 Hey, thank you very much Andy. It's first of all, it's my pleasure to introduce you and the panel The panel on R&D strategies and trends with focus on the big eye innovation And and as you already know, we have an impressive panel of R&D heads who have both big pharma and biotech experience And so it really promises to be a wide-ranging discussion The panelists themselves are Andy Klump from Takeda, Dave Reese from Amgen Hal Barron from GSK Mattai Mammon from Janssen And as alluded to it'll be moderated by the infamous Martin Mackay He's a co-founder of Rally Bio and also previously head of R&D at Alexion We're really looking forward to an exciting discussion. So Martin, please take it away Well, many thanks indeed Deval for that very kind introduction. This is always an exciting panel And as you introduced everybody, I won't spend time on that but just to say that Andy and Mattai are old hands at this They are they're used to the the roller coaster We're going to go through in the next 40 minutes And many thanks to David and Hal for volunteering as two rookies As it were on the panel not rookies in the world of R&D, obviously, but I'm truly delighted And I just wanted to pause for a second and really say to Karuna and Andy well done You have you have pulled this meeting off. I think everybody wondered, you know, could it be done in this format? Well, I've I've attended most of the panels and they've been terrific And the the fireside chats I found enthralling and and whilst I miss you all I miss the the the reception that follows this and the dinner afterward I actually think the content Is as good if not better than we've ever had at this meeting So well done to everybody all the panelists the fireside chats truly terrific And so then I've got some questions lined up here as Folks know I I try to make this as interactive as possible Even between the panel and with the audience It'll be slightly more of a challenge with the audience this time But I think there'll be a number of things bouncing off you folks and that will be great. I will attempt to moderate um At the risk of stately obvious covered has been a theme throughout it and it would be remiss of me if I didn't ask some questions Particularly with this panel of r and d heads and actually I have two questions for you But I'm going to take them in turn I'll tell you them both now, but I'd just like you to answer the first one and the first one is You know, just what you're doing around covered And you know, what's happening in your own companies and also really importantly This was a point that howl brought out when they were preparing for this how much you're collaborating as a group My second question then and as I say, I'll take it in turn We'll be To go into what impact has it had on your other programs And in r and d in general. So two separate questions So let's go with the first one, which is, you know, what are you doing in this covered a world about covered? and And I'm going to start with you david throw you, you know, right into the deep end david Sure, I'll try to swim and These are our big shoes to fill So I would say, you know, we're doing a number of things against covet 19 as are of course all of the panelists here In addition to participation in the, you know, broader efforts across industry and partnership with government and academia You know, I'd call out the few things our subsidiary decode and decode genetics in iceland Has published a couple very important manuscripts in the new england journal of medicine Detailing the initial course of the epidemic In iceland and then just a few days ago the antibody response of patients You know, we owe an enormous debt of gratitude to the icelandic patients who have participated in this And are continuing to do so on a longitudinal basis So this will actually provide some of the best longitudinal Data where we have essentially complete follow-up on patients in the island in addition, we are working on potential neutralizing antibodies as our, you know, other panelists companies here And then finally we are studying immunomodulatory agents. So, you know, across the the breadth of COVID-19, you know, feel comfortable and and quite happy with our contribution. This is an all hands on deck moment And I think we all feel that this is nothing that a single company organization government is going to solve We're all in it together. And much of this has to be viewed as a public good Oh, very very good. And davids. Thank you. How how was it going at lexosmithcline? Going through it. Well, I think thanks for asking our approach for cobit really is is taking three different approaches to seeing how we can provide solutions Underline these three solutions. Let me back up and say that, you know, such an unprecedented experience and we've Really took two philosophies. One is that, you know, we have to have multiple shots on gauze from multiple companies Because if we rely on anyone Approach it may or may not work and we need backup plans. So we were Focused on trying to figure out where we could add our unique value as one Um, and that because each company probably has expertise in certain pieces We believe that the collaboration was going to be key and you can see that through our program so the first area was vaccines and uh, we have numerous collaborations most Notable as one with synovia where we have begin the vaccine testing quite soon and we are providing For that collaboration And our adjuvant, which is a very special Adjuvant for the pandemic Which allows the antigen to be more immunogenic or more likely reduce the the dose of the protein so that When you think about the biggest problem that we thought we'd be facing was in addition to showing efficacy is having Be able to manufacturing at scale so that Reducing the dose of the of the protein by maybe even five fold possibly tenfold You can consider that to be the same fold increase in number of patients you can reach for the same manufacturing So that was something that we thought was very unique to gsk