natural treatment for the disease measles

good afternoon. welcome to the joseph j. kinyoun memorial lecture. niaid has hosted this lecture series since 1979 in honor of dr. joseph j. kinyoun who some of you may not know was a pioneer of modern biomedical research, founded the hygienic

laboratory in 1887. his work ushered in what we know and realize was a new era in america of combating infectious diseases through the rigid scientific research process. since then, his original one room laboratory which was housed in the attic of the staten

island marine hospital in new york harbor has evolved into the 27 institutes and centers which we now know today as the nih. his vision continues to inspire our approach to improving human health. next slide. it is my pleasure to introduce

dr. john mascola. the director of the vaccine research center and internationally recognized leader in hiv vatican research who has made -- vaccine research, made many seminal contributions to the field. he provides overall direction

and scientific leadership for the basic clinical and translational research activities aimed at developing and testing candidate vaccines not only against hiv, but other diseases such as influenza, ebola, chicken, zika and other issues of major global health

consequence. he also serves as one of my principle advisors on vaccines and related biomedical research affairs. and he is the chief of the virology laboratory at the vrc, where he directs research on structure based design and

testing of hiv and influenza vaccines on an optimizing the immune responses to and identifying the correlates of immune protection for these vaccines. in 1996, before john joined the vrc, he was one of a group of investigators who discovered

that easy to induce hiv antibodies neutralized only laboratory adapted viruses, and not those obtained from hiv infected individuals. and a few years later, he began a truly pivotal series of experiments on the biology of hiv neutralizing antibodies, and

was among the first to demonstrate that passively administered antibodies could protect non humate primary mats. what is a classic 2007 paper in nature medicine, he was the first to show cd4 binding site antibodies isolated from the plasma of hiv infected

individuals could neutralize diverse strains of hiv, and this led to further research culminating the desseminal 2010 paper that he coauthored in the journal science describing vrc01, the potent cd4 binding site antibody that neutralizes more than 90% of hiv strains

mature he, then, went on to lead to the development of vrc01 for preventive and therapeutic use, achieving a good manufacturing practice, production of the antibody, and launching ongoing human clinical trials to study it. in recent years he has led

efforts to map this, and isolated such antibodies in the setting. john is not only a visionary and highly productive scientist, but also a wise, generous person from whom countless researchers seek guidance including myself. please join me in welcoming john

mascola. [applause] >> good afternoon, and thank you for the kind introduction. i'm honed to ask to this give lect been and quite excited about a topic i discussed with tony, the idea of the talk will be to give you some sense of why

for some diseases for which we don't have a vaccine, we need to apply these tools. so the way i'll structure the talk is to give you background on vaccines in general. the successes but also remaining challenges. then i'll explain the terms.

i appreciate that b cell ontology is not obvious to everything everybody. the viruses are rsv. hiv and influencas. first let's talk about the public health impact for vaccines from centers of disease control.

the red bar shows the number of cases in a year of any given disease in a prevaccine era, the blue shows 2012. you recognize diseases, measles, mumps, hepatitis a. and note there were hundreds of thousands of cases per year in the prevaccine era, then this

are now hundreds or no cases at all. and one can look at the actual numbers, which i think is helpful. i use the example of measles, mumps, rubella, everybody knows their children get that at 18 months,s the get a booster

before they go to college. hundreds of cases go down to a handful. congenital rubella no longer happens in the united states, nor does polio, all cause of the success of vaccines. so how did we get there? we used what we now call classic

vaccine approaches. we can go up -- i'm using viruses as an example, bacterial vaccines use similar conceptual vaccine technologies. we can live attenuate the virus, we can kill it chemically. in more recent years, 1986 we had our first recomminate

protein vaccine, hepatitis b. in 200 oz we had our first virus particle vaccine. that was based on pioneering work of doug and john here at nci. so now we call these pretty classic approaches move and all of them stimulate the body's

immune system to generate both antibodies and i'll focus on the and i response here -- antibody response but also helper and killer t cells g vaccines give us life long protective immunity at a pretty high level. there are many human diseases for which we don't after

vaccine. i list some of them, not all here. we talk without the big 3 being malaria, tuberculosis and hiv. these cause such a huge global epidemic. these viruses, all diseases for which we don't have vaccines.

