Genetic Toxicology & Carcinogenicity: Insights on Drug Safety, Mutation Risks, and Cancer Risk

Carcinogenicity is a property of an agent that after exposure to it, it could increase or induce mutation or malignancies and it could result in formation of tumors or increased incidence of malignancies, or it might shorten the time of tumor occurrence and usually it occurs through either genotoxic pathways or non genotoxic pathways. So for genotoxic non genotoxic pathways it could be cytotoxic cell proliferation, hormonal changes, or it could be damaging some key proteins within the cell and genotoxic events could be adult formation, DNA strand break, faulty repair mechanism, base dimerization, and also chromosomal aberration. I have to say that not all genotoxic agents are mutagens and mutation is a heritable change in the DNA that stays in the sequence and it's permanent and it can be pass and not a single assay can detect all mutations. So that is why we have so many assays for detection of genotoxicity.

Hello everyone and welcome to the Emory Pharma Speaker Series. My name is Janet Liu, the Director of Biology here at Emory Pharma. Today is my pleasure to be welcoming Dr. Tufan Parma to Emory Pharma. She is a board certified toxicologist and has done extensive work in cell signaling and cancer research. She completed her Master's in Physical Organic Chemistry, followed by her PhD in pharmaceutical sciences in Toxicology at University of Toronto in Canada. She later worked as a postdoctoral fellow at the NIH National Cancer Institute, the University of Arizona and Stanford University School of Medicine where she focused on studying cancer cells and GPCR and cell signaling. Afterwards she spent over 15 years at Sri International where she led several departments including being a Senior Director of Toxicology and Immunology. She is currently the Senior Director of Non Clinical Safety Evaluation at Sangamore Thermopeutics, a biotechnology company that focuses on cell and gene therapies. We're excited to welcome here today and we look forward to your talk.