And and and something we wanted to ensure it was adopted by anybody developing the agent based Approach and so we have six or seven other collaborations using our pandemic adjuvant the second approach is Was early on when we started seeing this inflammatory syndrome this cytokine like storm and You know, there was a lot of Excitement about potential for aisle six being critical to that But but the truth was many cytokines were elevated And now I think we realize that based on the steroid data and based on the aisle six data That there's probably a need to quiet the immune system by targeting more than one cytokine The approach we took was to inhibited Using an antibody to inhibit a protein called gmcsf Which is the protein we think is most responsible for the monocyte macromage activation That we think might be fundamental and as well as unusual So we think this by reducing multiple cytokines may potentially be an option for patients to the trials to bear out for this More acute end stage sort of inflammatory situation And the third is um, as david mentioned, we're developing an antibody monoclonal for neutralizing the virus with a company called veer Collaboration that we're very excited about and this monoclonal We think is unique in the sense that it's identified from people and maybe we'll get back to why that's so important But from people who were covered from SARS and we were looking for an epitope that was conserved in SARS-CoV-1 and CoV-2 so that if the virus became Mutated over time that we would potentially hopefully have an antibody that would be effective despite all the mutations So that that's part three approaches Marvelous how and it's you know, it's interesting from a No, I'm almost political point of view that uh past leaders in gsk people that went before you are still at the sharp end of this pandemic And thinking particularly about monoclef and patrick valence. I mean there and I would rather be in your position than there as I must say Yeah, I think your your point about the sharp end might be absolutely correct Very good. Thank you matai Jonson j and j please tell me or so, you know We considered where we were going to be able to best contribute. Um, there are two areas. We have a long history in immunology and uh, you know many many ways of approaching, you know modulators Um, and then there's of course our vaccine platform. So first on the immunomodulatory side There was an observation by verily Um, Andy Conrad and the team they're verily pretty early on That there seemed uh, no correlation between uh viral load or level of iremia Estimated in a particular way and symptomology. So we knew that the we know still that the immune system Play some important role in both symptomology and ultimately maybe really bad outcomes And so exactly how to approach that is uh, is is not clear even to this day Uh, I remember hal and I had a quick phone call about that. We could equally rationalize activating or inactivating certain pathways and It's interesting, but we've we've taken the lead from certain registry data and um and other other data that we have access to and are making several attempts on the immunomodulatory side probably, um Better known right now from from some of the media is what we're doing on the vaccine side We have a vaccine platform that is ad 26 that has been quite useful for different vaccines including Zika and ebola. We have a major attempt in in uh, both rsv and hiv So it's been in a lot of patience And it works in a in a reliable way. It's stored easily. It has a good Fert one dose effect. It has all that so it's convenient and we spent time there We didn't jump right in, you know, we didn't jump right in with the spike protein But rather we made numerous variants with a collaborator dan baruch at harvard at the vethers of deaconess And we went through a selection process and we whittled down to a couple major candidates and then selected one In the june timeframe and then we went like mad to get this thing in the clinic We have a special cell line that can grow at 26 at very high volumes And so we used that and we entered the clinic on july 21st and we're You know waiting for all our data to read out from that phase one to a study to make decisions on how how to best Whether to and how to best go forward into phase three, but i'm feeling really good about where we are Very good matthew. Thank you. Um terrific terrific work going on there Andy at takeda Yes, so martin maybe just to talk thematically about how we approach this which was how could we enable The ecosystem and how could we work together to achieve common shared goals? And we reasoned at the very beginning as everybody on this panel feels that this wasn't a business interest This was a mandate that we had to step up and do what we could and so Everything that we've done along the way has been collaborative and you know examples We're a vaccine company, but we don't have the history that gsk does We don't have the a differentiating platform like j&j And so what we decided to do is step up our manufacturing capabilities and we'll do distribution in japan and asia partnered with other companies We've got four or five repurposed molecules Good mechanistic rationale and so actually david am gen takeda a couple of other companies are involved now in a series of platform studies really a novel way of working um in this space One of our capabilities is plasma, you know, this very arcane technology, but complex to Purify hyper immune globulin and so we're doing that with other plasma companies and we'll be stepping up a study with The n i a i d starting shortly Hopefully at the tail end it will look different than what we're seeing with plasma in terms of some of the background noise um, and then lastly working with uh, Novartis and gilliage rodinger Mushy and other companies in the research space and we're not a virology company But our scientists were so motivated to leverage some of the libraries and capabilities that we've had that we started You know, it doesn't take a lot