Executive Chair, Compugen
A Biopharma Pivot Story With Compugen's Dr. Anat Cohen-Dayag
Dr. Anat Cohen-Dayag transformed her company from a computational biology service provider to a clinical-stage biopharma with multiple cancer immunotherapy candidates, and she did it in a remarkably short period of time. Listen in as she discusses the human resources, financial, and capital strategies she deployed to build a biotech on the back of computational science, and shares her vision for the future of drug discovery.
#businessofbiotech #biopharma #biotech #cytiva
http://cytiva.com/emergingbiotech
Audio version available here: https://www.bioprocessonline.com/doc/...
Subscribe to th...
About Anat Cohen-dayag
Dr. Anat Cohen-Dayag, Executive Chair at Compugen, discussed the company's transformation from a computational biology service provider into a clinical-stage biopharmaceutical company focused on cancer immunotherapy. In a 2021 interview, she described how the company leveraged its status as a public company, using different financial instruments to fund its pipeline before having sufficient early-stage data to pursue public offerings. She noted that Compugen entered into a collaboration with Bayer, licensing to them the rights to develop drugs for one of the targets the company identified. Cohen-Dayag also outlined Compugen's strategy, which includes advancing its clinical-stage pipeline, developing early-stage programs to feed that pipeline, and using its computational engine to sustain its own discoveries. She stated that after presenting clinical data and establishing collaborations with Bristol-Myers Squibb, AstraZeneca, and Bayer, the company conducted another round of offering.
Transcript (52 segments)
M
Matt Piller0:06
Welcome back to the Business of Biotech. I'm your host Matt Piller, and today I'm bringing you a pivot story engineered by Dr. Anat Cohen-Dayag and executed at Compugen, the company she's led for the past decade.
Now, Dr. Cohen-Dayag is not your stereotypical biotech entrepreneur. She didn't follow the MBA, venture capital, pick a sexy and marketable indication du jour path to biopharma leadership. After earning her PhD at Israel's prestigious Weizmann Institute of Life Sciences in '96, she spent the next several years at the bench in various science and R&D capacities at companies including Orgenics, a subsidiary of then-Inverness Medical, which is now part of Abbott, Mindsense Biosystems, and eventually Compugen.
Once there, in just eight years, she climbed from VP of Diagnostic Biomarkers and Drug Testing to President and CEO. And in short order, she transformed the company from a computational biology service provider to a therapeutic discovery and development company focusing on cancer immunotherapy. Dr. Cohen-Dayag, we are thrilled to have you on the show today. And I want to open up our conversation by asking you, just where you started. A pivot from computational biology services to a true biotherapeutic company with a fast developing pipeline has got to require an incredible change management and personnel initiative. So where did you begin?
A
Anat Cohen-Dayag1:38
First, we started with cultivating the corporate culture of learning and making sure that people are going in a mode of self-learning and trying to open and push the limits as much as possible. We also engaged with key opinion leaders in the field, in each of the fields that we needed to address: immuno-oncology, oncology, immunology, business, financial, and learn, learn, learn. Get to know how to ask questions, how to get answers, how to blend all the information, and finally make decisions on our own. Because at the end of the day, the people that are consulting to you are not the people that can make the decisions. So that was one thing. And the second thing was to hire, and we actually hired in a gradual manner as the company progressed.
So first, we needed to make sure that we hire and complement not a drug development expertise, that was the second layer, but first, as a company that is discovering completely new drug targets, completely new, we knew that we needed to unlock the biology. This is usually stuff that is being done by academia, and the first time the drug target gets to therapeutic development in pharma, in biotech, usually is after 5, 10, 15 years of research by academia. We discover completely new drug targets, and there's no biology behind it. We needed to unlock the biology.
So we needed to build an infrastructure to do that, to build the capabilities, and also to enter into collaborations with academia that can help us. And you know, we have a long-term collaboration with Johns Hopkins that was actually working on all the drug targets in the pipeline. So that was one. And then we needed to incorporate drug development expertise, first pre-clinical, and in order to shorten the learning curve or to flatten the learning curve, we went to the Bay Area and hired employees that had experience with the antibody players in the Bay Area. And we then added also clinical capabilities.
