Name: Innovation Forum 2024: Shaping The Next Era Of Healthcare
Uploaded: 2024-03-27
Duration: 24 min 26 s
Description: Yin Ye, Chief Executive Officer And Executive Director Of Bgi Group Alex Zhavoronkov, Founder And Ceo, Insilico Medicine Moderated By Yvonne Xie, New Media Executive Editor, Fortune China

Yin Ye, Chief Executive Officer And Executive Director Of Bgi Group  Alex Zhavoronkov, Founder And Ceo, Insilico Medicine  Moderated By Yvonne Xie, New Media Executive Editor, Fortune China

Innovation Forum 2024: Shaping The Next Era Of Healthcare

Conferences

I, I'm thrilled to have Alex and Y with me today, the work you're doing is really shaping the future of healthcare. And please note missing, we'll be speaking in Chinese and headphones at your seat for you. Um As my, we just noted the fight against COVID-19 is really a worldwide collaborative effort. And now as we are entering the post Pande era, um which area of collaboration in the healthcare sector are most critical Alex. So I think that it's extremely important to collaborate in drug discovery. Uh That is the most critical area for all of us because we can live without a cell phone even. Well. Nowadays, it's difficult but we can live without a better car, we can live without uh you know, flying business class, but we cannot live without cancer, drugs or drugs targeting Alzheimer's or Parkinson's or infectious diseases as we've just learned from the pandemic. So I think the global collaboration in drug discovery is fundamental and of course biology is so complex that we need to work together to solve it. Um So Mr Ying, what, what's your view? Uh I will continue on that and talking more about data actually, we have mentioned that we are now in the era of artificial intelligence. So we will have a lot of data. If we not, we if we don't share the information from like China American and European, then we will enter the silo effect which will cause waste. Actually, when we first identify the genome of human being, scientists have mentioned that we need to share this kind of information uh when we protect the personal privacy. So I believe that such words are right for the present, whether it is for the anti aging research, we really look forward to that it's really critical and in which area can the results of collaboration can be seen most immediately? Alex. So again, I think that uh we need to collaborate on drug discovery, that's where you would be able to see the results of collaboration most immediately. Uh because patients cannot wait and when it comes to aging, all of us are patients and we should be very impatient patients. Uh So uh we probably will see the result of collaboration uh immediately in personalized medicine because uh that's where we can use already existing drugs to repurpose them towards specific diseases that might have not been targeted by those drugs immediately. Uh So that's one of the areas where we need to collaborate. And by the way, China is the world's largest developed uh uh market with 1.4 billion people. More than 1.4 billion people. So having a very good understanding of how different drugs work and different diseases. And so many people can uncover huge potential for us to live longer and better life and very rapidly because we are looking at already existing drugs. And the second wave of course, of course, innovative novel drug discovery. Uh they are China over the past 20 years, has developed the world's largest and the most efficient infrastructure for chemical synthesis, biological uh manufacturing and very rapid testing. So if you are designing anywhere in the world, you might want to come to China and do the testing and uh synthesis and uh you don't need to rebuild this infrastructure. Of course, it's a good idea to have that infrastructure in every country. But I think that right now we need to focus on efficiency and delivering really fast um uh very effective solutions to patients in need who are dying. Mhm um So Mr Yy, what would be your ideas? I think that's still about the uh data, for example, about those rare disease, it seems rare by the name. But if we um take the humanity, all the human beings as a whole, we have, for example, like we have 8 billion people in uh every year, we will have 90 million newborns every year. And the um the incidence rates of the birth defect about 4 4%. So if we can share all this data altogether, we can have uh like more insights for the precision medicine. So we will be a community of shared future. Alice. You mentioned that before the A I develop, develop drugs can really hit the market in silico. Your company still has a long way to go and you are still facing a lot of obstacles. What are these obstacles you're facing? So sure there are several obstacles that everybody is facing. And one of these obstacles is that biology is very complex. Biology is perhaps the most complex area that humans need to be focusing on collaborating and making a lot of bets that are going to fail. So in drug discovery, currently, if you start with a novel mechanism of