to start a drug discovery program We've really made great progress and now thinking about next steps of how we can continue to work together Not just to ensure that we're solving COVID-19, but to ensure that we're ready for the next pandemic No, it makes great sense and in and you've all mentioned a collaboration and um, I I heard From matai in hell, you know, you had a telephone call about this. I'm going to come back to andy To ask a question on it, but I mean just tell tell the folks that are listening How did that call stark how and what made you think of starting it and and what was matai's reaction? How did that go? Well, I mean these guys can add to it, but it was an early march day So early in the process for those of us in boston. It was typically cold gray and snowy outside It was a sunday morning And a handful of us got on the phone because we were we just were going in all different directions And you know, it was like we all had this epiphany that we're all going through the same thing We're all trying to do something good. Let's let's just work together and you know Something that many of us have said And actually i've had this great opportunity to work with matai for a long time how on and off i've really just You know dive deep in a relationship with david over this COVID epidemic These are really good people and yeah, we're competitors and we want to beat each other when it comes to winning But I think we all share a very common mindset and a common goal and you know The trust and the relationships that we have and we've developed really enabled a lot of this effective partnership Yeah, is that how you felt how? Yeah, I mean, I think yandy said it perfectly. I think I think it was just such a shock to the system to think about not only how big the problem was but how much responsibility we all had for solving it and It was sort of a unique situation where you feel Um time pressure Content knowledge pressure just a lot of different pressures. And so I think all of our instincts were Who can we call to help and so we leverage our relationships, you know, we I you know feel very close to matai and andy and to many people in the pharma industry It was just wonderful To just bounce ideas off the tie and you know, he said, you know It's funny you call because I just talked to indy earlier. We're setting up a group And so everybody was pretty much thinking the same thing. I'll tell one quick story just Leave your question that that I think was um Emblematic really of everything everyone's doing but one day I got a text from Roger perlmutter saying do you have a few minutes to talk about coven? and so of course I said sure and I don't know within a half an hour we were talking and he he He let me know about a screen they were doing On a halfway to class that they were excited about and they had some control molecules And one of them was a control molecule that we own that we had actually been in clinical development And he said, you know the funny thing about the screen is our molecule didn't really work that well But yours did and I just wanted to call you to let you know We'd be happy to send you the data do anything to help we hope That it solves the pandemic And just wanted you to know and the idea that you know companies are And heads of RDS are talking about sharing data and doing things without any Lawyers and IP and whatnot. It was just it was a very we actually had known about some of this data But but the point was really the collaborative spirits and I think everybody that I've interacted with in the farm industry Has has had since this broke out, you know on the data if I don't if you don't mind On the data sharing piece we had Anne Heatherington in the previous talk talk I wonder matai does matai has really stepped up in a space that's hyper competitive Theans helped to really drive information and data sharing. I don't know matai if you want to comment a bit about that I think it's been quite impressive up Yeah, you're pulling a chris vbacher. You're on me. All right at least if you're paying on the vaccine side even if you Are using different platforms if you're using a similar protein there is a logic for A commonality of an immune correlate for efficacy So there's no question that it's useful to share data and maybe share other information And at first it was challenging, but it's in recent times It's been quite satisfying to see the companies step forward talk about their experiences with the regulators choice event points You know where they are with how they think about recruiting and designing the study It's been wonderful. And then just you know the the comment maybe that that expands on that a bit is There's no question the collaboration and sharing is a good thing across the industry Where you are fiercely competitive about intellectual property, but there's so much that's not about that that is Useful to to get together So the silver lining of covet 19 of the group that Andy catalyzed and put together and lead Um, all of this is that it's created trust relationships And deepened trust relationships. They already existed among a whole bunch of people Yes Big pharmaceutical companies, but a whole bunch of people whoever was interested in in contributing And that bodes well for the future. There's no doubt Another important piece here martin was the group getting together. There was the real-time data sharing aspect, but also Asking the question. Where are the gaps? What's not being done that this group can potentially uniquely Contribute and then as the group gel that also became a vehicle to interact with government, for example the FDA Now operation work speed NIH active And provide a sort of unified voice Point of view all of which was in service of efficiency And i'm certain that the the relationships have been which have been quite intense I mean, I've spent more time with the folks on this panel almost in my own family Over the last four or five months And you know those relate relationships will be enduring Absolutely, and it doesn't