many parasitic disease i won't talk about, also in great need of effective vaccines. if we look again at cdc data for just the united states you can see the number of cases for these, and often times people don't think about hepatitis c or nora virus.

million 0 of cases for some of these. with sometimes thousands of deaths. even influenza, we have a modestly effective vaccine, tens of thousands of cases a year. i particularly like a little spot on the cnn website that

someone brought to my attention, it was during the 2014 ebola outbreak. i think each remembers we had very concerning imported cases, some treated here at the clinical center. but this little spot brought up the reasonable point that if you

think about the scariest viruses, even in the united states, may not be ebola. it was a nice educational piece, i thought. let's look outside the united states. so according to data that compiled annually by the worth

health organization, infectious diseases are the second leading cause of death and disability in the world, second to heart disease and stroke. we go back to these major disease entities, hiv, malaria respiratory diseases, cause millions of death

disproportionately involving children often times. in all these cases vaccine development could have a major impact. then lastly, we keep on learning about new diseases. we think of ebola as an old disease.

that was really recognized only in 1976. hiv only recognized in 1981. czars, chikv. mers, and you can assure there will be a disease down the line in a few years. need for vaccines persist, so we have many successes but we have

obvious clear viruses and other pathogens for which we don't have a vaccine. the question is why and what can we do? i'm going to explain what i mean and give you some examples. so by way of definitions, there are no dictionary definitions,

they're new concepts. i made these up and take full responsibility for any imprecision. for structure based vaccine design what we mean, and i use a virus as an example to make it more tangible, we use atomic level structural information to

manipulate or change a viral protein, to improve its antigenic profile and make it a better vaccine. around then b cell ontogeny was coined based on the classical definition of ontogeny that we know in biology, which is the developmental history of an

organism from fertilized egg to mature organism. we borrowed that concept and say that starting from a single b cell which produces antibodies, we studied the evolution of an antibody response, called antibody family, evolving through the process of

maturation. it may be oozier to look at it -- easier to look at it visually. we know hiving has a surface envelope lycopene. the envelope, we actually -- i'll talk about this, we have a structure.

we have antibodies that bind to that envelope. so you can think about structure based vaccine design, saying given a known antibody that we think is a good antibody, given the known viral epitope can we stein a vaccine to i listitate antibody.

if we know the structure, we should be able to do that. b cell ontogeny asks how did the immune system generate this the point, for some diseases, hiv and rsv, we need this kind of information or to make an effective vaccine. so one other point about

structure based b cell ontogeny, fairly new concept, a little more than 5 years. this is new. it's the work of several groups. i'll highlight hospital who have contributed but duke university, scripps research institute are major groups that have been

studying these things along with the vaccine research center and investigators at niaid. i think these concepts have been strongly influenced by hiv, but certainly now influenza and rsv fields have begun to use these concepts. all of this is still early in

the proof of principle state. the first vaccines based on structure design are moving into the clinic. i'll start with a couple examples. rsv to start with, which is probably the most straight forward example we have of

structure based vaccine design. an important point for viral immunology is that despite the fact that the immune system is complex, we need helper t cells, cd8t cells to control infection, a critical component of all vaccines and most in gem that we need strong antibody responses,

called them neutralizing antibodies. we have a very good vaccine for hepatitis b and antigens are critical, flu vaccines induce the antibodies to influenza. they could be better. and for hiv we seek to induce antibodies that neutralize hiv.

so i'll focus to a lawfirm -- large extent to the protein which you can see here. i'll start with rsv. worldwide, this virus kills more children, young children than any other pathogen except malaria, which is a major killer of children.

we have been working on a vaccine since the 60s without success. so we know a lot about viruses. we think we're good with vaccines but we haven't been able to make a vaccine for rsv. the question is why is that, you can read on the topic.