disease, so you never touched that target before the protein that is driving the disease and you want to take it to the market. The probability of taking it to the market is less than 1%. So it's not about the time or the cost is just the sheer probability of success is low. That's why I like to call it the molecular casino. Uh And in this molecular casino, the wheel is is spinning for 12 years and you actually, every time it touches some number there is a probability of success of failure. So 90% of the drugs that are in human clinical trials that enter human clinical trials fail. That is why uh people who invest in this field are extremely sophisticated and they understand the risks and benefits and the payouts if you've got the drug approved are gigantic because a lot of people start using the drug and the health care system reimburses uh your research and development expenses. Uh So just complexity of human biology and the probability of failure, that's our greatest threat. The second greatest threat is actually currently that something that I don't understand is geopolitical tensions. So previously, I thought that that would be off limits, right? Because we all agree that life is the fundamental human, right? After that, there are no rights after you're dead. So we need to figure out how to collaborate internationally, at least in this area. Um So uh so for Mr Yin, you have already um led the team to uh to, to serve for those rare diseases like a Os, including th uh he is a famous patient of a Os. So when you are helping these people, what are the biggest obstacles ahead of you? Can you introduction of Mr Tale? Maybe uh today, not many people know him in this event. Uh Actually Tale is really famous and when we and actually Tale is a patient of a Os and you may, when we talk about a Os, you may think of ice bucket challenge or Stephen Hawking. So during the past several cents centuries, we do not have a cure, direct cure for L OS. And at the beginning of this year, Tale had a conversation with Ban Ki Moon and as a patient, he has, he he he really advocates a big um research over this rare disease. So I think for rare diseases, the most important thing is that we need to acknowledge it, we need to know it, we need to under it. So for a rare disease, if there is only for example, like 1000 patients in one country, there are so many countries in the world, the total population of this rare disease will be really big. And also there is a cognitive misunderstanding thinking that there are no rare diseases in their ancestors for eight generations and they will not get them. It's a misunderstanding because no one is perfect on genome. For example, I personally carry seven serious defect genes. The important thing is that my wife cannot carry the same position. Otherwise our offspring will have high probability of getting sick. So the rare disease are actually born by normal people and the normal people is in quotation marks. And currently the cost of treatment is too, is very expensive. And now the gene therapy drugs is really is very, very costly, about 2 million to million. So we have a long way to go to conquer rare diseases. But anyway, we need to start from understanding these rare diseases. Yes, indeed. So the first step is to make all the rare diseases being seen. Thank you, Tyler. Thank you, Stephen. Um There is the first step and always the last mile and what's the last mile in application of A I developed drugs? What what factors influences um its scalability of these drugs. So A I discovered drugs in my opinion, are actually better than the drugs discovered using scientific serendipity. Because intelligence design allows you for multi parameter optimization. Where instead of looking for a needle in a haystack, you create multiple perfect needles and you test those perfect needles in a test setting that is geared towards a very specific disease or a broader spectrum of diseases. So we've demonstrated several times already that we can design molecules with surgical, with nano perfection that are like to become good drugs. So they are yet to be brought to the market. But we sold one for example, for $80 million upfront. So demonstrating that it could become a very solid business model. So it's not just blah, blah, you create a drug using A I uh in like a year, you sell it to a company that is specializing in drug development and you make $80 million upfront. Plus you take milestones and royalties. So we actually think that A I discovered drugs are likely to be better than the drugs that you can find using high throughput screening or from molecular libraries. So and the biggest bottleneck right now for us is actually testing. So very unlike images or that is A I generated where humans can very quickly look at the output and say, oh it's correct. I like it. In drug discovery. You actually need to perform the experiment and first you perform the experiment on animals, then you go humans, you can go more humans, you can go different humans, bigger human population. So that is the bottleneck and that