So this is, we're now actually a global company with the notion that you hire the expertise where it exists. Mainly as an Israeli company that don't have all the expertise here in Israel, we needed to broaden our search. And today we have, we're global. We have the clinical team in the US and the West Coast, and computational discovery management member in Singapore, business development in Switzerland, and R&D and G&A in Israel. Today, in the COVID-19 situation, I think that many, many companies are actually working in this way, but we started eight years ago.
M
Matt Piller5:13
When you, wait, so you, you know, you talked about sort of the personnel development post-pivot. You make this pivot in 2010. How long did it take from a personnel standpoint? How long did it take before you felt like, okay, we've got the right people in place to be functional as a company that's setting out to develop its own pipeline?
A
Anat Cohen-Dayag5:39
Actually, it was, I have to say that we first needed to know that we have the critical mass of capabilities. That was the trigger to the transition. That was the first trigger to the transition. And then we built the additional layers in a modular manner. So we didn't need to hire everyone for every step that we needed to take. We needed to ensure that we have the right people, and it's not always started by hiring, not at all. We started by engaging key opinion leaders that incorporated the knowledge to the company, to the employees that we already had, and then we went to look for, to hire.
I'll also share with you that, you know, it's not, when you go to hire and you're looking for real talents, you also need to be attractive. And how would you be attractive if you don't have yet the pipeline? So how can you attract drug development talents when you don't have a pipeline or you have a very early stage pipeline and that doesn't fit to the stage that they want to manage? So we needed to do many things with the current personnel that we had in the company and to grow the company to fit to the talents that we want to hire. And then at a certain point in time, it became, you know, the transition was to the point that yes, talents were looking to get enrolled, to compete. Yeah, it didn't start in this way.
M
Matt Piller7:22
Yeah. What, so let's talk about that pipeline for a minute and the early days. What went into the determination around what you would target, what indications you would target, how you would build that pipeline out? What was the thought process from scratch, like from the very beginning?
A
Anat Cohen-Dayag7:40
So we first needed to say, okay, which area and focus we're going to, where we're going to focus. And the idea was to focus on target discovery. And why target discovery? Because target discovery is very complex, even with respect, from my perspective, even with respect to other computational biology areas. It's very complex, it's multifactorial, and we felt that we have the capabilities to deal with it. So it was clear to us that it would be target discovery. We felt that our databases also fit this type of discoveries. And we moved into antibody therapeutics because we thought that that's a shortest path to the clinic, as compared to small molecules, or not to the clinic, to the market.
So the reason to pick immuno-oncology, actually in the beginning we picked immunology and oncology. And I'll tell you what were the reasons. Actually, there were two reasons for that. One, we understood that our databases, the data sources that we have, fits oncology and immunology. And second, by discovering new drug targets, we realized that there is a family of proteins that starts to get attention. These are termed, back then, immune checkpoints. The notion that targeting those proteins, you may interfere between the crosstalk of the cancer cells and the immune cells, where the cancer cells are actually inhibiting the immune system response against the cancer.
And the reason we're focusing on this is because Yervoy and Orencia. So Yervoy is an antibody that is targeting a protein, an immune checkpoint protein, and that's a drug developed by Bristol-Myers Squibb. Back then it was Medarex, but it was acquired by Bristol-Myers Squibb. And Orencia is actually built as a drug for autoimmune diseases based on the same protein. So we actually saw that there is an option for specific drug targets, and in this case, immune checkpoints, to serve for the generation of two different drugs: one in cancer, addressing cancer challenges, and on the other hand, for autoimmune diseases.