Thank you so much for being here to listen to this presentation and thank you Ron for inviting me to give this presentation to you today. The topic of my presentation is Genetic Toxicology and Carcinogenicity. What do we need and when do we need it? I want to try to wrap this up in 45 minutes. Might go over a little bit. Just a quick disclaimer that I'm an employee of Sangoma Therapeutics. Acknowledge John Marsalis here. Dr. Marsalis is a Vice President of Translational Development at SRI International. Then I'd like to thank him for significant contribution and input on this particular presentation for the genetic toxicology segment and providing several of the images that I have in this in this presentation for various assays. So I'm going to walk you through the outline first. First I give you an introduction, then talk about genetic toxicology assessment, timing and regulatory recommendations. Then I walk you through various assessments for mutagenicity and genetic toxicology, followed by genotoxicity relevancy to carcinogenicity and then carcinogenicity assessment, timeline and recommendations. I will continue with carcinogenicity study designs and case studies and finally give you a summary. What is genotoxicity? It is defined essentially as the effect of chemicals and physical agents on the genomic portion of the cell, particularly DNA that would impact the integrity of the DNA. Carcinogenicity is a property of an agent that after exposure to it, it could increase or induce mutation or malignancies and it could result in formation of tumors or increased incidence of malignancies or it might shorten the time of tumor occurrence. Usually it occurs through either genotoxic pathways or non genotoxic pathways. For non genotoxic pathways it could be cytotoxicity, cell proliferation, hormonal changes, or it could be damaging some key proteins within the cell. Genotoxic events could be adult formation, DNA strand break, faulty repair mechanism, dimerization and also chromosomal aberration. I have to say that not all genotoxic agents are mutagens and mutation is a heritable change in the DNA that stays in the sequence and it's permanent and it can be passed on. Not a single assay can detect all mutations. That is why we have so many assays for detection of genotoxicity. In terms of timing of when to conduct genotoxicity assays, you can start right from the screening of your process during the drug development. During the time of lead compound selection you can use quantitative structure, activity relationship or high throughput screening in order to knock out some of the compounds that are potentially genotoxic and may not be suitable to move forward. Then once you've moved your lead compounds forward, you have in the non clinical development prior to IND submission you would need to do GLP in vitro mutagenicity assay and chromosomal aberration in mammalian cell lines. Then some people prefer to do their micronucleus assessment in vivo before ind, but you don't have to. You can do that prior to phase one. However, by the end of phase one your micronucleus Assay should be completed as well. And then finally you have to complete any additional bioassays that is required before submission of the NDA, which is the new drug application in order to get approval with respect to drug discovery. And when you do genotoxicity assay, you probably want to start with an AMES assay. You want to make sure you continue doing high throughput Ames assay until you are confident that there are no structural alerts. Some of those structural alerts are here, but most chemists no longer use these structures in the compounds. But every now and then you might find one. Then the other option is to utilize in silico predictions such as quantitative sar. Or you can use AI which is now large and people. Everybody nowadays use artificial intelligence here during this time you want to kill early and kill often. You don't want to carry something that has potential alert past the development because it's expensive. Development is expensive. Now in terms of modality and entity, what do you need for these types of entities? I'm going to walk you through some of these entities. If you have a small molecule and it is planned to be a single dose in clinic, you need to do an Ames assay before phase one and then you finish the rest of the panel before phase two. If you have a small molecule that has multiple dose is planned to be multiple dose in clinic, then you need to do both AMES and mammalian cell assays before phase one and then you will have to complete the rest of the panel by phase two. The next thing is that for anti cancer drug you don't need to do any full panel until the phase three. Biologics don't have a standard genotoxicity assessment the way we do it for small molecules. But if they have any concerns with biologics, for example, if the biologics is gene editing, then you probably need to do genotoxicity assessment. But the current panel won't work. So there are specific genotoxicity assessment for biologics and then for DNA reactive impurities, you don't have to do any assessment until phase three. Today I'm really focusing on the top two which is mainly small molecule. But I have to tell you that irrespective of whether you have single dose in clinic or multiple dose, most companies choose to actually do all of their genotoxicity assessment before phase one during the IND submission. Because these assays are cheap. It's safer to go with a peace of mind to the clinic and not worry about genotoxicity. In terms of the panel of assessment there are two options according to the ICHS 2R1 guidelines. The option one is to perform Ames assay and then to select one of these assays, either a chromosomal aberration or mitonucleus assay or mouse lymphoma assay. These are in vitro assays. Then also to select one of the in vivo assays in mitronucleus assessment. It could be either acute standalone or it could be integrated into your 28 day or standard GLP toxicity. That will be option one. Then option two is again you can choose to do an Ames assay. You can select to do one of these micronucleus assays and then you select one of the assays listed down here. Either comet, uds, transgenic mouse model, rodent model or pga. But most companies, what is most favored by everyone is to conduct option one, item one. So aims micronuclease in vitro and micronucleus in vivo, particularly as part of the repeat dose toxicity assess. That's what most people do. If all of your assays are negative in genotoxicity assessment, then you potentially have a green light to go forward. There is no further assessment is needed. However, if it happens that your human metabolite is not there in the preclinical studies or preclinical model, if you see carcinogenicity that you cannot explain based on mutagenicity, or if you have a structural alert, if you have any of those, then you will have to do further assessment even if everything else is negative. Now if you have