#455: Is animal testing in drug development on its way out?

0:02We all do want to remove away from animal testing, but we also realize it's not an overnight thing. We can reduce study animals by 30 to 50 percent, which is a step in the right direction. Through computer algorithms, machine learning algorithms, you can then do the comparisons with the treated animal. How would I as a patient get to a stage of trusting it? How would a regulator get to the stage of trusting it? Over 20 studies. In every single time, the conclusion did not change.

0:34Hey, really. Delighted to be joined this week by Steve Ballera, Chief Scientific Officer for Safety Assessment and Toxicology at Charles River Laboratories, which is one of the largest CROs in the world. And a company that almost every new drug will pass through on its way from the lab to human trials. steve spent nearly two decades at charles river before that decade in pharma advisor and bristol miles squibb toxicologist by training background in cancer research and is now responsible for scientific conduct of safety studies that determine whether or not a molecule is safe enough to ever reach a patient he's also at the center of one of the biggest shifts happening in drug development which we're going to talk about which is the move away from animal testing towards virtual and human relevant models.

1:23It's funny, I've got Claude to do this intro, and it now says, which makes him a fascinating person to talk to because he's helping lead that change from inside a company built in part on the very thing it's now trying to replace. What an AI intro that is. Steve, welcome to Health Tech Podcast. How are you doing, sir? Good. Thanks, James. Pleasure to be here. Glad to talk about science. It's going to be fun. It's going to be fun. Whereabouts are you speaking to us from today, Steve? Where are you based? So I'm based actually in Reno, Nevada. Oh, nice. In the desert. Yeah, I'm an East Coaster by the United States.

2:01By birth and had an opportunity presented itself with Charles River, I moved out to Reno, Nevada. And meanwhile, my wife and I still say, can you believe we live in Reno, Nevada? Who in the wildest dreams would ever think of that? that might be your home but where's your heart then where are you uh where are you where are you brought up with my career i've lived in so many different places we moved on average every four to five years this is the longest i've been anywhere so we've been out here in reno for about 16 years now which is very bizarre and like like most like most people who are fascinated by television you know i've got friends and family up in montana and i always tell my wife we should moved to montana and she's like be great if you go you can go i'm staying here so oh no but i do like it out here the weather is beautiful i'm if if i could show you the outside my front yard sits right at the base of the sierra nevadas the ski areas are 12 minutes away so make a big plug for reno if you're a big skier you can work at charles river and reno you're nine minutes from mount rose ski area glorious glorious what a plug for hiring that is so Steve I'm looking forward to this um we obviously have people from all across the health tech space and I use health tech very broadly there and so um to speak to someone in pharma in a CRO that's spearheading something relatively new in that industry it's going to be quite an interesting chat but before we get into of that stuff.

3:34I want to learn a bit more about you. So where does, where, where's your, where, where is your career taking you? How did you begin? How did you rise the ranks? I mean, it's a heck of a title you've now got and, and, uh, you know, in a heck of an industry. Um, but what does that, what does that ascent look like? How does one, how does one become Steve Bellera? My favorite subject. Um, so yeah, I've had a very interesting career. Like most scientists did my undergraduate. So I did my, I'm actually originally from just outside of Buffalo, New York, went to school at a small private Catholic college for Canisius college, did my graduate work at the university of Connecticut.

4:12So if you're a big basketball fan, college basketball fan, UConn is very well known for both men's and women's basketball, where I got my PA, my master's and my PhD in, um, I'm actually a, I'm actually a protein biochemist by training. And we shared a grant with a toxicology group that was working on acetaminophen. So I know more about acetaminophen than probably most people would care to know about. Um, and I did speak English, by the way, I'm just going to throw that in there. Okay. Yes. The paracetamol. Uh, so when we shared this grant, I did sit in on a toxicology course and, um, I was like, toxicology, this is the most boring discipline ever.