And from our perspective, as a small company that wants to make a breakthrough, and we wanted to get the best out of what we discovered, we thought that it would be good for us to enter the field of oncology and immunology with the focus on immune checkpoint proteins, where we could, if we can identify new such proteins, we could come up with two types of drugs, one for oncology and for immunology, from specific drug targets that we'll identify. So that was the reason for us to enter this field when we realized back then, and the immuno-oncology field in 2010 really didn't exist. There was some data published by Medarex, it was the trigger for later, you know, for the acquisition of Medarex by BMS. And we thought that if this is correct, and immune checkpoints may serve as the basis to generate drugs for cancer immunotherapy treatment development, then maybe, just maybe, we can find out additional immune checkpoints and generate new treatment solutions.
And from our perspective, it was the combination of the business opportunity and the capabilities that we have in this field, and the notion that we can come up with two drugs per product.
M
Matt Piller12:16
Back then, were you required to develop any, like your laboratory presence? As you move, as you discover, you know, you set some target indications, you get into discovery, you know, you start to develop a pipeline. Did you invest in lab space and sort of move away from the computers to the wet lab?
A
Anat Cohen-Dayag12:46
Yeah, we didn't need to move away because we wanted to keep them close to us, but we needed to add. So I'll take you step by step. So first, you know, first we needed to come up with novel targets that we want to invest in. And obviously, there are big questions: what is it that you want to focus on? Why it will be differentiated at the end of the day? Which type of drug you'll generate? You know, when we come up today with drugs to the market, no one is looking, oh, the significance of the drug is just because computer discovered it by computer prediction. No, the significance of the drug is that it can compete with other drugs and it could outperform. So we needed to know all this back then. How are we going from step one to step 100?
So focus on specific drug targets and building a laboratory that can test it experimentally. And then when you test it experimentally and you have a hit rate, but only some path, what are you going to do with them? How are you going to reach from a drug target that you validated, initial validation, to generate a drug, to generate an antibody that will serve as a drug that will target this drug target, and how it will fit specific indications? So we needed to build the experimental systems and tools in order to do all these research, first starting by in vitro studies in the cube, then moving to animal studies, what is called in vivo, also testing in human tissue samples to be translational to humans, and then launch programs that have, do all what is called IND-enabling studies, studies that we can submit to the FDA in order for them to approve to us to get to clinical studies, and then the clinical trials.
So for each portion, we needed to add the relevant capabilities, and sometimes it was in-house, so building the laboratory in-house, yes, and we moved to new space. And we needed to make sure that the experimental team know how to speak with the computational team in order to make breakthroughs. And then we've built a site in South San Francisco for pre-clinical and clinical capabilities.
M
Matt Piller15:31
The Business of Biotech is brought to you in partnership with Cytiva. Together, we're committed to helping the leaders of new and emerging biopharma companies navigate the financial, organizational, human resources, and regulatory waters you'll encounter on your way from discovery to the clinic and beyond. Check out a host of useful resources for biotech leaders at Cytiva's Emerging Biotech Accelerator at cytivalifesciences.com/emergingbiotech. That's C-Y-T-I-V-A lifesciences.com/emergingbiotech.
That, is that all in-house or are you outsourcing any of the development and production capabilities?
A
Anat Cohen-Dayag16:15
So actually, we started by having, you know, on the experimental front, on the early stage, and mostly in-house. Some is done under collaboration with academia, as I mentioned, the Hopkins collaboration, long-term collaboration with Professor Drew Pardoll. And we also outsourced some of the studies in the early stage. Pre-clinical, up until two years ago, we had it internally. Two years ago, we decided that we pursue it by outsourcing. So most of the pre-clinical work is done by outsourcing. Clinical, we have the clinical team, but obviously, you know, the studies being done in different clinical sites all over the US.
M
Matt Piller17:22
What, how did you fund some of these capital investments during the pivot? Right, like you've got to do some, we talked about the hiring you have to do, you've got, you know, some lab development. I know we're covering a lot of ground here, we're going from the server room to the lab to funding, but I'm curious how you went about funding the pivot in its entirety. Was there a concerted fundraising effort attached to it?