positive. So for example, if you have positive aims, usually it's a rare event because most people eliminate those ahead of time. You could consider halting the development. But for aims, if it's only in one strain, I wouldn't halt the development. However, if you have multiple, for example five of the strains are showing positive, then you may want to consider do I want to keep moving this forward? If you have another candidate, might be able to move that one forward instead of the one that has Ames positivity, then you have mammalian cell assay which often you get positive results. They are prone to having false positive. If you do have a positive result in mammal cell assays, these are in vitro assays, you have to really do additional studies to clarify clinical relevance. Then if there is a positivity in in vivo assays, for example in micronuclease in vivo, it's usually very rare. But if it occurs, you kill the program. Nobody will move forward with that. That's because, I mean you could Move forward if you wanted to, you could add additional studies, but bottom line is that eventually after multiple studies, FDA may still not accept moving something with micronuclease positive forward. What do these assays measure? Mutagenesis is measured by Ames mouse lymphoma assay or MLA and transgenic rodent models. DNA damage or DNA repair is measured by direct measurement of adducts or covalent bindings. COMET assay or UDS Cytogenetic or chromosomal aberrations are meas by chromosomal aberration in mammalian cells. Chromosomal aberration in spermatozoa or in bone marrow or in rodent bone marrow assay. I'm going to walk you now through each of these assays in terms of Ames assay. First, a concept that these assays are conducted in five bacteria strain. Four of them are salmonella strain and one is an E. Coli strain. These bacteria are essentially engineered to have a mutation in the hist operon. This mutation essentially results in formation of an enzyme that is mutated and this enzyme cannot convert histidinol or any of the other histidine precursors to histidine. Now if a mutagen comes along and interacts with this bacteria and eventually remove this existing mutation and fix it, we call this reverse mutation. Then this enzyme becomes active because it's no longer mutated and it can now convert histidinol to histidine. Therefore, these bacteria, before they are mutated, they cannot live in an environment that doesn't have histidine. They have to have histidine to grow. How do you run the assay? You essentially incubate salmonella bacteria, whichever bacteria you want to use in a test tube with a histidine containing environment. Then you add rat liver microsome S9 fraction or human liver microsome S9 fraction which have the metabolizing enzymes in it. Then you split this into multiple containers. One can be your vehicle control. Another for example, can be your test compound. What you're going to measure and test, you always include a positive compound here, evaluate functionality of your assay. Then you plate these in a histidine minimal histidine containing plates. Of course they should not be able to survive, but you have natural and you incubate them for a while. There are some natural revertants that occur here in the vehicle control. Here for example, you are seeing natural revertants that exist. Then if you look at your test compound you can see that you have significantly more revertance than the control group. If you look at Your positive control, you can see the plate is covered with the revertant. This one is actually a very strong AMES positive control. Here's an example of how we present the data. Essentially you have on the Y axis you have the number of revertants per plate and on the x axis you have the dose. In the tubes that do not have the S9 fraction, you will see no mutation. This is for Benzoyl Pyrene in TA100 you see no change, no mutation happening. But when you add the metabolizing enzyme, you can see that there is a dose dependent increase in revertant formation. What this is telling me is that benzo pyrin requires metabolic activation to become a mutagen. There is a bioactive material there. Moving on to the MLA assay. This is very similar to the AMES assay with the difference that this is now a forward mutation. This assay is conducted in L5 1,787 lymphoma cells. Essentially, these cells, when they are in heterozygous format, the TK plus minus, they have the ability to incorporate thymidine kinase into the cell and then they will die once they incorporate. Now, if you add a mutagen to this assay, it can convert the TK plus minus heterozygous to the homozygous version, the minus minus, which essentially would be a. Now, these particular cells that are mutated, they cannot incorporate TFT and therefore they can survive. Now this assay usually have large colonies and small colonies. Large colonies are showing essentially point mutations. You count those for point mutations. Small colonies are representative of cytogenic changes and you count those for those they're very sensitive to pH, sensitive to cytotoxicity and also osmotic changes. Because of that, there is a huge number of false positive in. One of the important things for you to do when you use this assay is to do a dose range finding to make sure that you select the top dose correctly. Here is an example of an MLA where we are doing a cytotoxicity assay first to select the top dose. Usually the standard is that you either do 24 hour exposure of the drug to the cells without the S9 fraction, or you may be doing 4 hour with and without S9 fraction. What you want to really see is where you find your 20% RSG, which is essentially the relative suspension growth. At 20% RSG you have 80% cell death and you do not want to be more than that. That's according to the guidelines. Here you can see that for the 24 hour we have achieved the cell death at around 25 microgram per meal. This is of course more than 20%. If you look at the 4 hour with and without s 9 fraction, you can see that with s 9 fraction we are at 200 and with s 9 fraction, sorry, without s 9 fraction we are at 400. In order to select the top dose from here, we need to look and see where the 20% falls. However, we didn't have any dose levels at which we for this particular assay we had 20%. So they elected in this experiment to go with the 65% RSG. That means they have 35% cell death here. Then here you can see that the top dose for the 24 hour would be somewhere around 5. And the top dose for the 4 hour with and without S9 fraction would be somewhere around 160 microgram per mil. What they did is they did exactly that in their MLA assay. Now for oxfendazole they conducted the 4 hour exposure without S9 fraction at doses of 10, 20, 40, 80, 160. As you can see, there is no mutagenicity here. They did a four hour with S9 fraction and for this they did not only one assay but two. Because they wanted to have a confirmatory assay. There's two sets of four hours with S9 again