4:57Who would want to be a toxicologist? Funny how fate has it now that I part of the, one of the world's largest toxicology organizations on the planet. Um, fate has a funny way of laughing at you. So, so I did my graduate work on acetaminophen, looking at proteins that bound to acetaminophen. And then I did my postdoctoral fellow at the University of Wisconsin, actually for the McArdle Laboratory for Cancer Research, which has the privilege of having a number of Nobel laureates there. So I was in very fine company working in Dr. Henry Pito's laboratory. So I did my postdoc there and then I was hired into Park Davis Pharmaceuticals.

5:38So I was brought in as an investigative toxicologist. Basically, I was brought in to work on solving mechanisms of toxicity. So as a drug goes through development, you may run into some kind of issue with it. My laboratory was set up to be the investigative to try to figure out what that mechanism was. Did it apply to humans? Did it just apply to animal models to see if you could save the compound, save the molecule, or find a better backup. I also was voluntold to be the head of genetic toxicology by our department head at the time. And so I became the head of genetic toxicology as well. So I became a quick study and he was running GeneTox for that organization.

6:22And that continued through the Pfizer acquisition. So Park Davis, Warren Lambert became part of Pfizer and then was recruited over to Bristol-Myers Squibb, where I got back into general toxicology and kind of back into the in vivo side of the business. And throughout both companies, I've sat in mostly small molecule drug development teams. And really kind of what guided me always has been kind of the, I'll say, nature abhors a vacuum. So I've volunteered to do a lot of different things, which I got the attention of Charles River and was recruited to Charles River and was working in the Shrewsbury facility that was opened in 2006, seven area time where they opened the new facility there and was head of talks there as well as pretty much everything because it was a new facility.

7:11Talks, lab sciences, things of that nature all fell under my purview. And when the economy tanked in 2009, 2010, the company decided to close that facility and was trying to figure out my next step. And funny story, my boss at Charles River, we were in this weird matrix organization, was my boss from Bristol Myers Squibb. He's like, hey, I could use you in Reno. And from our earlier conversation, I was like, where? Where is that? So I moved out to Reno as the head of Toxin Reno. And through time, I rose to become the global lead for toxicology. And then when he retired, I became now the chief scientific officer for discovery and safety assessment here at Charles River.

8:00So it's been a very twisty career, a lot of different sites, a lot of different locations. But I think it's, I think it's, my career has done well for me. It has. And I think there's a lot to learn for people listening on that as well. I have a lot of friends that are obviously doctors. I went to medical school and a lot of people are leaving clinical medicine for various reasons, to be fair. But one of the major places that they're going is pharma. And there are many clinicians having great careers in pharma. um I actually went for dinner with um my friend uh Johnny Matthews you might be listening uh the other day and he's you know having a great career in various different pharma companies and working his way up and I guess that's what I want to ask you which is you've had a career where you have you've you've consistently gone upwards and had more and more responsibility and you've I assume looked after more and more people in departments and um and had more and more impact frankly with what you've been able to achieve what what what was it about you or what you wanted or how you went about it that that kept getting you promoted quite frankly what what's your secret what would you tell people what do you expect of people that you manage now or that you mentor now like what what is the key to it interestingly enough I don't manage anybody right now.

9:29Interesting. In most of my career in Charles River, I have not managed people. As a head of talks, of course I did. I had my own department. But as I've risen to global talks and now the CSO, I work in... Everyone makes fun of me because I work... We joke that I work in the matrix. I live in the matrix. Time. Because of our sites across the world, I travel a lot. Times don't mean nothing to me anymore. I'm kind of on call 24. I don't have any direct reports, but I, so because of that, you have to develop a whole different skill set. You have to develop really, how do you make change? How do you have impact without having direct report lines?

10:09So, you know, leading people through influence. So it's a very different set of skills. And I've had to hone that over the years to kind of figure out how do you get things done when you can't just tell people because you report to me, you need to do this. So it's a very, it's been it's a different like i said different skill set so can i ask you about that then how to i mean the obvious question is how do you lead people through influence when you can't literally tell them what to do i think a lot of it is one tool that i've found is asking questions okay you know you never throw anybody under the bus but you can ask them questions and ask them to kind of put pride the rationale and they may say aha you're right i didn't really think of that maybe I ought to consider X, Y, and Z.