A
Anat Cohen-Dayag17:56
Yes, so as you said, everything needed to work in parallel. All the challenges were on the plate all the time. So yeah, so looking at from the capital perspective, we were a public company, so we went public already in 2000, and it's a different company obviously. And so we could leverage the fact that we are a public company, and we used different financial instruments to bring money in order to initiate the pipeline first. And then when we had enough proof or early stage data, we could start thinking about public offerings.
So after showing some data and entering into our first pipeline collaboration, when we entered into a collaboration with Bayer, where we licensed to them the rights to develop drugs for one of the targets that we identified, and following this collaboration, we went to the first public offering, a secondary obviously. And after, and recently, after presenting some clinical data and having three collaborations in place with Bristol-Myers Squibb and AstraZeneca, on top of Bayer, three leading pharma companies, we went to another round of offering. So that was more or less the way how we progressed.
M
Matt Piller19:43
Yep. Do you still, so, you know, your background seems fitting to your, your background, like pre-computing seems fitting. I, how do I want to put it? I guess it feels as though you're the right person at the right place at the right time to lead Compugen through this pivot back in 2010 and the ensuing years that it's taken to, you know, mature the company, given your science background, right? Like, do you find now though that you're, like this combination of computational biology and drug development gives you a competitive differentiator moving forward? Like, are you still leaning heavily into that computational biology sort of discovery process that the company was rooted in as you move forward and build out your future pipeline?
A
Anat Cohen-Dayag20:32
Definitely, definitely yes. And I think that we have two key differentiators. And today, by the way, I'm so happy to see in the last two years that there is some more focus on computational discovery in the fields of life sciences, on many fields of life sciences. And I'm completely thrilled and a great believer on this front. So from our perspective, there are two key differentiating factors. And one, the fact that our engine, the computational engine, today is proven in a way. Actually, we have three drug targets that moved from computer prediction to clinical studies through successful pre-clinical studies. And I'm not aware, at least in the field of target discovery, I'm not aware of additional such successes. So this is one that is differentiating us.
And you know, we can discuss what's the status of the company today and why I'm calling it a proof of concept. And the other thing is that we're not only an in silico discovery company. We've integrated drug development expertise, immuno-oncology expertise, into these capabilities, and that matters a lot. And I'll tell you why. And I can say that because I worked for many years as a computational discovery only company, in silico discovery only company. When you start with the end in mind, it matters. You are less naive with respect to your computational output. Obviously, there is hit rate, and you don't know that until you understand the other side of the story. And you can better select, you can better appreciate what is needed and what is not needed when you're an integrated discovery and development company. And also, I think what differentiates us is the drugs that we have today. We have three first-in-class drug candidates that we're very proud of.
M
Matt Piller23:01
Yeah, yeah. You brought up that competitive landscape, and that's a very interesting point given that, you know, earlier in our conversation we talked about the fact that Compugen was ahead of its time when it was in computational biology, and now you find yourself in a pretty heavily populated competitive environment. What do you, you know, I just want to...
A
Anat Cohen-Dayag23:25
I just want to add that is correct. But if you ask me, okay, would you want now to get back to this situation where you're only focusing on discovery collaborations? And no, I think that what we were doing in trying to push forward to drugs, to generate drugs based on our capabilities, I think that's the right path for the company. And acting on the edge that I just described, that's the right path for patients, for shareholder value, and really to fulfill this dream that we all had in 2010 to translate our computational discoveries into drugs.
M
Matt Piller24:18
Yeah, yeah. So let's talk about those drugs. You alluded a little bit to the pipeline. You've got, how many projects do you have in place? How many candidates do you have in place right now in clinical studies?
A
Anat Cohen-Dayag24:32
We have three programs, and behind it, we have additional multiple programs in early stage.
M
Matt Piller24:39
Okay. And so give us a, I guess, a clinical update on the three that are in clinic now.