you see no mutagenicity compared to the positive controls. Then for 24 hours they used the 1.25, 2.515. Here you can see that again, they have no mutagenicity. Aximedazole is not mutagenic in this assay. As I mentioned earlier, there is high incidence of false positive in this particular assay. Around 2006, Peter Gasper presented to the EMA Subject Matter Expert meeting a set of data that showed percentage of mammalian cell studies submitted to German authorities. In German regulatory authorities between 1995 and 2005, they looked at 804 studies, 596 compounds. All of these studies were divided into three. Either they were MLA human lymphocytes or Chinese hamster cells. Of these 804 studies, 219 of them were positive. That's 27%. And they were divided to pretty much 9% of MLA, 9% of Chinese Hammerster and 9% of positive human lymphocytes. This is a pretty huge rate of positivity. And that resulted in elimination of 181 compounds. That's 30% of all the compounds evaluated got eliminated. You have to be careful not to have these false positives. How can you avoid that? You can avoid it by not exceeding the recommended top concentrations of 1 millimolar. Evaluate your cytotoxicity. Select the top dose correctly. Make sure it's no more than 20% RSG or 80% cell death. Do not test in the precipitating range because you're not going to get accurate reading. And then use aporia target cells. Most people use p53 deficient cells, which are essentially mouse lymphoma and Chinese hamster cell lines. They're okay, but you could have more false positive there. The best cell lines are human lymphocytes and human lymphoblastoid cells. Moving on to the big blue assay. For this assay, I'm going to walk you first through how the mouse itself operates before I tell you how this experiment is conducted. These mice are transgenic mice and in the normal state they have a LACZ and LACI gene. The LAC I gene, when it's expressed, it actually produces a protein that sits on LACZ and prevents LACZ from producing beta galactosidase. It doesn't work. Then if a mutagen comes along and mutate the laci, then this protein is no longer made to prevent LACZ from producing beta galactosidase, therefore it can produce it. Then you would be able to visualize those colonies that have beta galactosidase using X Gal. Now here, if you treat the mouse with a mutagen and Observe it for 28 days, you can collect various tissues. The good news about this is that you can actually look at each tissue separately. This is the only assay that you can use to look at tissue specific mutagenicity. You can collect your tissues, extract DNA, put it in the phage and then infect the E. Coli, plate it and then incubate it for a while. And look out for the colonies that express beta galactosidase. When you incubate that with XCal, you can see these little blue colonies that are formed here. Your mutation frequency is essentially a number of blue plaques divided by total number of plaques that you have. Here's an example of mutation induced in mouse liver. We're looking at mms, which is methylmethanosulfate and dimethylnitrosamine. Here. The way the experiment was conducted is that mice were treated as a single dose injection of one of these compounds, MMS at 20 milligrams per kilogram or DMN at 2 milligram per kilogram and then they evaluated mutations on day one, eight and 22. First of all, DMN and MMS, they are both AMES positive, they shine in Ames assay and they're also both methylating agents. However, if you look at this assay, you can see in this particular Big Blue, MMS is not mutagenic, DMN is mutagenic. But what is interesting is that DMN is a known liver carcinogen. MMS is not a known liver carcinogen. What you can see here is that first of all, Big Blue correctly identifies the tissue specific mutagen and that not all mutagens in Ames test cause mutagenicity in vivo and not all mutagens that are AMES positive are carcinogens. So that's why you don't want to throw the baby out with the bathwater when you're doing the Ames test. Moving on to the pig a mutation assay. First, the concept pig A essentially stands for phosphatidyl nocetylglycan class A. This has a gene, pig A that can pump out what is called gpi, which is a protein that can connect the gpa, which is GPI anchored protein. GPI anchored proteins could be many. I'm here showing only CD59 as an example. In the endoplasmic reticulum, these GPI and CD59, they can combine and they can then come out and be presented at the cell surface. If a mutation happens in the GPI in the PGA gene, then what will happen is that this GPI is no longer formed while CD51 can still be formed. Because there is no GPI, CD59 cannot be presented on the cell surface. Now this is essentially happening in the progenitor cells in the bone marrow. These progenitor cells, whether they're normal or abnormal, they will then cross from the bone marrow into the peripheral blood stream. Now, if they are normal, you can detect the CD59 on the surface of the cell red blood cells using an antibody that could fluoresce green for example. Or you would have a mutated form, in which case There is no CD59 and you probably detect them using a different kind of antibody and they will show a different location. Here is the forward light scatter for floor cytometry. And down in the fore fourth quarter quadrants, here you see the mutated one, whereas in the top upper part or the third quadrants you see the wild type red blood cells. Here's the example of how these studies are conducted. One thing I forgot to mention here is that actually Pig aging is expressed in both males and females. It's on X chromosomes. So males have one and females have two. But in females one of them is silent. So you can generally use males, you don't have to use males and females. And you can use control and three dose levels. The top dose will have to be a maximum tolerated dose or a limit dose. Then you collect blood throughout the 28 day study or 56 day study if you will, at pre dose day one, day 15, day 29, this is the dosing period or during the recovery on day 50, 42 or 56. If you're looking at red blood cells, the mature cells, you want to look at greater than 10 to the 8 cells. If you're looking at reticulocytes, you want to look at greater than three times 10 to the six cells so that you can have some power of assessment. Here is an example of ethyl nitrosourea mutagenicity. In this PGA assay here the animals were treated with 40 milligrams per kilogram of EU one time six rats per groups. If you look at predos, you can see here there is no mutant red blood cells. If you look here you can see that now after dosing they have some mutants are formed. Then you can present that in a graphic format here. On the Y axis you would have the CD59 negative RBCs. On the X axis you have the days of collection and you can see on day 0 and 7 they see nothing. But suddenly on day 14 they begin to see mutagenicity of ENU. And then of course that increases as the time goes by.

