E88: Rethinking Toxicology: The Future of NAMs in Agrochemicals

Robert (12:25):
And this what three R principle dictates. in regulatory testing you could use it more, maybe not ultimately replacing all animal testing, but devising a testing strategy which is saving animals that is not only true for in vitro methods, it's also true for other methods like using big data in computational approaches or grouping and read across approaches.

Mary (00:21):
I'm Mary Parker and welcome to this episode of Sounds of Science. We are joined by Robert Landsiedel and Susanne Kolle from BASF. Welcome Robert. And Susanne, Robert (00:42): Thank you for having us. Thank you.

Mary (00:44):
We're very honored to have you. So can you tell us a bit about yourselves and your current roles? Robert, would you like to go first?

Robert (00:51):
Sure. Susanne and I would both work at BASF Chemical Company with the headquarter in Germany in Ludwigshafen. And we are doing tox studies for the company. We are an experimental toxicology and ecology unit and actually in the unit of special toxicology, which I am heading. And that unit has different labs to perform toxicology studies, mainly in vitro studies, but also biokinetics and inhalation and special toxicology. We do about 500 plus toxic studies per year under GLP. And we are having research projects on the development of new methods, new approach methods, alternative methods, publicly funded projects, internally funded projects and collaborations with outside institutes.

Robert (02:03):
We also have students about eight to 10 PhD and master's students working on these projects. And above that we are also engaged in education. So we are giving lectures, collaborating with universities, having master studies in toxicology. And just to finish on me personally, I'm also the current president of the German Society for Pharmacology and Toxicology.

Mary (02:40):
That's amazing. Susanne, how about you? What's your background?

Susanne (02:44):
So my name is Susanne Kolle. I'm a bio technologist by training and I have joined BASFs Experimental Toxicology and Ecology as a postdoc. After that, I was a lab team leader for more than 16 years, starting in the development of alternative, going through the application of alternative methods and biokinetics and local toxicity. And my current function is to coordinate the mentioned research activities on alternatives, so NAMS and also the chemical strategy to sustainability. And one other function I have is to sit in OECD expert groups, for example, for local toxicity and validation.

Mary (03:29):
I actually used to live in the Midwest, which was in soybean country, so I'm actually very interested in agrochemical safety. So is that the kind of thing that you do or do you do other kind of chemicals that might end up in the environment?

Susanne (03:55):
We do test anything BASF produces, so that includes agrochemicals but also industrial chemicals and cosmetic ingredients or anything that's produced within the company can be tested here in this amongst our study portfolio.

Mary (04:11):
And I noticed that both of you had a lot of entries on ResearchGate. You've helped author a lot of papers. Do any of them stand out as particularly interesting pieces of research that you're really proud of?

Susanne (04:34):
So, I guess in my portfolio it is the in vitro sensitization world because we have developed the two out of three approach that ended up in an OECD guideline after many years of hard work. Back then it was written by one of our PhD students because Robert mentioned we have students that we have. So I think that's one of the most cited papers. And then also one of the areas that ended up in an OCD test guideline eventually replacing animal testing for regulatory purposes.

Mary (05:12):
That's excellent. Well, speaking of which, the use of animals and safety research has really evolved over the years. So what are your views on this journey? Robert, do you want to go first?

Robert (05:22):
Yeah, maybe important to say first that most of the testing we do required by law is still animal testing. And of course most of our safety assessments rely on animal data as a consequence. And that has worked well, more or less, but more. Well it has worked well as we have not experienced a lot of great failures in the safety assessments. Of course there are spectacular failures, but few, so it worked overall it protected us, but it is inefficient, it is unethical, very questionable, and it's not at all keeping up with the progress in biomedical science. So these are three good arguments why we feel we could do much better.

Mary (06:24):
Yeah, I would agree with that. What about you Susanne?

Susanne (06:27):
I'm sure agreeing as well. And I think human relevance and also reproducibility are just two of the points that make animal experiments from the scientific side not as good as alternatives that are available.