surprise me, you know knowing you guys and knowing the way you work I think there's sometimes a thought abroad which is you're all at each other's throats But I've known people in this industry for long enough and particularly a time like this Where folks really get together and solve the problem and you know, roy spoke about the power of science coming through here So i'm delighted Now now that i'm in a, you know, tiny company we Andy and I had a discussion A few weeks ago and I was saying, you know, we've partly missed a beat in a small organization We're pre-clinical all our work is done outside with collaborators and and they kept their doors open So for us, it's been, you know, just more of the same that for you folks, you know running large Development programs, but what's that been like? matai I think the on the development side it varied depending on the part of the portfolio or the the indication that we're talking about It I mean overall we we we treated each Trial as special, you know, what were we able to do to keep it going to keep it moving? Um, and in some areas we were remarkably successful. In fact in some areas, I would even say we may have sped it up Um, and then in other areas where for example in our pulmonary hypertension space They're all pulmonologists and the pulmonologists were the first to be pulled off whatever they were doing and asked to treat the pulmonary disease of COVID-19 So they were that was tough and certain very potent immunomodulators that had unknown Activity really those were a little tough But if you look at oncology, if you look across the board in our cardiovascular space, it's it was remarkably good We used tactics. We did special things. Um, you know, we got drug to patients in non-standard ways We did some remote monitoring for example with our antidepressant, which is supposed to be delivered Uh in an in an observed facility We had rapid regulatory review of a of an ipad-based home monitoring system Which worked out and kept people on drug So we pulled out all the stops and in most places we were successful in keeping things Very good. Well, uh gsk. How? You know, I don't have much to add from what metai said. I think um, you you basically go looking for creative solutions um And some of those solutions are going to be solutions that stay with us Because they're creative and useful and happen to solve this problem But probably should have been in place before just the incentives and the push and whatnot wasn't there So I think you know keeping the patient at the center of the clinical trial is always something all of us are trying to do And this just I think further accentuated the need to do that. So Uh, I don't go into any specifics, but but we had many challenges most of what came as well with that creative ideas Like from time mentioned You make a great point about many of these things will stay with us now hal and maybe it's things We should have been doing, you know previously. Um, what about uh, and gen david Yeah, you know to me that I think of this in a couple big categories first is the virtualization Of clinical trials, you know, whether it is a telemedicine visit Directly shipping investigational product to a patient instead of having them come to a site Uh virtual investigator meetings site initiations centralized monitoring All of those things were trends pre kovid kovid Greatly accelerated it and I think we've learned that many of these things Each one of which might be fairly trivial But when you put a dozen or 20 of them together it can have a dramatic impact on Trial efficiency, uh, and I would also say in the democratization Of trials a patient whose 300 miles from an investigational site may have an insurmountable barrier To participating if they have to do weekly Trial visits if they only have to come every couple months That may be a game changer. And so what do we retain from that? That really allows a broadened reach going forward shame on us if we don't permit that to happen Yes, indeed. I'm going to move us off. Um, kovid But I want to give you a chance Andy just to just to close out on the the impact that it's had in your organization Then we're going to move on I mean, I won't there's not much to add. We're all going to say the same thing I'll just emphasize and double click on what David just said Which is we now have an opportunity to use this as a forcing function To change how we behave and to improve efficiency Would we just get so caught up in Inertia and legacy ways of working and so much of what we do is just intrinsically inefficient And I think this is a chance. We just can't can't lose the opportunity Absolutely, so I'm going to move us off and kovid Thank you for that really great insights in terms of the fighting the virus but also the impact on your organization I'd like to move to my Question that I ask every year and it's I must confess it's my favorite question I usually get some really good discussion and it's about what are you most excited about? You know in your organization just just take the pandemic aside for the moment. What's going on? Medalities technologies, I think we may even have some questions from the audience So let's go around and ask that and then we'll we'll go to the audience. So how at GSK. What are you most excited about? Well, there's quite a few things. I'll I'll just pick one combine a few things into one which is And maybe I'm Reaching to some of you on the choir here But I think that what we find most exciting is to make this to use Andy's words inherently inefficient system more efficient the problem that we're trying to tackle is the 90% failure rate of Potential drugs entering the climate that come out and we tried to simplify that problem to ask What is the of course? There's many drivers for that What is the biggest one and we arrived at being that the targets that we choose are perfect for mice and not very good for humans I mean, I'm oversimplifying and and so we said, okay Well, what are you going to do about that and and we've tried to