At what stage? You mentioned that you try to. As you're developing your compound, you want to be judicious in terms of when to do these studies. So at what point are you doing this? Are you doing this with. At pre IND stage?

I would do those all at pre IND stage aims and in vitro mutagenicity assays, I would do them all before ind.

The mouse study.

The mouse study also before ind. The only thing that I wouldn't do before ind if at all, I like to do all of them before ind, but in vivo micronucleus assay you can do it post ind, but financially it's better to do it before IND as part of your toxicity.

Cost wise.

Cost wise? Yeah.

What kind of cost?

Oh, it depends. It depends on what you're doing. But usually these studies are not that costly. We're talking about somewhere between on early stage, if you do high throughput, they're you know, 20, 30, 40,000. If you get to the later stage, if you have to do them GLP, then you're looking at, you know, 100 to 100, $200,000. With the micronucleus in vivo micronucleus assay, if you do it as part of your toxa study, it's a lot cheaper because the animals are already paid for through that. And those studies are usually very costly otherwise.

Yeah, for the antibacterial, which is the best.

I say for the antibacterial. Well, for antibacterial compounds, it depends if you can get it solubilized. If you can get it solubilized, I would say you definitely have to do, let's say an Ames assay. You probably have to do the panel of aims. Although actually it depends on what you use it for. It depends on what you use. If it is short term use of antibiotics, not long term, then you probably don't have to do that before ind. But in the end, if you have to do it, you have to be able to make sure it's soluble. All of these cell assays are not solubilized. I mean, it's hard to solubilize into the aqueous form. Which one is better? Does it really matter?

Some of the antibacterial, they show the false positive.

