Breathing New Life into Toxicology: Human-Relevant Models in Action

Mary McElroy:
There's a big change, and then things rumble on for a while and then something else happens and there's another spurt to stimulate change. So for us, for example, it's the developers who are making very good lung models in vitro that we can actually use. So that standardization, I would hope that our grant, because we have designed it to answer the regulator's concerns, would be a step change. It would be one of those step changes that they will provide data that shows it's reproducible and predictable, and then provide the protocols that enable others to reproduce what we're doing and then be brought into an OECD guideline for greater uptake

Mary Parker:
I'm Mary Parker and welcome to Sounds of Science. In this episode, we are diving into a powerful new alliance between Charles River and MatTech Life Sciences, an exciting collaboration focused on advancing human-relevant non-animal models for toxicology testing. Joining us today is Mary McElroy, Head of Discovery, Toxicology and Pharmacology at Charles River, who will walk us through the science and real-world impact behind this innovative work from one Mary to another. Welcome to Sounds of Science. We're honored to have you on the show.

Mary McElroy:
Oh, thank you very much. Mary P, I'm honored to be here.

MaryParker:
Luckily we don't have to tell each other apart, so we should be just fine.

Mary McElroy: 
Absolutely. Thank you.

Mary Parker: 
So tell us about yourself and your role at Charles River.

Mary McElroy: 
So I'm Head of Discovery Pharmacology and Toxicology, and that's kind of my day-to-day role, but I also have an important role more globally as chair of the 3Rs Advisory Group. So three RS is a general term for replacement, refinement and reduction of animals in safety and discovery, toxicology and general scientific studies. I'm also on the Global AMAP Group led by Julie Frearson, the Alternative Methods Advancement Project, and it's an umbrella term for all things 3Rs and replacing the use of animals in tox studies and discovery work.

Mary Parker:
So what is your educational background? What got you interested in this line of work?

Mary McElroy: 
So I have a varied background, but since my PhD, it's all been lung-related at various institutes around the world, and I became an inhalation toxicologist here in Edinburgh, developing lung-related work. I had the opportunity to grow a business based in lung efficacy models and as my team grew, I acquired an in vitro group that focused a lot on lung models for safety assessment. So along with my prior work and my current group, we've decided to explore and develop the in vitro group.

Mary Parker: 
From my understanding, the safety around inhalation is not just drugs that you might inhale like through an inhaler, but also maybe during the manufacturing process if the drug becomes airborne, they need to make sure that that would be safe as well. Correct?

Mary McElroy: 
Absolutely. If you think about it, we breathe in a lot of chemicals on a daily basis, whether they're environmental, but anything we spray or if we're in a work environment but something becomes distributed in the air, it'll come into our lungs. So we want to be able to breathe in compounds that are nontoxic that we can easily breathe out and not cause any adverse effects.

Mary Parker: 
So along those lines, could you provide an overview of the collaboration between Charles River and MatTech life sciences?

Mary McElroy: 
Yeah, so this collaboration came about because we were approached by the American Chemical Council and they are a stakeholder group for about 190 chemicals in the us and they as part of the long-range initiative for the American Chemical Council, they wanted to investigate and build up alternative in vitro models for inhaled chemical assessment. In regards, they wanted researchers to provide evidence to build confidence and safety of these alternative methods for chemical inhalation safety assessment. And the collaboration with MatTech came about for two reasons. Firstly, because the principal scientist and my team, Joanne Wallace had worked on and off with MatTech over the years because they're one of two suppliers that provide very high-quality lung in vitro models. That was a key. There was a relationship there already. And secondly, they were US-based and we were applying for US-based grants. We thought that would be strategically important.

Mary Parker: 
And they are lung-like tissues that can help mimic the actual human inhalation experience, so to speak. So how does this initiative, this partnership, align with Charles River's commitment to new approach methodologies or NAMs and the 3Rs principles?

Mary McElroy: 
So we've been working in a commercial environment with these models for probably 10 years because chemicals or perfumes that are cosmetics can't be tested on animals. So if you were looking for the safety and assessment of an ingredient that's in a cosmetic product, it would've been done in a model such as the model we get from MatTech. So there's been background. We have experience, it's been there growing for 10 years. The collaboration with MatTech came about because we saw a reason, a way that we could work together to define certain objectives for the grant that would ensure that when we generated data, the regulators would say that's good quality data, for example, doing the same experiment both labs and getting the same result.