Mary (06:46):
What pathways currently exist that provide viable alternatives to animal testing? And maybe can you touch a little bit on the regulatory angle? I'm curious about the EU side of things.

Susanne (07:26):
Yeah. So it very much depends on the area you're touching upon. So there's more or less low hanging fruits like those in the local toxicity area where we have NAMs available that have OECD guidelines and well that really can make a replacement with limitations because they are not one size fits all. So not every NAM that has an OECD guideline can be applied to the whole chemical or product universe. So there's limitations to the story, but there's guidelines for local toxicity. But then if you talk about systemic toxicity or more complex endpoints like reproductive toxicity, that's of course a different story. And I know Robert has an example he'd like to make here.

Robert (08:14):
Yeah, the current paradigm, the systemic toxicities approached is why adverse outcome pathways, and that is of course linking the two observations. Of course you want to prevent adverse effects in humans and animal studies deliver just that or exactly that, adverse outcomes in animals. So in a way, the comparison between the outcome of animal study and what you really want to know is conceptually easy because you compare adverse effects to adverse effects. When you look at vitro methods, it's not the same at least from both methods, the local toxicity taking apart, but for systemic toxicity, look at early events, something in the cell happening which will progress to affect the organ and then the whole organism or not. So it's early and it's not directly linked to the adverse effect and the chain between the early observation in the cell and the adverse effect you want to prevent, that's the adverse outcome pathway.

(09:26):
Now the idea is that we can build assays to look at cellular effects and make the connections to adverse effects. That's currently done at a speed of few assays and few adverse outcome pathways per year. Now if you think that we have about 35 organs, we examine in an animal study and we do, even if we did five assays per year in the OECD and just to mention it, we usually need more than one individual assay. So three for the sensitization, probably more for other organs. It'll take decades until we have a full replacement. And that is a conservative estimate. It's not painting it black. That will happen if we continue as we go. So we think we need to find another way to use the data we are getting from in vitro methods other than linking it to adverse outcomes, which are then the ones which are regulated because we have to be clear for chemicals. Clearly it's a highly regulated area and the regulations are tailored to the results of animal testing, namely adverse outcomes. You come with results which are not adverse outcomes, but bioactivity in a cell, you need some way to make a connection. But that's a detour. And so we believe AOPs are really a detour and quite a long journey if we want to make progress fast and if we want to have a direct approach, we need to find a way how we channel bioactivity data directly into regulations.

Mary (11:31):
Would you say that possibly in terms of more immediate goals, it might be more realistic to not say, not try and replace animals entirely, but get to a point where a lot of this work in the early stages is done in vitro or in silico so that by the time you get to animals, you're pretty sure what's going to happen and you'll have fewer failures at the animal level, which will also of course save money.

Susanne (11:56):
Yeah, I think that's one of the common practices done in during development of substances. And I think that's the same for pharma or similar approaches are taken for pharma as for chemicals and agrochemicals. So because we cannot do animal experiments for all the compounds early in development and we don't want to because it's of course using animals for ethical reasons. So I think that's one way to go and that's something that is being done within an industry

Robert (12:25):
And this what three R principle dictates. So of course we were up till now talking about full replacement, but reduction and also refinement are the two other elements. And just to be complete, of course you can use animal testing in screening to avoid non-animal testing. Of course, you can use in vitro methods to avoid animal testing and screening because you can sort out candidates for example, right away and do not have to do the whole tox testing program. But also in regulatory testing you could use it more, maybe not ultimately replacing all animal testing, but devising a testing strategy which is saving animals and to be complete. Of course, that is not only true for in vitro methods, it's also true for other methods like using big data in computational approaches or grouping and read across approaches.

Mary (13:28):
Well speaking of that, one of the new alternative methods or NAMS that we've talked about a few times is virtual control groups or VCGs where basically big data sets from previous control groups are used instead of a live animal control group. And Charles River has a collaboration with Sanofi to explore the use of VCGs in nonclinical toxicology for example. But in your opinion, how will collaborations like this help move the needle towards regulatory acceptance of alternative methods in general?

Susanne (14:04):
So I think collaborations like that in particular bringing together CROs and industry are essential to bring the field forward.