is a very well regulated area. You can, you can basically like Tesla, you can accelerate from 0 to 60 then you move with the speed of traffic because it's very regulated. So we would of course like to figure out ways to go around that. And on that point, if I do have a little bit of time, I'd like to expand on als. So one really cool example that I can bring of our global collaboration. Recently 2022 in March, I go to a conference at Johns Hopkins called Answer A LS Consortium. It's the head of Neuroscience Chair of Neuroscience at Harvard MG eight JH Merit. If Merit is listening, uh the B and Jeffrey Rothstein for counterpart at Hopkins, they created a massive data set of A LS patient data openly available. You can apply for it and take the data. So we applied for that right after the conference, took this data for um cells collected from IPs for LS patient and reprogrammed into human neurons. So it's a very, very good data source. We combine it with many other sources of data, run it on our A I platform identified new targets and possible drugs to repurpose, tested it in Mayo clinic's model of flies, fly neurons in real flies show that some of those targets worked. And then uh we met Dr Lu Bai Bai Lu, very famous Chinese American neuroscientist at Tsinghua University who is pioneering multiple neurodegenerative diseases including A LS. He found one of those targets to be very promising to one of those already existing drugs and started a phase two clinical trials in China. Um Principal investigator in this, the target was to enroll, I think 60 patients in the US. It would take you years to enroll so many in China A S is not a rare disease, they are fully enrolled right now waiting for the outcome. So even if it fails, just the sheer speed of that exercise and global collaboration can take this process of trial and error that fails most of the time anyway to the next level. That's why global collaboration is extremely important and that's what we need to do more and more. So collaboration is definitely our keyword for this discussion and the keyword for this whole summit is A I. So uh Mr Yin has BG I already deployed generative artificial intelligence. What role can I play? Actually, we're doing it. We are actually helping doctors and even public quickly understand the language of life. Uh language like languages including English and Chinese or languages used by human beings. Um And English is a um unilateral, it is actually a one dimensional language. And for Chinese, I think it's a two dimensional language because we have characters. But for DNA, it's a three dimensional language and it has a or it is a four dimensional one because it has a space. So I think we can use N LP or A IGC to do development and we can also use uh and also for A I, there are three pillars uh algorithms commuting power and data. And now I think we still do not have enough data to uh to support the development. And for the, for the first, for the first human genome, we spent $3.8 billion to produce the first human genome. But now it will only cost us $99 and we can have a multi uh we, we can have a multigene, we can combine all of those different materials altogether. And nowadays, um nowadays, we have the evidence based medicine, they only accept inductions. That's so, so even if we can use A I to predict things, we, we should still prove that in the real world. So we need to have a digital twin here. So we will be very time costly. You need to have a lot of time, spend a lot of time to prove that. Um Alex I noticed you recently published a study in nature biotechnology. You called it the most important paper of your life. What's the big deal of this paper? Can you explain. So sure we publish a paper every 10 days in in good journals. But that one was very, very special because it combines uh the story of drug discovery from the very, very early stages where you try to formulate the disease, identify protein targets using A I uh then generate novel chemistry, novel chemical matter using A I. Then we performed many, many, many experiments in vitro in a Petri dish, done many, many, many experiments in multiple different animal models in mice and rats and dog showing that the drug is effective in lung fibrosis and multiple other fibrotic diseases. So we perturbed that type of biology. We also did something really special called hallmarks of aging assessment for the first time. So we looked for protein targets that are implicated on aging and disease at the same time because my philosophy is that if the drug is powerful enough to work in some hallmarks of aging, it's likely to work in multiple age related diseases. And then we took in of phase zero human clinical trials just to see if it's nontoxic for, for humans, did that check mark and then we did two phase one human clinical trials demonstrating the safety. Now, this drug is in phase two. And in this paper, we are outlined this entire journey, the probability of success of that of having this uh go go to phase two is less than 1%. And I bet my life on it so it can