so the vaccine that was tried in the 60s fascinating story, made good sense. we had successes with polio and other viruses to make a formalin, activated fax. chemically kill the virus, it was done, made as a vaccine, tested in the 60s.

it turned out to induce a weak antibody response. we'll talk without that in a minute. importantly, the vaccine failed to work. did not project children. there was cases of enhanced pneumonia in children, a

complicated story that i won't talk about in detail. importantly, clearly, the vaccine didn't work. so despite more than 50 years of research we don't have a vaccine for rsv. i borrowed the slides on rsv to show you that viruses have

surface proteins, and a key surface protein on rsv is called effusion protein, what we called f protein. and so i'm going to talk about this f protein. and so we can look at surface proteins of many viruses. what's fascinating, all these

surface proteins actually are related. they're all trimeric, all what we call type 1 fusion proteins. they mediate fusion. and even though the viruses, influenza, hiv, ebola are quite different there are certain similarities and they all

mediate into the cell. so how does that happen? here is a viral fusion protein, could be any virus. let's take rsv. it has to contact a cellular receptor. the virus wants to get into the cell.

first, the protein engages the receptor, then the protein changes its confirmation and inserts a peptide into the cell surface and pulls the virus and host cell together and injects the viral cell fusion. so structure buyologists cause fugs proteins entry machines.

they have a thermo dynamic entry mechanism, they change confirmation and pull the veeris and host cell together. you can see that here, again, and the critical components of entry machine fusion proteins, they can be metastable. they're designed to altar

confirmation. they're often unstable, called metastable. so you can see the fusion of protein here. you can see after it changed confirmation, you have the post fusion confirmation. that becomes very important for

respiratory syncytial virus. investigators for years have known that if you look at electronic mikegraphy, and you look closely, you can see two forms of the fusion protein. one form turns out to be prefusion. the other post fusion.

and so investigators have color coded that in red and post in a little bit you ask why does a virus have post fusion on the cell surface? that's not what it needs to enter the cell. and part of the reason is that it's a metastable protein.

depending on the virus some of the protein triggers. so you can see that depicted here where over time, with temperature or leaving the virus in place at 37 degrees or low salt preparation, many of the prefusion proteins trigger to be post fusion.

this is work done by barney gram at the vrc, so in retrospect, one of the key explanations related to the vaccine that didn't work is that the pref protein flips to the post fusion, during the process of enactvation, 37 degrees sent grade and low dose term meld i'd

used to activate the fires. it's likely that the vaccine preparation looks like this. most of the protein on the surface in the post fusion form. so again this was tested in the 60s, induced a weak response and didn't work. what do we understand about that

now? so what we think in, and the hypothesis from that work, was that the vaccine that we would like to have given should have been a vaccine that had mostly prefusion f on it. that can be done by using a life attenuated vaccine, still active

research area, or by using more common, recombinant protein technologies, hepatitis b and others, around for years. we can express the recomiant pref protein. what investigators realized, we need to stable the protein. this is structure based vaccine

design to make a stabilized rsv protein. how does that work? so first of all, one needs structural information. the post fusion form of the protein was solved as a crystal structure in 2011. the prefusion was solved in

2013, so again, this is pretty recent work. often times in structural biology we knead and antibody to be linked to or bound to a protein to get good crystal. that's what was done here. this is an example of a crystal from peter's group, and this is

the publication by the laboratory, led by jason, where the title is the structure of rsv gleo protein trimer. we have the structure of prefusion f. what can you do? we want to make a vaccine. we have all the structural

information. so the goal is to maintain this and prevent this. if you know enough about the chemistry, that's possible. so what structural biologies biologists often do -- remember, this is a trimeric prepare. 3 idea call subproteins.

they use a ribbon diagram or ribbon appropriation to pull one of the subunits of the picture and see what the prefusion structure looks like in detail. this is the protein here in the cell membrane. this is the protein. and when the prefusion protein

triggers post fusion, the fusion peptide goes from here to way out here, sticks into the cell. so you need this confirmation change in order for the virus protein to work. so what we seek to do is maintain this structure and not allow this.