10:55So it is a very different way of working. How does one get the attention of those above them in order to keep getting the promotion, getting higher up in the organization, being able to make more impact? Is it simply doing the job? Is it more than that? Again, nature abhors a vacuum. It's volunteering. It's delivering. So for those that, I guess, you know, don't live in preclinical, what is a chief scientific officer for safety assessment? What does a typical day, week, month look like? What are the responsibilities? It sounds like multiple countries, multiple time zones from a very practical element.

11:35What is the job? it's kind of a jack of all trades job in the sense of, and again, I've been allowed to kind of create my own role. And, and, and one of the things that attracted me and it's kept me in the contract world, as opposed to pharma is every day is different. Every day brings new challenges. My job, I kind of divide into like three buckets. One is definitely looking at our division strategy and science and being defender of the faith of science, saying, well, that's a great business thing, but scientifically that may not be the best direction. And as Charles River is a very scientific animal welfare-based company, is making sure that I'm kind of the conscience of that science and it's not everything is a business decision.

12:26It's just reminding people of that. So that's one part is the strategy, the science, where do we go? What are the technologies we need to bring in? What scientific strengths do we need to be current? Those things. And then that's one bucket. And the other bucket would be is really I liaison with a lot of our large global pharma clients, figuring out them. How do I help them? How do I help them get their drugs to patients faster? That's a lot of what I do on a daily basis. Then, of course, I work internally closely with operations about harmonization of processes, optimization of processes, making sure that we have captured the best data, have low error rates, and bring the best science and data to our clients.

13:13So it's kind of an everyday is some combination of that. I love that. Sorry, I'm just finishing writing notes here because there's lots I want to ask you about here. Firstly, the first bucket, essentially defending the science in the organization. for a scientist that is that must be a wonderful job that a wonderful part of the job because actually i think so many people that i talk to on here that are scientists have to very much battle with the commercialization side of things and of course you'll have an appreciation for that and of course you're optimizing operations in part for commercial reason because you know the organization can have more impact and that sort of stuff but i guess to be able to go to work and actually defend the science is a really nice part of the job that must be very enjoyable for someone with a background like yours and i also have the unique perspective of not only being in the contract world now but i was a client i used to go to cro's as a client running studies so that is a unique perspective and you mentioned about your friends and people you know leaving the clinical space to go off and do you know being rather than a clinician going to others so going back you know 20 years now, leaving pharma and going to a contract world was kind of unheard of.

14:26And we're seeing more and more people looking at that now. At one point here, we had seven or eight of us had done that. And that was just unheard of. Most people left contract worlds and went to pharma. But going the other way, you bring in a unique perspective. I was a client. Is this going to make the client happy? It makes sense for a contract world, but does it make sense for the client and kind of reminding the company that you know we are very customer centric organization and we keep reminding ourselves and every day that our clients come first which brings me to bucket two um looking after those large global pharma clients you know how do you get drugs to patients faster i guess for people listening that perhaps might not know the complete ins and outs of a cro like what's what's your remit there like how what what's in your decision making capability there of how you're getting drugs to patients faster.

15:17What can you change? What can you do? What is your role in that? So there's a couple of things in there. I can peel the onion back if you let me. So I remember back - Take your time. We've got a while here, Steve. So feel free to indulge. Feel free to indulge. A number of years ago, Jim Cramer, who has mad money on, I think, CNBC, once called Charles River the arms dealer of the pharmaceutical world, which that's always stuck in my head and it kind of it is true we bring we have the capabilities to help every pharma company both big small what we call the the two men in a molecule companies the virtual companies we have the capabilities to help every company take a drug from basically end of discovery all the way through registration so we bring those capabilities there uh that's one aspect of it.