Mary Parker: 
Now in your experience, this is only out of my own curiosity, how often are new cosmetics products tested? Are they mostly just recombination of old previously tested chemicals, or are people actually innovating anything they put into perfumes and makeup and things like that?

Mary McElroy: 
I would say there's always innovation and new combinations and new mixtures. So that is ongoing.

Mary Parker: 
So as the field of inhalation toxicology continues to evolve, we are shifting from the more traditional animal models to human-relevant predictive approaches. So what is the limitations of these new sort of more human relevant models and how do in vitro alternatives address them?

Mary McElroy: 
So the current in vitro models, if I understand you correctly, they are generally what we call in 2D monolayer culture. And quite often they are from derived from a cancer cell line of some sort. So they have a deranged genetic makeup and possibly abnormal phenotype are more inclined to proliferate and maybe expressing less metabolic enzymes. So the more complex models, particularly the upper airway models are three dimensional structures. They're multilayered in the upper airway. They have cilia, which is important for mucus and goblet cells that make mucus. So they replicate the upper airway, they look morphologically like an upper airway in a human from a human trachea, and they have more representative metabolizing enzymes.

Mary Parker: 
And naturally the human-derived models would maybe have more relevance than animals who are a mouse, say for example, doesn't have a similar airway to a human when you come down to it.

Mary McElroy: 
There's a huge number of differences. But one of the main ones is that animals breathe through their nose and when they breathe in an aerosol, a lot will deposit in the nose. Whereas if a person were to breathe in the same aerosol, a lot would ndeposit in the nose because our noses are simpler and therefore more of the aerosol goes into the deep lung and therefore is potentially more damaging.

Mary Parker: 
And of course, not even to mention that sometimes we breathe through our mouths.

Mary McElroy: 
Yes we

Mary Parker: 
Do very little filter at all there.

Mary McElroy: 
And our noses are much less complicated, so there's less trapping of particles and aerosols.

Mary Parker: 
So I understand the collaboration between Charles River and MatTech also involve the development of a rat epi airway model. What is that?

Mary McElroy: 
Yeah, so the rat upper airway model was developed by MatTech really to fill a regulatory gap And I would say from the regulators, they want to understand that if you have a complex model and you observe toxicity in vitro, how does it translate to toxicity in vivo? And that's difficult to do in humans. But there is very good rat in vivo data from GLP inhalation studies with many hundreds of chemicals. So the question is if we have a nice rat model and we work out a concentration of that toxic, if we do all the in vitro to in vivo comparisons, will it come up with the same value as toxic in the rat? And that's how we're trying to fill the gap. So while MatTech has developed the model, the grant and the collaboration are enabling us or providing us with funding.

Mary Parker: 
So how does this model contribute to understanding inner species? Variability in toxicological responses

Mary McElroy: 
It will provide lots of data, but what I think in practice when we come to work through our chemical list, we'll probably find some chemicals that fit the mold. If you find it's toxic in the rat, in vitro, it'll be toxic in the rat. But I expect there'll be other chemicals that don't fit nicely into that bucket and probably will come up with reasons. But I think when we work with human models and we work for human assessment with a human model that's representative, we'll find out more about the toxicity of that chemical to the human and we'll understand, we'll get more of what we call a mechanistic, we'll, a better understanding of if a chemical's toxic, why is it toxic? Or we'll have a much more understanding of why something is toxic in human. Whereas we just don't get that in the rat to the same degree because when you do a toxicity study in a rat, you are generally looking for very general tox. Sometimes it's even just a body weight change or sometimes it's a pathology change, which is quite general.

Mary Parker: 
How does the mat tech epi airway model enhance human relevance and testing?

Mary McElroy: 
Because let's say we find something as toxic, let's say it causes the cells to explode, we can maybe investigate that's very dramatic. We can investigate why in human why is that, and that might give us better way of understanding tox and maybe a better way of screening for it earlier. If we understand what the initiating events were, then we can look for changes in the initiating event rather than the cell explosion. And we maybe provide evidence that a lower dose is protective rather than a safe dose is going to cause very little harm. We can change the way we think about assessing toxicity. So this is dose, if you don't go above this dose, you're in a protective region.

Mary Parker: 
What role does dosimetry modeling play in this and how does it improve the accuracy of inhalation tests?