what investigators did is they looked at this. i'll give the simplified, non structural biologist view. you take these two amino acids, they're close together. i labeled this wrong. this is post fusion. ignore that.

in the post fusion these 2 amino acids are separate. very different space. so instead of keeping these, they were changed to sistine bond and made an engineered sistine bond to hold this structure in place. it doesn't allow this to change,

allow this to happen. there are other mutations used. cavity filling, hydro phobic amino acids that help keep the sometimes we put a domain on the protein to hold it together. so there were several mutations in total. we didn't just want to do this

in a structural space. we wanted to make the protein. these mutations in total allowed us to express a stabilized prefusion f protein and keep it stable. now we have something that can be used as a vaccine. when that was done, the

important data are that this protein is more immunogenic. several groups have tried rsv vaccines in the older sense and the vaccine in the protein was so the immunogenicity of post fusion vaccines are not very g what was hoped and was shown clearly in monkeys here in, in

immunogenty. prefusion protein is a much better immunegen. so this type of work is now progressing. this work is just 2013. this type of product and technical has been licensed. the vaccine center we'll do a

phase 1 study in early 2017. the concept of the phase 1 study will be to proven the concept that in humans this type of protein can boost the immune response. most people, in elderly populations, they would be preimmune.

you'd have to boost the immune in infants, this would be a better vaccine than going to go with a [indiscernible]. so this was one of the first examples of using structural biology to make or change a protein and change it its ontogeny profile.

for hiv, we need to use both concepts, structure based vaccine design and b cell ontogeny. why is that? hiv is a more complicated pathogen from the immune standpoint. for rsv we had an unstable

trimer, that could be fixed. if you look at the protein the surface is exposed. there are aify glycans or suggests on it. the opposite part of the spectrum is hiv, it's also unstable trimer. we would need to stabilize it.

completely covered in sugars or glycan, and very diverse, so the purple, more purple is a lot of antigenic diversity. with rsv it's simple, the immune can see it. here is a more complex situation. why that bad for vaccines?

the antibody needs at some point to contact a protein surface. and the gly cans get in the way. and antibody that works against one strain may not work against another strain. so antibodies, what we need for hiv are antibodies that can penetrate the glycan shield.

and this has been a major obstacle of hiv vaccine. it's not easy in hiv vaccine space to get the right kind of so this is evident by looking at natural history. point that dr. fauci makes, we learn from the natural history of a pathogen.

for hiv, one gets viremia and very quickly, we get antibodies to the virus. that's how we diagnose the infection. it takes month to get neutralizing antibodies. if this was rsv, polio, these antibodies would come up in

weeks. with hiv, it's month to years. that's the first clue this type of pathogen is dif. the type we really want, more cross reactive can take years to come up. we have -- for hiv, we have an immunological problem in

addition to a structural problem. this translates in the real world because we have tried several types of vaccine candidates in hiv, as vaccines, and they have failed to work. so gp120 is a simple recombantprotein.

if this is a trimer, this is one of the subunits. we have used gene based vaccines, jeeps into a genetic vector. or combine dna and adenovirus. we've combined pox virus effecters with protein. these failed to work.

they did not protect in placebo controlled trials at all. this protected at a 31% efficacy, not good enough yet. we need to build on that. so importantly, we get poor neutralizing antibody responses in all these vaccines. by that, low potency and limited

cross reactivity. so we have a sense from the antibody perspective why that happens. again, we immunize with gp120. the antibodies elicensed do not effectively bind the trimer. we gets lot of and is to gp120, but the surface is buried on the

trimer. if this was rsv it wouldn't be a we want antibodies that wiped to the trimer and neutralize the virus. how do we get there? so first generation vaccine proteins, gp120 induced high binding, more neutralizing

natural infection genrates strain specific neutralizing. but the broadly neutralizing antibodies that we want with the vaccine which cross reacts with many strains are found only about 20% of hiv infected people. we don't understand why that is

but part is the chronic nature of hiv and it invades the immune system. years ago, we were working with mark here at niaid, and studied subjects that were part of nih clinical trials. so here these are 80 individuals all in the x axis, hiv infected.