16:09And the needs of big pharma are different than the needs of the smaller companies. The big companies have a lot of infrastructure, a lot of processes. So how do we work with those processes and to give them what they need? And a lot of it has to do with speed and acceleration and things of that nature. The biotechs on the other end of the spectrum don't have that. So they rely heavily on us for advice, guidance, what's the best thing to do, things of that nature. So we really, we try to cater to what those companies need from us. And again, provide them that service. The third bucket then, operations, processes, the best of the science and the best of the data to the clients.

16:50That's interesting because you've been in this game long enough to see, I imagine, the science change, how we deal with data change. change what what's different now than when you started how how how do things look differently now there's been changes over the in throughout the years one of the things is Charles River has grown by acquisition part of our job is to say okay when a client works at our multiple sites do they get the same experience so there's a lot of harmonization not that anyone's doing anything wrong but the client wants the same level of service the same techniques so we do a lot of that internally and trying to harmonize that.

17:32So that's part of it. Some of the biggest changes are, um, how do you deal with new modalities? So, I mean, back in the day, everything was probably 80, 90 % small molecule and biologics started taking off in the late, probably somewhere in the, you know, mid two thousands. And now biologics, you know, monoclonal antibodies are just a small portion of the biologic area, but now we're also seeing ADCs, which is monoclonal antibodies linked to a warhead to deliver a toxin to, say, a cancer cell. We're seeing, you know, combinations of not just toxins, but bivalent, trivalent antibodies. We're seeing a lot of different, you know, oligonucleotides, all these different modalities.

18:19So you've seen a huge shift in the modalities and that kind of distribution has proportionally changed over time. So that's been a big change. Technology has changed. I mean, I went on a tour of one of our labs recently and I remember pouring a gel. I did a lot of Western blots and Northern blots and DNA sequencing gels. I remember I was in a lab and I remember pouring DNA sequencing gels that were two feet by three feet, took like a gallon of acrylamide, pounding on to make sure you didn't have air bubbles. And now they show me, well, we have this machine now that can do it all in like seconds.

18:58Or the acrylamide gel are like the size of a driver's license now where I had these things that were the size of a piece of eight by 11 piece of a paper. So the technologies have changed a lot over time, smaller, faster, more reliable. So that's another big one. So those are probably the biggest things that I've seen change over time. Toxicology, by nature, has been a very conservative. You know, what works, don't break it. But we are starting to see, with such an effort on animal reduction, we are starting to see clients looking at how do I reduce animal usage? Whether we talk about virtual controls later, about different study designs, micro-sampling as opposed to taking full blood samples.

19:46anything to do to make animal welfare better and that's the other change that you've seen excellent and i definitely want to talk to you about that before i do just a couple more questions so as you as you rightly say you sit at this really um interesting like choke point where every drug will basically pass through from that vantage point where's where's the real bottleneck in getting a molecule from discovery to the clinic? Like what actually is the biggest bottleneck that you're seeing currently? Time. Interesting. I mean, when I say time, I mean, if you're still running your clinical trial, you still need to see, most people are supporting it with a one month study.

20:40So you can't run a one-month study in 21 days. You still got to do it. It's a month. A six-month study is six months. So what we've been focusing on is how do you start studies faster, and then how do you report them out faster, which also leads to a conundrum. I had a conversation with – I've created my own advisory board, and I had dinner with them at the American College of Toxicology last year. and I sat down with them and I said, how do we balance as a CRO? You want us to go faster. You want reports, you want data faster. How do you balance that with quality? So quality is, you know, I use an old advertising line, quality is job one from both, from an execution of study point of view to accuracy and reports.

21:25How do you balance that? If you're taking all our time away because you want it faster and that is a big trend in industry right now. How do you go faster? How do you do that? How do you balance quality with science or quality with speed? I should say. It's a great question. And actually, so let's say. I'm going to say that nobody had an answer for me. Well, I was going to ask you, you know, when in that case, when a company does come to you, they've got limited cash. They've got, you know, a regulatory clock that's ticking away. What is the conversation that you have with them in that scenario?