Mary McElroy: 
It's a whole different field as part of our grant with the ACC and collaboration with MatTech, we also have specialist computational modelers who will take our in vitro data and model what we call deposition. Or that's when you breathe in an aerosol where it lands in the lung and if it lands, how long it's going to stave there and how will it accumulate and then how that will predict a toxic dose. So it's very, very key. And some of these models, these computational fluid dynamic models are very sophisticated and they can mimic as well as the structure of the lung. So they can also mimic breathing the dose to the lung depend on the aerosol concentration, but it will also depend on how you breathe. I mean, if you're running fast, you're going to bring in a lot more air, so your dose will be much higher than if you were asleep and sedentary. Or if you have an underlying lung condition, let's say it's fibrotic, then your breathing is pattern is different and it'll probably be shallower and faster. So that means your upper airways will get a higher dose. So you can better model chemical deposition if you like in the lung in health and disease. Yeah.

Mary Parker: 
What are the next steps in this collaborative effort? Are there plans to expand these methodologies to other organ systems?

Mary McElroy: 
Absolutely. I think we have a lot of opportunities develop the lung models further because we are currently looking at the predictivity of chemicals that are likely to harm the upper airways. So we would need to do the same approach for the lower alveolar space. So that's another kind of section of the lung. And we are modeling at the moment in the absence of immune cells, so we need to add in the immune cell component. And the bigger thing we are testing, we are testing a panel of 10 known chemicals. It would be better to have a bigger panel. But what we are showing is that we can make, if we join protocols and joint standard operating procedures, we can get reproducibility between labs and we can predict the toxicity class of chemicals using in vitro methods. So we are kind of halfway through the grant and kind of moving in the right direction, but still objectives to complete. And with that we hope that would be a springboard to more just to build up a whole kind of prediction model for the lung, like both upper and lower airways.

Mary Parker: 
So as in vitro models gain traction, obviously regulatory acceptance and cross-industry adoption are really important to make sure that these are actually implemented. How are regulatory agencies responding to the data generated from these models?

Mary McElroy: 
So I'd say first off, I was at a micro physiological conference about a month ago, and I think this is fairly reflective of all the conferences I've been to recently. It's a very high, regulators are very engaged in the process. They are part of the panels, they are leading sessions and they're asking questions and they're providing advice. So they want to see data from alternative modelsthey want to understand what's good about a model, but they also want to understand its limitations. And for that they realize these data, particularly in the drug discovery space, can be as confidential. So they're talking about safe harbors for data so that they can see raw data and understand it maybe in relation to reference items or a particular known chemicals which are known to cause harm or less harm. So that's one thing they're doing their presence, they're providing advice. They're also running workshops and providing frameworks, frameworks that they're advising companies such as ours and the developers to follow to accelerate the uptake of these models. So I would say there's a lot of enthusiasm and I think as time goes on, it's moving towards maybe more concrete approaches people can follow in a pragmatic fashion to help build confidence.

Mary Parker: 
Do you know of any specific examples where in vitro approaches have influenced regulatory decisions or have even been accepted in lieu of traditional animal studies?

Mary McElroy: 
Yes. So there's two big things So one case in the lung space, it was largely driven by Syngenta and my team here in Edinburgh actually provided some of the raw data from the in vitro models. And Syngenta had a pesticide called chlorophyl and it was known to be toxic. And they argued with the regulators that a 90 day inhalation rat toxicology study would be pointless. So they made a case to use in vitro lung models just like we are doing, look for that concentration that was toxic, understand the actual exposure in the field, what people who are spraying the pesticide might be exposed to along with computational modeling. And in a hundred page document, they came up with a safe dose and through lots of discussion that was accepted as an OECD case study in lieu of a 90-day inhalation study.

So that's not a guideline, it's a case studies and it's an indication of approach if you wanted to show a pesticide was safe you could take. And the second example is that I've been privileged enough to be on the BIOS safe NAM committee. And in the past year and a half, which finished off with a paper about a month ago, showing that NAM-based approaches can be used to set the first in human dose setting for disease indications which are severe, such as cancer and driven by scientific reasons. So if you have a therapy such as a very specific antibody therapy, which only binds to human cells and not any of the tox species, there's no point testing the toxicity of that antibody in an animal, absolutely no point. So the large pharma and companies, they worked with the regulatory authorities in scientific discussions that were within an IMD package to say scientifically, this is justified scientifically, this is a safe dose. And within that dialogue and between the regulators and the developers, they came up with first-in-human doses based on non-based approaches. Now it is caveated, it was for disease indications that were severe and it was for disease for which the target was well known. So the toxicity was already known. So I think those are two important examples.