(14:13):
And we can learn a lot from the pharma sector here. And because it is industry and the CROs that provide the quality data, they have standardization, they have experience and also new quality standards like the good in vitro method practices. So I think this kind of collaboration is really essential. And practices like the Innovative Health Initiative, Vict3R, which has more than 30 partners, including a little bit of chemical industry including BASF, are also looking into the virtual control groups. And I think there's really high potential saving round about 25% of the animals and repeated toxicity studies, but we need regulatory acceptance. So we need cases and we need to learn and we have to discuss with regulators early on because they will have to accept it in the end. And if we cannot change OECD test guidelines to allow virtual control groups, it's a nice thing to do to learn and build cases. But if it's not in the guideline, we'll have to do controls groups on and on.

Mary (15:27):
So the virtual control groups are replacement, but they're also reduction. Do you think that it's possible that regulators might be willing to, if they're not willing to do the virtual control groups entirely, would they be willing to do a compromise of fewer animals in the control group with the addition of a virtual control group as well?

Susanne (15:54):
That's looking into the crystal ball if I have, knowing what regulators would be willing to accept. Yeah, I don't have an answer to that at, but I think we should try because there's so much data out there and it's worth a try and have also pairwise comparison because it's an easy thing to validate. So if you do virtual and real control groups and you can see how the results compare, I think that's an easy thing to do.

Mary (16:29):
Robert, anything to add to that?

Robert (16:33):
Well, just legally they are obliged to take it if it saves animal and whether it's scientifically valid and sound enough to be demonstrated. But I think if, and of course you can always have doubt and there will be gray areas, that's probably where the real decision is taken. But if it was a clear-cut scientific sound solution and saving animals, I think there wouldn't be an alternative that would just be the delay by the regulatory process to implement it.

Mary (17:16):
the 3Rs of replacement reduction in refinement was developed over 50 years ago to provide a framework for performing animal research. And Charles River obviously takes the 3Rs very seriously, but we also add a component of responsibility. So where and how do you feel that role of responsibility comes into play when thinking about new alternative methods?

Robert (17:40):
Well, you mentioned the 3Rs and we see that as part of the responsibility and animal welfare in general. And I do believe in Europe the regulations are at least as strict as in the US probably more, at least the democracy is more for certain. But on top of that, we want to not only, and of course, but it goes without saying we are complying with animal welfare regulations, but we want to reach standards which are above that. So we have an AAALAC certification, which is the association taking care of animal welfare standards above the legal requirements.

(18:35):
And then part of the responsibility is also that we take care that the experiments we do and that would apply both to animal testing and in vitro testing, have a good quality and are reliable. Obviously it's the greatest waste if you do an animal study and cannot use the experiments in the end either because it was the wrong study, the wrong study type, the wrong dosing, or it was performed that the data are not reliable because the procedures were not strictly followed or things like that. So quality control and careful planning is definitely part of responsibility. And again, when it comes to animal testing, there's an ethical aspect to that as well, but same applies to in vitro methods.

Mary (19:32):
So how can NAMs be expedited and be accepted amongst regulatory bodies so that organizations like us can make it accessible and make it an accessible research option for clients.

Susanne (19:45):
So I think key is really global harmonization for the classification requirements and we touched on that previously. So currently the classification rules are made for animal experiments. And as long as we have those and we have legal requirements that are not harmonized in some demanding animal tests, I think this is not going anywhere because then you have your partner, some will have global businesses in different sectors and then you can do an alternative for one area, but then you have to do the animal test for another area. So I think that's really key and that's also one of the things that are discussed within the EU roadmap for phasing out of the animal testing. So we really need rules that are norm based.

(20:36):
One of the key things and things we may not forget is the assessment of the relevance which will need to change. So we cannot stick to the one-to-one replacement. We need ways to define relevance for bioactivity or more on the effect side and not so much based on the one-to-one replacement, but still we need to look at reliability so we cannot forget about reproducibility. So that's key, even for the simplest method that ends up in regulation, we really need that. I think one other key aspect is really the predictability. So we can also not change regulation to have a NAM requirement in parallel to the animal testing requirement. If there are new rules to the game to do classifications based on NAMs, we need really predictability. So if we shift to NAMs and have requirements there, so authorities have to accept that and that's meant by the predictability and guidance for business more or less here.