A
Anat Cohen-Dayag24:46
Sure. Before I focus on the specific clinical data, I just shared that we picked to focus on the discovery of drug targets for cancer immunotherapy and focus on those patients that are not responding to cancer immunotherapy. So I don't know if you're aware of this, but today it's only 20 to 30 percent of the cancer patient population that is responsive to cancer immunotherapy drugs. The rest of the 70 to 80 percent of the population is not responsive to the current marketed drugs. So we decided to focus on this front and try to discover new drug targets, new biological pathways that are actually ongoing in the human body that you need to target in order to help the immune system to fight the cancer.
So with that in mind, we discovered, so now we have three programs in the clinic. All of them are addressing new drug targets for cancer immunotherapy. The leading program is COM701. It is addressing a completely new biological pathway that we discovered that has its own merits. And while we're very excited about it to serve as a cancer immunotherapy treatment, but basically what we discovered is that this pathway is working in parallel and in complement with another pathway that we discovered in 2009. Back then, as you know by now, we weren't a drug development company, and we just sent it to publication. It was published, it was sent to publication back-to-back with another big pharma company. And when we established our own pipeline, we decided not to focus on this specific pathway. We didn't feel that we can compete with the big pharma that is developing a drug for this pathway.
But when we realized that we discovered the new pathway that is actually having a molecular intersection with the old one, we decided that we're going to focus on both. And what we also identified is that the two of them are actually interacting in a way with a third pathway that is the pathway that has a drug in the market, and these few drugs in the market that are addressing the same pathway. This is the PD-1 pathway. I'll just say the PD-1 pathway is the backbone of the industry today. The patients that are responsive to cancer immunotherapy drugs are those patients that are responsive to the PD-1 pathway drugs.
And we came up with additional two pathways. One of them is already addressed by another pharma, and now, you know, with few pharmas. And the third one, the one that is targeted by Compugen, is only addressed by us. So we felt that we bring to a three-pathway story, one of them, a pathway that is very unique and different and can serve as the missing piece in order to treat patients that are not responsive to current drugs. Okay? And based on that pathway story that we identified, we've built all our clinical strategy to test the drugs that we have as a monotherapy, as combination therapy, in dual combination of combining two drugs or three drugs.
That's the basis of our collaboration with Bristol-Myers Squibb. Bristol-Myers Squibb has a marketed drug to target PD-1, which is one of the current blockbusters in the market. They also have a drug targeting the second pathway, and we're using our own drug for the third pathway. And this is how we're actually having a study that is addressing this three-pathway hypothesis that we came up with.
The data that we actually showed up until today, supporting the computational discovery that we've made, is actually very encouraging for COM701. We've already shared data from just very early stages of the study, from the dose escalation stages of the study. We treated patients that are actually exhausted, have exhausted all other treatment solutions. These are patients that are really suffering and they have no other treatments. And we actually had a study that treated what is called all-comers patients from different types of cancer indications. And what we saw is that we had not only that the drug was safe, well tolerated, and safe, but also it showed some anti-tumor activity.
We saw that there is a high rate of response to the drug, and we also found out that this response is durable. And in some cases, we saw a deep response that has to do with specific indications that are not responsive to the current marketed drugs. For example, colorectal cancer and a type of ovarian cancer that are not necessarily responsive to these checkpoint blockers, to these drugs that are in the market now, and actually showed some response to our drug. So these are very preliminary clinical data and based on very small number of patients, but we're now expanding these patient populations and we're testing different combinations of drugs and looking forward to share more data.
I'll just say one more thing. The target that I stated that we published and sent to publication in 2009, back-to-back with another pharma company, already showed by this other pharma company encouraging clinical data. So this target was also showing two things: that first, our prediction was correct, the computational prediction was correct, so that's proof of concept for us. And also that out of the three-pathway story that we came up with, with PD-1 and this specific target that was identified with another pharma company, and our new pathway, that now we have data supportive of all these three pathways, suggesting that our hypothesis is correct. So the PD-1 pathway is validated, the other pathway was already validated by another pharma company, and we have encouraging data for our own third pathway. And from our perspective, that's a big, huge plus in order to think about the future of the strategy that we're taking.