22:01Because I imagine it happens fairly regularly. so that's where kind of you know well i'll say the word ai starts to play how long did we get 23 minutes 23 minutes without saying i know we can we can talk on ai for the rest of the 23 minutes but ai starts to become part of that conversation and right now i think ai to me i we call it i call it ai assisted decision making we know that ai is not perfect but it definitely can give like our scientists a something to think about a heads up a check on what the data is saying to them have they missed something that maybe the ai picked up of course they have to verify it and things of that nature but i see that's where ai is going to help our scientists save time um ai can help with quality so that people don't have to do a lot of qc that if it's validated and under the regulations, so if I'm familiar with the FDA CFR regulations, you can talk about having validated systems that if the AI and the computer system says that this data is accurate, nobody has to question it anymore.

23:07So that saves time. Combine that with AI and AI-assisted decision-making, you can start to save time. Still can't, you know, AI will help with pathology. You know, maybe they don't have to read every slide. The AI can say, hey, that's normal. Don't look at that. But this, there's something funny happening here. You need to look at that. And when you talk to a pathologist, the number they always tell me is about 80 % of the slides have no change. So then if the AI can say, focus on 20%, that'll give them a time savings. So that's where I think time and quality will become, we won't have to worry about quality because the AI will help with that and it'll give us that speed.

23:48Are you starting to see AI having an impact there? Or is that something that you're looking at and planning for now in the hope that it's going to? Whereabouts are you on that timeline? I think we're at the beginning. We are seeing, I would say in the last couple months, we are starting to see a bit of nervousness around AI. Now we have companies coming to us like, hey, you're not allowed to use AI on our data. And so what we do is we have a conversation with them. It's like, so if I have a reporting tool that can cut your reporting time in half, you don't want me to develop that tool. They're like, oh, no, we don't mean that.

24:24Well, your language you're sending to us says no AI. So we're starting to have these conversations. And I think it's natural. It's a natural response to what things are happening. They don't want us taking their data and putting into some software. I won't name anyone. And then just sending it out to the worldwide net to look at it. But if we can develop tools behind our firewalls to help write reports faster, bring better QC, then they're like, oh, okay, that's okay. But we have to have those conversations. Absolutely right. And that is the sensible, patient conversation, patient as in time conversation, nuanced conversation that definitely needs to happen, which sounds great.

25:06I mean, the devil's in the detail with that, isn't it? Because it is complex. A lot has to do with the speed of adoption. I think we just have to be careful. with what we're doing. And like we're, I was, we've been talking with our legal team a lot and talking with clients a lot about the language we want to put into master service agreements. And the lawyers all excited because this is a brand new field. You know, how often do they get to start thinking about language and rules and laws and things around what you put into contracts? Because this is all new space. We're all inventing it because it's coming to us fast and furious.

25:38And not everyone is comfortable with, you know, AI for administrative tasks, AI for prediction, AI for discovery, AI for this. They're all a little different and they all have different, I'll say rules and regulations around. Definitely. Definitely. And look, you're a scientist by background. So you're of the evidence-based community where you want to know something works. We're also in healthcare. We, you know, the stuff that we're part of eventually reaches patients. And again, exactly. We have internally this alarm that goes off whenever anything new like this comes along and threatens our processes we all you and i and everyone that's got you know clinical or healthcare or scientific backgrounds will go is this safe exactly completely appropriate that things are slower it's completely appropriate that there's more conversations that we we indulge in that nuance to try and find you know the devil in the detail that you know we we want to get to the right answer not necessarily the fastest most efficient answer because of what it might miss because in our industry that that ends up causing a lot more problems than going slightly quicker did cause us a slightly better line on the p &l or whatever the end of the day it's all about patient safety it's all about patient safety and nobody wants to jeopardize that whether it's as you said the p &l is better it's faster if patient safety is jeopardized then we're not doing our job right exactly that exactly that great so i want to talk about what you just mentioned previously actually about animal testing and this obviously being incredibly important to your company and looking at what you're part of specifically you like very important to you as well with virtual control group so could you could you just give us the lie of the land in terms of the industry at the moment and where we are with animal testing of of drugs and medications and things and And I guess where we're moving to in your role in that.