each dot is virus. you can see that some donors make weak neutralizing antibody responses, for most veerises they don't neutralize. codons neutralize most virus at high titres. this was a first clue, interested in these donees.

these donors made potent antibody responses, the kind we need with a vaccine. what can we do? we can isolate the antibodies from these advice. i'll show you how. we can study how they neutralize structurally and ask how the

immune system makes them. we can do the structural virology on these donees. the technology has advance sod we can take an hiv probe, mix it with b cells. these are hiv specific b cells. we can take an individual b cell and amplify the antibody genes,

that technology has been routine in the last five years. so from a b cell we can express in the laboratory, the antibody made by that c bell receptor. so if we do that, and i'll just ahead some years now, since 2009 to about 2016, the field has isolated literally hundreds of

potent neutralizing antibodies against hiv. these define the epitopes on the so each of these blue patches, they bind here, at a sited called b3 gly can. it interacts or touches the glycan and touches thispa or a green patch.

or the c de4 binding site. so we understand where these viruses attack. i can show you a very nice video made by peter at the vrs, where knowing this information should make really very specific videos like this showing the antibody in the right size, so this is

antibody pinning to one eptowed or another where the cd4 binding site. you can see how big antibodies are. you can, i think, appreciate that at the chemical, structural level, we know these interaction actions, attic for attic.

how they wind into the virus. so i borrowed a slide from dr. fauci to make the point that the challenge here is translating the fundamental challenge, epitopes to vaccines. we know the epitopes, we need a vaccine, a specific way to immunize to induce these

how? we basically need to move away from just recombinant proteins to being able to use the trimeric structure. that's what antibodies bind to and what's required to neutralize. if we're not immunizing with

native structure. so rsv call it prefusion, the hiv people call it prefusion structure, the native virus we need thing that are native, otherwise we won't get antibodies that recognize the so just since 2013, a lot of this is new biology,

investigators have solved the crystal structure of the native glucoprotein envelope, done here, the wilson and ward groups. in fact, that is a stabilized so similar to rsv, really independently, hiv researchers realize if you had to put a

sistine bond, and here antibody is bound to the protein allowing crystal structure. we can see the crystal since 2014, peter and other laboratories have given us atomic level structural details of not just the protein, but actually, the glycan shield

around the protein. so we know exactly what we want to immunize with, with the immune system. the problem still lies in the diversity of hiv. if we immune nice with a trimer, we get antibodies to that but these are too strain

specific. they don't neutralize diverse viruses. that's what we need. so we're still left with a diversity problem. we crossed the threshold in that we can do structure based vaccine design.

we have to solve this immunological problem for hiv. so we go back to looking at our very important donors, so we have examples in natural history, in natural infection, of people that macros reactive these neutralize 90% of strains. so what we want to do is follow

that b cell response, we call b cell ontogeny. and we want to follow it from the beginning. so we would like to study an individual from early on in infection, 2 or 3 years. we can do that. there are very important

clinical cohorts. i won't go through all the names. but these are often nih sponsored or funded cohorts where individuals are followed, high risk. followed from the time before infection through the time of

often times through several years. so we find an individual, we study that individual, we pull out the antibodies year 2 or 36789 and we find antibodies that ra very potent and broad. we go back, pull out out a lot of antibodies.

now we're talking about antibodies in one family, binds the cd4 binding site all we want to study how that first b cell made that antibody population. that's going to tell us how the immune system did it. so the example of that is here.

here are naive b cells before they've seen antigen. one of the b cells sees hiv when that becomes infected. that b cell divides so there are hundreds of thousands of b cells all daughters of the same b they make what we call family of antibodies, all these antibodies

are the same clone. same colonial family. so we can study this whole family by just pulling out one once we pulled out one, we know the genetics. and we can use pcr to find all the relatives. we can find any sequence.