Mary Parker: 
Those are very illustrative. Thank you. So what unique value do human-relevant models bring to these sectors?

Mary McElroy: 
I think it's the human relevance. I think the capacity to look at mechanistically, why something is toxic and not toxic. I think the ability to look at toxicity in disease state because a person may respond differently if they're not healthy versus diseased, the ability to look at different nationalities, different ages, things like that. So more complexity, more refinement and understanding of the response in different human subpopulations, whatever they be.

Mary Parker: 
Yeah, that's a good point because when you talk about the difference between rats and humans, obviously there are a lot of differences, but there's even differences between people who live in different parts of the world that are exposed to different things growing up that have different metabolisms, what have you. So those are absolutely relevant as well to safety testing to make sure that whatever chemical it is, is actually safe for everybody.

Mary McElroy:
Yeah, absolutely. Absolutely, yeah. I mean in the short term, I think that it's going to be hard to replicate all aspects of human physiology in a dish at the moment, including systemic exposure. I think that's accepted. But I think for sure introduction of in vitro models can certainly compliment animal models and reduce the number refine number of animals used. So they are already being used for internal decision-making, and that's in both pharma and the agrochemical space because why wouldn't you use a human model to assess toxicity if you can before you start deciding on a lead candidate?

Mary Parker: 
Absolutely. I mean, regulators aside, a company has to make a decision to go forward with a chemical or a drug themselves before it even gets that far. So that's a very good point.

Mary McElroy: 
Yeah.

Mary Parker: 
How do you envision the role of human relevant in vitro models evolving in the broader context of toxicology and safety assessment?

Mary McElroy: 
Well, that's a big question. But yes, greater complexity and with vascularization. And in that way you can then start to bring in the immune cells and have understand the relationship between the tissue and the immune cell response. But as the models develop, we'll go hand with the development of science. So the more we understand how our bodies stay the same in health and how we respond to infectious agents, I mean, that's a huge large part of science that we don't properly understand. If we don't properly understand, then we can't model it. So it is going to be hand in glove with the development of the model, development of the science, then the improvement in the model, and then more knowledge of the science. So it's in that very steep learning curve because we have more sophisticated models to actually understand ourselves better, really. So we're in that sort of a very steep learning curve

Mary Parker: 
Still in the early stages of that feedback loop.

Mary McElroy: 
Yeah. Well, in my opinion, and I would certainly say in relationship to lung, which I've studied for so many years since my PhD as I mentioned, so I can probably talk with most authority about the lung

Mary Parker: 
Well, if you're curious, several episodes ago I talked with an expert on multi-organ chips systems linking several different organs on a chip to create something approaching facsimile of an organ system. So recommend checking that out.

Mary McElroy: 
Great, thank you.

Mary Parker: 
Any final thoughts on the significance of Charles River and Matt Tech's collaboration and its potential to transform inhalation toxicology?

Mary McElroy: 
That's another great question. I suppose the way I think about it is that change in this space, it kind of goes in, fits and starts. There's a big change, and then things rumble on for a while and then something else happens and there's another spurt to stimulate change. So for us, for example, it's the developers who are making very good lung models in vitro that we can actually use. So that standardization, I would hope that our grant, because we have designed it to answer the regulator's concerns, would be a step change. It would be one of those step changes that they will provide data that shows it's reproducible and predictable, and then provide the protocols that enable others to reproduce what we're doing and then be brought into an OECD guideline for greater uptake. So I'm hoping that we will contribute to that step change. And with an OEC guideline, anyone who wants to assess the inhalation risk of their chemical have something to follow so it's not guesswork and follow based on what the regulators want to see. So step change,

Mary Parker: 
That would be a big, huge deal. That would be very cool.

Mary McElroy: 
Yeah, that would be cool. And that's what the team here wants, and we all want as part of this team, I'm committed to doing our best to achieve that aim.

Mary Parker: 
Sounds like it. Well, thank you Mary M for being part of Sounds of Science from me, Mary P, it's been a pleasure having you on the show.

Mary McElroy: 
Thank you very much, Mary P. It's been a pleasure to talk to you. Thank you.

Mary Parker: 
Mary McElroy, head of Discovery, toxicology and Pharmacology at Charles River. Stay tuned for the next episode of Sounds of Science. Until then, you can subscribe to Sounds of Science on Apple Podcasts, Spotify, Stitcher, or wherever you get your podcasts. Thanks for listening.