27:40Again, there's a lot of caution, a lot of confusion, a lot of what ifs, what about this? Nobody wants to slow down their drug getting to patients. So people are asking lots of questions and they're not sure what the answers are. I think the prevailing opinion in the industry is, yes, we all do want to remove away from animal tests. We also see that happening. We want that to happen. but we also realize it's not an overnight thing. Yeah. Not everything can be done in a 96-well plate, a 364-well plate. It cannot be done. But where we see things happening is making better decisions to put better compounds through so you don't waste animals that way.

28:27You can help make decisions on investigative work. Is this the right thing? Again, the word NAMS is being used a lot. as your new approach methodologies. Those of us who are kind of in this area and talking about it, it's not new. I mean, companies have been working on in vitro screens for decades to pick better compounds to develop. We've done in vitro testing or assays. As I said about, you know, I was brought in as an investigative toxicologist. We were using in vitro methods to elucidate mechanisms on toxicity. So NAMs in the regulated space, I call that investigative toxicology. The stuff in the preclinical or the pre-IND phase, that's just picking better candidates.

29:16So we've been doing it for years. What we're doing is creating better assays that pick better compounds, are more predictive of the human condition, and again, helping us develop drugs that are better, safer for patients. So that's all new. That's not new. It's an evolution. And that's not a revolution. It's an evolution of what we've been doing forever. So those are the kinds of things. And so the industry is looking for ways to do that. But yes, I think what will be happening for at least the near term, and I don't know, I'm not going to define near term, is we will be doing in conjunction animal testings with some kind of in vitro for a period of time.

29:55And any way we can reduce animals, whether it's using one sex, using one species, weight of evidence arguments, to do that is the way where I see things happening for the near future. And part of that is what we kind of mentioned earlier is virtual control groups. So for those who don't know what we're talking about, it's kind of been done in the clinic for years where you use historical control data to compare to a treated group. So the idea would be is one day to replace all the control animals on a study with data from previously collected data and use those as comparators to your treated groups.

30:37Where the industry is heading in the short term is what we're calling a hybrid model. And there's a reason for this. So a hybrid model. So on a typical four-week toxicology study, we use 10 male, 10 female rats. Where we see the industry going for the near term is reducing each group by half. So five live animals, five virtual animals. And there's a reason for not going to zero. One is in talking with regulatory agencies, experts in the field, that everyone's concerned about, I'll say, genetic drift of the animals in your database. So if you're going to continue to repopulate that database, five animals, 10 animals is better to repopulate, but five will allow us to keep refeeding that database.

31:26So that's one aspect of it. One, having animals on study allows you to have animals in the room as your treated group to control for procedural changes, things that happen while you're actually executing on the study. Also, it happens that there are infections, diseases that come in with animals. having those live controls as comparators, but we can reduce study animals by at least control animals by 30 to 50%, which is a step in the right direction. And then using that virtual data with live animal data, and then through computer algorithms, machine learning algorithms, you can then do the comparisons with the treated animals.

32:15So you still get the same scientific equality the same set of data you're using matched animals to the study data to the antutreated animals so that you can still do these comparisons it's a step in the right direction it is it's a huge step in the right direction i mean you know 30 to 50 percent reduction i mean that's that's huge numbers i mean any anything in the you know drug discovery drug to market processes that would give you a 30 to 50 percent uplift on anything would be you know considered almost impossible so i i yeah i think um it's obviously a very large step in the right direction i'm intrigued though like how do i mean think about it even from even if i put myself in in the in the the shoes of a patient right i i'm i'm a patient this new drug has come through this type of process and let's say I'm part of the trial, I'm part of the human trial now.