M
Matt Piller32:41
Yeah, yeah. I can, it's interesting, you know, you talk about the fact that there's some good valid data coming from a pharma company, you know, back in 2009 that you sort of developed that IP for, hand it over to them, and now it's kind of come full circle back to you, and that's a good thing. But I can't help but wonder, Dr. Cohen-Dayag, if, you know, given your long history with the company, if you look back at pre-2010, some of the computational biology that back when you're a service provider, you hand it off to other pharma companies, I can't help but wonder if some of those you kind of think to yourself, oh, I wish we could have that one back right now. You know what I mean?
A
Anat Cohen-Dayag33:21
Oh yeah, I totally understand. One correction: we didn't hand over the IP to big pharma on this specific asset, and that is described, they discovered it simultaneously. So you know, that's just to be fair and with them as well. But on the service model, you know, we didn't hand over the capabilities. We were doing in some of the cases, were doing the analysis for them. They were exposed to the capabilities. But I don't think that anymore that this is a situation where we, when we're looking backward and say, oh, maybe they can compete with us with these capabilities that they became aware of. No. And I'll tell you why, for two reasons. First, the world moves so fast in terms of technology, computational capabilities, biological data generation. It is, you need to be updated all the time. You can't go backward and look backward and say, oh, I needed this one. So that's one thing.
The second thing is that we stopped this model, to work through this model, actually in 2004. So you know, we've built for additional six years. We had time to build capabilities in a mix-and-match manner. So now we actually have a suite of tools that we can combine in a mix-and-match manner, and we also add what we need. And so this is just, you know, just to be fair, I think that this field of computational discovery is moving so fast. It's whatever you developed years ago, not necessarily this is the right thing to use today.
M
Matt Piller35:24
Yeah, yeah, very, very good point. So we're running short on time here, Dr. Anat Cohen-Dayag. I'm sorry, Cohen-Dayag. I told you I'd mess that up.
A
Anat Cohen-Dayag35:37
And I told you that it's fine.
M
Matt Piller35:39
Yeah, we can cut that part out. As we run short on time though, share with us what the next big step is for Compugen. What's the next big step on your horizon?
A
Anat Cohen-Dayag35:53
So there are a few. I think that definitely on the front of the clinical studies for COM701, for COM902, which is the second program for us, and we're pushing forward few clinical studies, as I said, monotherapy, combination therapy, doublet, triplet. That's from our perspective in the next few years. We should have multiple drivers on this front. Obviously, you know, things need to work and the clinical data will tell us the story. But definitely on the execution front, we should have data readouts in order to drive the company forward. And also the fact that we have additional early stage programs that should feed our clinical stage pipeline, and hopefully this will translate to additional new programs generating new breakthroughs in the understanding of new biological pathways. And the third one is, you know, relates always to our computational engine. We hope to continue in a sustainable way to feed our own pipeline with our own discoveries, with new programs that will hopefully generate new treatment solutions.
M
Matt Piller37:23
Yeah, well, it's an exciting time and an exciting story. I congratulate you again on that pivot way back when and the success that your company has found in the years since.
A
Anat Cohen-Dayag37:34
Thank you very much, Matt. It was great speaking with you.
M
Matt Piller37:38
Thank you. It's been our pleasure. Thank you for joining us. So that's Dr. Anat Cohen-Dayag. I'm Matt Piller, and this is the Business of Biotech. We're produced by Bioprocess Online in partnership with Cytiva, a company that's committed to nurturing new and emerging biotechs on their journey from concept to clinic and beyond. Tangible proof of that commitment exists at cytiva.com/emergingbiotech. So two tabs open right now: cytiva.com/emergingbiotech and bioprocessonline.com where you can subscribe to my newsletter. If you want to make one more click, make it the fifth star on the right of this podcast landing page. In the meantime, thanks for listening.