected who then get vaccinated uh having more potent immune reactions to the virus but again those have generally been easily managed another question that has come in is uh what are your thoughts about the mrna vaccines versus the other uh you know adenoviral or protein-based vaccines any reason to suspect that there will be differences in efficacy or duration of effect so not really i think you have to take each vaccine sort of as it comes and basically follow follow the science the it it's it's hard to say that a protein vaccine is better than an mrna vaccine or vice versa or better than an adenovirus vaccine they all have their their their sort of strengths and weaknesses in terms of um how easily they're delivered but you can't really say just by just knowing what kind of vehicle it is whether it's going to be a better or worse vaccine so if i understood your closing comments uh about the impact of these variant viruses my impression is that you think that there will be more people who develop symptomatic disease than compared to people who don't have these variant viruses but that perhaps the incidence of severe disease where people end up in an icu and at high risk of dying will not be completely attenuated and these will still continue to be efficacious vaccines is is that a fair assessment of your view that's fair um the caveat being that i don't have a crystal ball i i think it's true to say at this point that even in situations where the vaccine has has been the original strain and the ch the infecting virus has been one of these new variants nobody has yet died in that situation so it's clear that there's still um in in every case that has thus far been looked at um protection from the severest disease and and death now we obviously hope that that will will continue it may very well be that the the vaccines need to be update updated to maintain that situation um what where the uncertainty is at this point i think is whether those whether that degree of mismatch between the vaccine and the um the circulating virus will enable more transmission chains to to continue whether it will enable sufficient levels of virus replication and mild really mild disease to to be propagated through the population that that's really important for uh and for achieving things like herd immunity okay does the vaccine not just protect you but other people that that's that's a situation that we really would like to get to uh but you really do need a vaccine that that squashes virus replication enough that you're no longer a transmitter that's that's sort of a key distinction i think then my last question uh seeing the uh multiplicity of mutations that uh are emerging on these uh resistant uh viruses uh there's been a lot of speculation in a couple of cases documentation uh that they may be coming preferentially from people with compromised immune systems where you can mount an immune response but can't clear the virus and so the virus keeps getting multiple shots on goal to evolve in a single patient how do you view this and do you think there's an effective response that we should be making to take special precautions with people who have compromised immune systems yes one could argue that at this point it might be closing the door after the horse has bolted but yes i have seen a couple of reports where there have been i would say impressive amounts of sequence evolution in single individuals who have had compromised immunity and have been treated with poorly characterized convalescent plasma exactly the sort of situation that i outlined where you're putting sele enough selection pressure on the virus to allow it to continue to propagate and adapt to the antibodies that are thrown at it exactly what we do in the laboratory as we're studying um this process now on the flip side if you have a sufficiently large population of people who have weight immunity that's waned to a level where they can be infected but can still propagate the virus you have effectively the same the same situation right so whether the evolution occurs in a single individual or a population of individuals who have uh diminished immunity is is is sort of moot we it's going to be i think difficult to trace these variants back to their source and i actually i think it'll be impossible um you know but to answer your correct your question directly that yes i think individuals who have compromised immune system and persistent infection should be treated quite cautiously um in terms of isolation for example great well there are many gear heads who have stayed with us all the way to the end and we could go on for a very long time but i think it's time to draw this session to a close i want to thank paul for a spectacular talk incredible work uh congratulations and thanks for joining us today for those of you who have been joining us for these virtual discussions we will continue to keep you informed in the coming months your philanthropic support has been and will continue to be vital to our continued success uh to combat covid19 please see the next slide for more information about how you can contribute we invite you to continue visiting our website for more information as well as for videos of our past sessions we'll also be announcing future webinars on that page thank you once again for participating today have a good evening [Music] you

Channel: The Rockefeller University

Views: 14,927

Rating: 4.609756 out of 5

Keywords: The Rockefeller University, Dr. Richard P. Lifton, Dr. Paul Bieniasz, COVID-19, SARS-CoV-2, viruses, COVID-19 vaccines, SARS-CoV-2 Variants

Id: 8HhFDK07SZ8

Length: 90min 39sec (5439 seconds)

Published: Wed Feb 10 2021