make the human the model organism? And by that I mean using data to the extent possible from humans to really figure out what What is wrong with the patient with a given disease? What's the pathophysiology process? What are the therefore the targets that need to be modulated and we're basically focused on using human genetics As well as functional genomics I mean, so we now have with 23 me and uk biobank and fin gen and open targets and a couple of the things that we'll announce soon But one of the largest is not the largest human genetics databases, but but 90 plus percent of these GWAS associations sometimes very strong. We don't really know What's in the causal pathway? What's the confounded by things being in ld or what might be? Not in an introgenic region. So you don't even know what gene it's actually affecting You don't really understand exactly what the association what's driving the association and of course without knowing that there's no way to really be effective Developing medicines are finding the target So this whole field of functional genomics to be very specific about what I'm most excited about the ability to do gene gene interaction maps in a mammalian cell A dream that I don't think most people thought would be realized in there any time soon And now we can do this 200 million combination with endophenotypes that that Can be upwards of Quite quite quite the intensive outputs like transcript profiles, etc And when you look at that just to round off the area most excited about human genetics times functional genomics Times this trillions of data points per experiment this this machine learning has the opportunity to deconstruct the Kind of the semantic representation behind all these data points And so we think that really going deep on all three of these human genetics functional genomics machine learning are What can allow us to lucidate targets that are novel and can maybe have a significantly higher probability of success Than the than the industry averages at present I saw hell this week that you opened a center in in london On the on the machine learning side really speaks to that third leg Yeah, we've we've we've tried to collaborate the best and brightest we have Opened up the laboratory for genomics research with its jennifer dow the jonathan weissman center In san francisco. We've got the machine learning group there as well as well as the machine learning group now in london And uh, you know, we've got a collaboration with the road to do this variant to function assessment So we're we're trying to hire the best to work with the best and and uh, you know Double down on the whole that whole area of that technology Very good. Good. Good luck. Matai. What about g and g So howl and I are thinking similarly in some ways here But uh, you know, there there is lots of opportunity for the use of data and analytics in what we're doing there's um On the let me try a different kind of example on the diagnostic side So when we talk about diseases that are really well known to most physicians coming out of their training Uh, they don't have a hard time identifying patients as soon as you dip below a certain frequency of occurrence Where it's unusual to see a patient with that condition or the condition just may not be known at all It becomes extremely challenging. And so, um, as an example We have a we treat a disease with a couple really effective agents for pulmonary arterial hypertension And um, but these patients are sometimes never diagnosed or if they are diagnosed or diagnosed in years Not straight away So what we're what we're working on is a use of machine learning methods That are applied on top of data that are collected routinely Like for example, my cardiogram might be an example where in there is intuitively the data Uh to to sort of diagnose something to do with, you know, your your how you're pumping blood And maybe there's a clue in there that you do in fact have some pulmonary arterial hypertension Uh, so it's not a done deal But it's an example of something where you can apply machine learning to immediately Do something that people can't naturally do like they they're not these are very subtle reads. Sometimes they're not Possible even with human eyes But with these machine learning methods, especially applied to images and video There's there's the possibility of pulling back interpretation. That's very difficult for humans So in addition to the great work some of what hal referred to on choosing targets that make sense The connection between a specific mechanism or path and and a disease pathophysiology It's really important to say that machine learning methods Data science broadly can be applied everywhere And we're we're fortunate to have a great group under a great leader. Najat was on the previous panel Very pragmatic approach. We've taken under her leadership to build up the data science capability and application with engines Marvelous. Thank you, Matai. Andy. I must confess I've lost track of the collaborations the joint ventures and Just seems to be hundreds that you've done since since taking over it to Kada What's exciting? All right, I mean not to restate what Matai and how I've already agreed to I'll just add a third leg to this idea of Human genetics identifying strong targets functional genomics allowing us to understand mechanism and then this realistic Toolbox of modalities that allow us to go after any disease target You know if we go back to when many of us joined the industry around 2000 you had small molecules than mice And we succeeded. We cured every disease in mouse with small molecules. You congratulate But unfortunately we found a lot of that just didn't translate And now we're at this juncture where realistically we could say that by the end of this century We we could have cured every known disease or managed every known disease That I don't think that's a fallacy. I think that that's a reality But when I look at then the challenge and then this is exciting is what's going to get in our way You know still