Robert (21:46):
Maybe to add to that Charles River and general CRO labs or BASF labs or general in labs, we're the only ones in the whole world of labs dealing with NAMs which apply them in the end, and which have a good quality control like GLP or good in vitro method practice or ISO standards or all of that. And I think we have a unique role

(22:15):
Because in a way you can develop a in a university lab, but you will never find out whether it's fit for GLP for example, if you don't run it under GLP, if you don't have a ring trial with a GLP lab and a lab which is using it in routine use, you will never find out. So I cannot see it. I hope it will not happen that it's happening without labs like yours and ours because it's a way, in a way producing a product without asking the customer, which is you and us. And the other aspect, and I guess that's part of the question at the moment, there are so many NAMs developed in different stages, all of them seem to be either successful or promising. Funny enough, there's not a single failed methods, though there is, but it's not that communicated. And of course bringing it to regulatory acceptance, that is a hurdle and I think we need to keep high standards, but I think the process could be more smooth, probably faster and better chaperoned by people who know because you are pretty much left alone when you go for regulatory acceptance. The third thing is that when you want to use a method, which is not standard regulatory, you sit in front of a pile of hundred methods and don’t know what to do. So I think for us, same for you, it would be great if we had some guidance on which methods to use and which not below the regulatory acceptance level.

Mary (24:18):
It does come down to being practical in the real world, not just applicable in an academic lab. We've cured cancer a thousand times in mice and it doesn't always make it into human patients. So I think that's kind of a similar idea. Would you say that having good data is kind of part of the solution, having data that is consistent and entered well, I mean especially for something like a virtual control group where it's similar to machine learning or at least it's analyzing big data sets in order to give you a real world readout of what the control group might be acting like?

Susanne (25:17):
Yeah, I think that's really key. And then we need to learn how to deal with these things and it's not only us but also the regulators. And I think then it's always good to learn from experience and to compare to the things we've done we've been doing in the past.

Mary (25:37):
So in your opinion, what role do contract research organizations play in supporting the evolution of next generation safety assessments?

Susanne (25:47):
I think it's the CROs and industry that really frame the field for the quality. So the standardization, the quality measures like ISO certifications are also the good in vitro method practices. And I think there's early involvement really needed during the development, but also carrying that through to regulatory acceptance because there's so much experience out there. And I think it all ends also in the accessibility of methods. So if there's a random method, whether it's relevant or reliable, but if it's only available in a single lab in the world or the test system or any essential component is not available, there's not much use if it's the best method ever. So because it's really sharing knowledge and experience and cases,

Robert (26:43):
We are really concerned with the quality of methods and data. As you know, there is well described the reproducibility crisis in biomedical science and there's no reason to believe that it will not happen to toxicological methods if we don't take care. And again, the only ones who could are CRO and industry labs. So that's a great responsibility. And of course our wish is that CROs get involved more. Same wish to other industry labs of course, but I think for some reasons the door is not that wide open. I think we have to push to get in and we have to fight to keep quality standards up because if you never had a quality system, you don't know what you're missing out. I'm trying to put it, academic labs have a different purpose, so there's a danger that that aspect gets lost. So our wish would be that CROs, like Charles River gets involved more and that you along without push for quality good in vitro method practice, to have robust and reliable reproducible to have robust and reproducible methods. Now having said that, we have a bit of an advantage because we have internal research money to develop methods and do that. So we are at least not from the start, do not rely on making actual business or generate money out of the new methods.

(28:42):
But in the end, it's always the same question, when do you take a method aboard and there's zero risk. If it's regulatory accepted, there's more risk the earlier you do it. But again, if we don't get involved early, we'll miss out on the quality and at least.

Mary (29:04):
Yes, that's definitely a good point. So in your opinion, how will NAMs disrupt and advance safety assessment?