if we have longitudinal samples going back in time we can study how that antibody went from its first b cell to becoming potent, and that's what was done here. work we did with part haines. first we pulled out the then what bart haines did, he studied 2 action.

he studied first the virus that infected the person. the beauty of these cohorts. one has the virus. so you can see that early on in infection, a virus is what we call -- very colonial. the virus diversifies. i can't?

it's mutable but also invading the immune system. you can show specifically as the virus mutates the antibody responds to a particular epitope and evolves. the virus mutates further and the antibody mutates further. this focuses in on the more

conserved epitope. this whole process, we call virus coevolution, very important because what we can do is we can see for any given antibody, the antibody mutates. this is the process of -- somatic hyper mutation. as it mutates it gets better and

better. it becomes neutralized. we can show that here. transmitted virus and the virus mutates. and the antibody mutates other time. we know that antibody virus coevolution happens.

in detail to understand as a so if we merge the b cell evolution, we have an understanding of what happens. and this is one example of some complex examples, but if we take an antibody to the cd4 binding site, something we studied very heavily, the cd4 binding site is

buried in here, where the virus engages cd4 on the cell surface. there is somewhat of a glycan hole. the binds site, it has been learned, that they all approach the virus in a specific restricted angle. that's the only way in.

if an antibody -- if an antibody to the cd4 binding site tries to come in a different angle, it may bind cd4 as a protein with you not the right angle as a this is metastable, so during infection, it's she had. the immune system sees cd have. the antibodies are around.

that sees a real trimer, and it can't get in. but over time, there is a maturings, intermediate yeahs and mature. the mature antibodies have the right angle approach, accommodate glycans, find the protein surface and neutralize.

the immune system has to go too a lot of work to neutralize hiv. that's what we need to do withi]a so i can see that here. what i just showed you. so we can see in this donor who made very good binding site antibodies what the virus looked like at the beginning, we can

see that the virusw3 mutated,how the antibody responded. so it gives us something that could be a sequence ofã±rvaccines. if we recapitulate the process here with just changing the virus a little bit, we should be able to push the antibody response toward the glycan hole

and what we want. this concept is called linage based vaccine design. we have to know the linage of the antibody to get there. it is frankly something that's unprecedented in vaccinology. we don't have to do that for rsv.

we don't care about linear based vaccine disease. when you immunize that whole we care about it less for flew, but it's critical for rsv. so the approach for hiv is quite complicated. it starts with neutralizing antibodies from people that bind

the stable trimer. requires structure based vaccine designing we've done that. we can stabilize the trimer and makeã§ã³ subunits. we know what that trimer looks like in some atomic detail, we can make parts that actually fative.

we've now -- we have examples of b cell ontogeny, and we are starting to do b cell linage design. the first product for b cell linage design are going to the clinic. so that sequence of immunegens that i showed you has been made

and proof of concept cities will be done. when we talk about antibodies and hiv, we have this whole area of vaccine design which will take some years. it's obvious that for rsv, we may we on to something that could translate in a few years,

and for hiv, all this work is going to take experimental medicine over some years. it's hard to do. so another approach is what we call passive immupization. this takes advantage of the fact that we have antibodies from hiv infected people that are very

potent. so active immunization, we induce the antibodies, passive means we passively infuse antibodies a byproduct of this beautiful structural work, we have all these antibodies bound to the virus, is that we have antibodies in the lab, we started to make them clinically.

so there is currently an advanced clinical trial that asked the question, can a passively infused monoclonal antibody prevent hiv infection in high risk adults. this was isolated at the vrs, administered at 2 different doses every 8 weeks, and a

pretty large phase 2 clinical trial conducted by clinical networks, one called preventions trial network or vaccines trial network. and first of all, appears in friday populations where in april -- african populations. you can see an volunteer getting

infused. the antibody would be given by a subcutaneous method in the long run. this is enrolling with over 600, 200 enrollees and going well. we hope that this trial will give us an answer, several answers.