stochastic It's not as as as quantitative and precise as the physical sciences, but we can overcome that with mass action What's going to get in our way ourselves? Our behaviors our inability to control the landscape around us You know, we talked a little bit earlier today about some of the shenanigans going on at fda and some of the politicization of some of That could be brutal and and delay Significantly we've seen same discussion earlier around pricing and not just pricing but health care costs If we can't get in front of that and own that someone else will and that has a chance to really disrupt our forward momentum No, great great points. And you're very kind Andy about when we joined the industry How we'll enjoy this I I joined beach and pharmaceuticals in 1979 So you're a little off with me. I'm sure the others are more content What modality we're using then leeches Believe or not infectious diseases Earth samples Greening, you know dead cells live cells and we made some medicines. It was amazing David a question for you on what's exciting Then I'm going to go to the side to look at some questions from the audience in the last few minutes. David Sure. Well, let me pick up on the couple threads that The other panelists have started One is the notion of data and I would postulate and we have a core belief that this is the century of human data From genomics and other omic technologies. So we're investing heavily in proteomics, for example through integration of these technologies and clinical trials to Real-world data data from the marketplace at the other end, but it all has one thing in common It's human data In a way that we haven't had before how do we use that exploit that to understand and get a much more foundational understanding of human disease So to us we are trying to actually build an integrated unit which remit is human data Now what that will do is of course give you new targets And one of our challenges now is that about 80 percent of the targets of interest aren't currently tractable with current modalities as Andy is pointing out and so You know, we have a very strong belief that that one part of the future will be multi-specific Or multifunctional. I'm reflecting on hearing Roy Earlier today. He really was the avatar of the paradigm Single target single drug often in an enzyme that's inhibited or agonized By a drug Well, how do we broach transcription factors other? hard to reach targets Well targeted protein degradation Other mechanisms where you have a binder and you have an effector and you bring those together to create new biology I think is just going to open up that universe of 80 percent of targets that we currently can't reach So to me that's tremendously excited exciting that gets us towards Andy's utopia Where we can go a long way against a large swath of human diseases Terrific, let's go quickly to side and see if we can just get one question in from the audience It might not be to every speaker of its side Yeah, thanks. Yeah, so I think expanding on what What what Hallis started and and then David talked about I mean, I want to use Oncology as an example for emerging trends And as David you just pointed out, I mean Look at Ketrude becoming the first drug approved for a tumor agnostic indication like msi high And then other histology agnostic approvals followed for multiple tumor types You've got twerk TRK fusion by locks of bear the rk fusion ross one by genetic tech And all the excitement around the development programs focused on k-ras mutations and other oncogenes I mean, I know I would love to hear from the panel. Do you see this trend emerging? In oncology to treat tumors based on common biomarker expressions rather than site of origin And if if that's the case, what are the implications for future oncology drug development strategies? You know for the right drug to the right patient I think we're pretty much out of time now in in truth It's such a great question that could derive many answers But do any of the panelists just want to jump in quickly is there anything that you would like to say? It's a great point side Thank you Go ahead. How I was just going to say I think we're all probably vehemently agreeing that this tumor agnostic approach is going to be Not only present in the future, but probably even more robust where we use not just genomic biomarkers But lots of different ways of deconstructing the pathophysiology of the tumor itself and thinking about ways of attacking it My own personal belief is that that's going to become even more sophisticated, particularly with gene gene interaction The concept of synthetically valid. So I I know all of you pretty well. I imagine that that probably captures what everybody's thinking This is a place to double down Yeah, absolutely Marvel, thank you side and thank you to the panelists Terrific as always we could we could spend a long time together and normally we do after this and Hang out and you know enjoy each other's company, but really many thanks indeed I think I've got to announce the poll results from the r&d question Then I'm going to pass on to Andy. So do we have a slide on that? Oh, there you go Am I supposed to read that? I'd love to see that on a big screen Can you see it clearly Andy it click on it and then you hit speaker view and that'll that'll light it up I was clicking just nothing was happening in my my fingers have gone numb So covered 19 has led to an unprecedented level of collaboration among stakeholders in the biopharma industry Where do you expect to see the biggest increase in collaborations post pandemic? And the leader about something about a third is discovery pre clinical and then another third on clinical development manufacturing, you know 20 25 percent and then interestingly the lowest the the sharp end of can't get a competition Uh less than six percent on commercialization. So very interesting and and I must say not too surprising With that again, thanks to the panel. I'm going to pass on to Andy now for some concluding remarks. Thank you, Andy