Susanne (29:15):
We need to switch to NAMs more and more. I don't have a good answer on the disruption because there's NAMs already have them in regulatory places as well, but I think it's NAMs themselves that will disrupt anything here. I think other pieces need to be disrupted, like the classification systems and harmonization of global requirements and things like that because there's so many NAMs out. I think if we searched for them, there should be, yeah, we could solve many problems if the NAMs are sufficiently relevant and reliable. But yeah, I think it's other places that need disruption.

Mary (30:04):
I think their potential is to enhance and obviously for the betterment of the 3Rs I think they're there as a boon where the disruption would be in the regulatory area and in like you said, global harmonization.

Robert (30:25):
I mean since we talk about the word disruption now and in the community, there's this discussion, will it be a revolution or an evolution? It'll be an evolution, which of course from the standpoint of NAMs, it's still a good thing because NAMs are the more advanced species if you want it. We will see a lot of parallel testing firstly because NAMs will probably be introduced before the animal tests will be phased out. And even if Europe would be successful in that or the US or North America in general, it will take quite some time until the last region of the world will follow or have their own system of changing. And as globally active companies, we'll do double testing just because that for quite a while, that is of course maybe scientifically interesting because they have parallel data, but it's not helping animal welfare at all. And it's certainly hampering and being a burden for the business.

Mary (33:58):
So what does the future hold for NAM innovation and regulatory adoption in North America and in Europe?

Susanne (34:10):
In Europe there is a lot of changes ongoing. So, we are awaiting, well, it's no longer called the revision, but the simplification or remake of the Reach regulation, the chemicals legislation, yeah, but that's still within the drawer, so we don't know the details of that. But it's anything between nothing will change and a lot of will change adding data requirements and adding more and more substances or product classes on the lists of the to be tested materials. And then we have the Commission’s roadmap for the phasing out of animal testing for safety assessment. And that's really work in progress. So we don't know how this will, because it's work in progress, we don't know the details and what will be in there. But I think one particular issue in Europe is the hazard-based classification. So we do, it's not like in the US where you can do some exposure assessment and then do a true risk assessment. In Europe, we base classifications based on a intrinsic hazard of a compound, and I think this needs to be changed as well, so we can have a more realistic assessment of the risk really.

Mary (35:33):
Okay. That would be nice. Robert, any final thoughts?

Robert (35:37):
Maybe adding to the regulatory aspect from a scientific standpoint, that would be, first of all, we were very successful with the local toxicity methods, skin and eye. Now the other target of concern when it comes to local toxicity, at least for chemicals and agrochemicals, is of course the respiratory tract. And we don't have methods for that, but that seems to be close to the low hanging fruits,

Robert (36:13):
I think the US is ahead and we are collaborating almost exclusively with US partners because we do see a chance that might be implemented easy without the complications of having lots of organs and biokinetics and all that of course is not an issue with local toxicity. So respiratory I think should be given a priority when we may want to make fast progress. And the other is that regulatory demands and also scientific questions of course change and usually you add new aspects to that. Immunotoxicity is certainly underdeveloped when it comes to animal testing. Of course, we have some immunological relevant observations in the standards, animal tests, and we have skin sensitization testing. But for example, we have nothing for respiratory sensitization. And you can name others, certainly neurotoxicity area. There are things also in eco toxicology like bio concentration, but these are fairly new or change demands. And I think it would be sensible instead of discovering that and developing animal tests for that to say, well, if we have something new, we should start with NAMs right away and not have animal tests. And that’s just pragmatic. I would say the lower hanging and the areas where there is no good or no established animal tests, that's where we should go first.

Mary (38:13):
Well, Susanne and Robert, thank you so much for being part of sounds of Science. This has been a fascinating talk. Thank you for your time.

Robert (38:22):
Thank you very much.

Susanne (38:23):
Thanks for having us.

Mary:
Robert Landsiedel, Vice President Experimental Toxicology and Ecology and Susanne Kolle, Research Coordinator, Experimental Toxicology and Ecology at BASF.

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.