can an antibody block or prevent hiv infection in humans? everything i told you about this presumes that to be true. so it would be nice to experimentally confirm it. it would tell us how much antibody we need, critical for vaccinology.

if the level needed, if it circulates through the system, gets to the mucosal surface, we believe that the amount doesn't need to be that high as long as it's potent. we need to know that in human trials so it helps set a bar for so that gives you a sense foer

for where we are with hiv. i'll end with an example that someone in the middle of rsv and hiv. so we have influenza vaccines, i think everybody should be in the process of getting their flu vaccine on campus here. that's classic.

or you might hear the term split virus vaccine. what we do transplantation or what manufactures do is [indiscernible] the vaccine so it releases the protein. that's activated subunit, so semi purified. and there is also a life

attenulated internasal vaccine, more often given to children. not recommended this year. but interestingly, all of these vaccines lack durability. we get immunized every year. you get polio as a child, you continue get it again. essentially get close to life

long immunity. flu, every year. we change the strains in the vaccine most years. andrtheã§ã³ efficacy of influenza vaccine is 30 to 50% at best. so clearly, we would like todoã‘i so let's go pack to our principles of structural

biology. what do we know about the flu? it's a stable trimer. even though it's related, so one of the reasons we probably have a vaccine, one doesn't have to stabilize. if you make hemoglobin. >> it's stable.

look at apology -- key areas of and then lastly, way too many people to think. the laboratories at the vr.c that have participated in the hiv part of the work, the labs, [list of names]. i will stop there. thanks.

>> thank you for for that elegant lecture. questions or comments? do you have a question? good. >> [laughter] john, that was spectacular. thank you. do you think the rsv enhancement

problem will go away because of the higher titres, or what do you think will be the outcome there? >> i think it should potentially be solvable in a situation where you have the predominant antibody response to be neutralizing.

with the prior vaccines, the response was weak and non left for the possibility of viral aggregates and deposition in lung tissues. together with that, mediating cd8 and responses. i think the sense is, you don't get the off target responses and

should be able to avoid [indiscernible] >> john, in the sequential vaccine model of hiv, where you immunize and sequentially coax the immune response, the question that i get concerned about is that if you have the initial priming with the

immunogen, is the secondary ard tertiary and fourth one one ihow common is that from one individual to another? how do you know that the immunogens are inducing the next stage? >> this is obviously a fundamental question.

when we study a single individual that i showed you, you could wonder and i hope you did wonder, that's one person. so is this generalizable? i would have to say when i came into it, we studied those individuals. my bias it may not be

generalizable. we learned and i changed my view. what we learned is that we have these fantastic examples in flu, for example. that the responses are often -- there are classes of responses common among many people.

so i think we used this term b cell linage design to say if we're smart enough, for certain categories of antibody, cd4 binding site for hiv, most people can make the response we're interested in. if our initial immunogen and secondary immunogen are designed

the right way, they will stimulate category of antibodies that are common to everybody. it still needs to be proved. that it's generalizable. but i think that the b cell lineage data, and now tens of people rather than one person, tell us that it should be

generalizable in that regard. >> for that coaxing along i assume you want to accelerate it, not take 2, 3, 4 years. can you give more opwhat the regiment of the trial design would be. >> this is a fundamental problem with hiv.

you saw the nice graph there. is that in part because the earn' is infected? cd4, the germinal centers are destroyed. we hope in the setting of immunizing a healthy donor, the germinal center response will be normal and the immune response

quicker. but also there are certain principles like adjuvants, and the protein, and learning how to stimulate a response that has to be applied for hiv. there is a process of maturation, so the antibody in rsv does not have to mature very

much. doesn't have to mutate much. it can go through a reaction and be a memory b cell and be effective. for influenza, it may have to go through several reactions and mutate a little bit. for hiv, we probably have to

promote mature ration at a greater level. potentially months of time, hopefully not years. but probably months. okay. thank you very much, john.