still fail. But that milestone now published in Nature Biotic, we also created a resource even for peer reviews. When people were reviewing the paper, we provided a form of charge GP T where you can talk to the paper and ask it questions and many other in silo papers and get the response because it's such a multidisciplinary paper. When medicinal chemists look at it, they ask medicinal chemists questions but they overlook the A I part and the clinical part and the biology part. So it had many, many reviewers with many, many questions. Nowadays, if you go to five papers in silico gpt.com, you can actually talk to this paper and ask it questions today. It was accessed more than 42,000 times in the first two weeks were very popular paper. So do look at it and the gene is called tick that we've targeted. I think it is going to work in many diseases and hopefully aging recently, the University of Copenhagen showed that mice that lack those genes are they when they are put on high fat, high sugar diet. So they basically live their live on cheese cake for their life. They don't get fat and they actually muscle. That's what you want, right? So serendipity since you've mentioned about agent several times, I know that the ultimate goal of your company is to extend human life lifespan. How so sure, I think it should be the goal for everybody here because it's the most altruistic and egoistic goal. At the same time, I think that all of us are going to age and die at some point in time, it would be great to die younger later. And the way we are approaching it is in a very traditional way within the pharmaceutical, traditional paradigm where we discover drugs, put them on the market and sell and then use the proceeds to reinvest and research to understand the fundamental human biology. And actually, I think it's a shared goal even with BG because from what I know one of the founders, Wang Jian, I made a statement that he wants every BG I employee to live to 100 or 100 and 20 right? That he was criticized for making that claim, but it's not a bad claim to make, right? And it's not a bad goal to set for yourself. Because one thing that I want all of you together from this conference, from my talk at least is that the longer you expect live, the younger you are going to behave and by extension, the longer you're going to live. So just psychologically, you should already prepare for a much longer lifespans. And if you think that technology is going to stand still for the next 20 years, you don't understand exponentials because what we have done on the paper is pretty revolutionary. But look at some of our other papers for many other drugs that we are developing. Uh And uh they have like really, really good fundamental understanding of aging um come to our conference in Copenhagen. Uh And um I think that uh you should already set your longevity expectations way beyond what you have planned, what you are planning, what you uh parents live to or what the country average is. It's a very good thing and it's also a very noble goal. So if we can agree on one thing, uh you know, in the US and China is that we need to live longer and healthier. Maybe the next generation of us can be immortal sometimes. Um Anyway, uh so uh my race for Mr Yin is that for extending human life, how uh gene editing can play a role? Actually, back in 1990 the average life is VT is 30 years old. However, today it's already at 73 years old and in 1949 the life expectancy in China is 39 years old and now it's 81 years. So, and I have a good news for the women here. In any era, almost every female will have four or five years longer of life expectancy compared that with men. So I think extending life expectancy is very important and it relies on the nutrition, better medical conditions and the development of some drugs. All of that extend our life expectancy if we can intervene with tumor earlier, if we can treat the new degradation earlier, if we can treat Alzheimer's, I believe that all of you can extend your life expectancy to 90 or 100 years old. Uh However, you need to have saved enough money. Right? Because the last 10 years of human life is quite expensive. And actually we have established a uh we have a, we, we, we think, we think that now we can frozen the gene. Uh We don't talk about the ethics for now. I think we don't see any limits, maybe live for like 200 years old. That's possible. And actually, personally, I, I I'm very clear that I personally oppose personal immortality because if there is no death, every birth is a tragedy, right? It seems that the resources is quite limited. And second, I also talked about the immortality. I think that I also make a joke that the first immortal person might be physical, be killed by the second immortal person. So to some extent, we should pursue a kind of spiritual immortality, a kind of cultural immortality rather than the immortality of individuals. So that we can, we can say that we can save some resources for the next generation. Thank you for your discussion. Thank you. Thank you. Thank you.