They do sometimes. It depends on what assay you use. I would say stick with the option one, which is the Ames assay, the micronucleus in vitro and then a micronucleus in vivo. I would do those instead of any of the other ones. But again, you have to figure out if it goes into solution. Comet assay. With comet assay, this is the measure of DNA strand break. It's a single gel electrophoresis that you conduct. And it's very simple assay. You essentially put your material into your cells into a tube. You add a mutagen or your carcinogen or whatever into the tube. If it causes any DNA damage, then when you run the assay on a gel electrophoresis, you're going to have this comet forming DNA. Essentially what happens is that the head of the DNA move forward and any DNA break will show as a tail of the comet. This assay is actually no longer. I mean, it's being used, but it's not frequently used. The reason is that it's being used sometimes it's inexpensive, it's sensitive and it can work, but it's very prone to cytotoxicity. And so a lot of people prefer not to use it at this time. So I'm not going to give too much details on this one. Moving on to the UDS assay, the unscheduled DNA synthesis assay. It's very simple. Mutagen carcinogen can interact with the DNA and form the adducts. And you guys are all very familiar with these adducts. During the excision repair process, the excision endonucleases will cut out these adducts that are connected to DNA. And now you have this disjointed DNA. Then during the repair process, if you have a treated thymidine, it can go into this DNA and repair the DNA with the repair enzymes. And now you have DNA that has treated thymidine in it. And you can measure that by evaluating how much thymidine you have incorporated into the DNA. In terms of the assay, it's very simple. There is in vivo and in vitro assay. I'm going to walk you through the in vivo. The in vitro is very similar. So in the in vivo assay, you expose the mouse to the carcinogen or mutagen and then you collect the hepatocytes. After perfusion of the animals, you put the hepatocytes on the slides and then take the slides and incubate it with the treated ethymidine. Then the cells are washed out, the COVID sleeve is put back on and you insert that into an emulsion that would then eventually shows or highlights the cells that are granulated. The cells that are granulated are cells that are damaged. In terms of the in vivo assays, the same thing, except that you buy the hepatocytes that are normal and healthy and then you incubate it with the mutagen and the treated thymidine. You wash it, you put it into this emulsion to highlight the cells. Then you look and count the cells that are affected. Here is an image of a normal control that has all the cells in a circular nice shape format. Then here is a dimethyl nitrosamine exposed cells. As you can see, these cells are all granulated. These are all affected cells. So that means this drug is a mutage or it damages DNA forms adults. Chromosomal aberration assay is another assay that essentially directs assessment of the ability of the compound to damage chromosomes. It's typically conducted in CHO cells and sometimes in human lymphocytes or some other cells. The exposures are usually 3 to 4 hours with and without S9 fractions, or 21 or 24 hours without S9 fraction and then it measures essentially structural changes in the chromosome, whether it's break, whether it's translocation, polyploidy or whatever that it could cause. It can also be done in bone marrow cells. It is also losing popularity because this is very subjective. It's very hard to train someone to be able to detect these chromosomal aberrations. It takes forever to measure it because someone with a great eye can measure 50 cells at a time. You can imagine if you have to go through stacks of slides, it would take forever. There's a lot of false positive. And the in vivo assay, if you actually have to do this in vivo, is horrendously expensive. We're talking about millions. Here is an example of the exchange that you can see here or a chromosomal break here that you can see for some compounds. How does micronucleus assay work? Micronucleus assay is actually very simple. If you have a normal cell and expose it to a drug that could cause DNA damage or chromosomal damage. During the cell division, usually the chromosomes split and go to two different cells. Sometimes the chromosomes don't do that. They either break or they form these anaphase bridge. Or they might just have a kind of a chromosomes stranded somewhere. Then what happens is when the cell divides, one cell will have the normal nucleus and then another cell will end up with a normal nucleus plus a micronucleus that contains these broken pieces of chromosome. This is this concept of micronucleus assessment. How do you do that experiment? Essentially you dose the animals with a potential drug. You observe them for 28 days or something like that. Then you collect bone marrow, put the bone marrow on a slide, stain it with acridine orange and then visualize it under microscope. Usually you're looking at RBCs or reticulocytes. You measure both here you look at mature cells that are not stained. The immature cells are usually stained orange. And then the immature cells that have the micronucleus that have a yellow dot in there. Here's an example of an in vivo mouse micronucleus. You can see red blood cells are not stained. White blood cells are stained, but they don't have any nucleus. Then here you can see that the reticulocytes have these yellow micronucleus in them. And reticulocytes are immature red blood cells. This is an example of a 28 day, twice daily dose administration to rodents for a compound. A. I cannot disclose the name of the compound. The positive control was cyclophosphamide. And we were looking at polychromatic erythrocytes, which is essentially the red blood cell cells that have micronucleus on the Y axis. And then on the x axis you can see the dose levels of the compound. A 300 milligram per kilogram and 1,000 milligrams per kilogram, so two funds.

So, for example, you have your pre ind, you're conducting this micronucleus assay. You see this spot. What is the decision at this point? Is the compound dead?

Dead? Other compound is dead?

Yeah, but is that. But what if this is the only thing that's happening? Everything else is.

It doesn't matter because that's. Yeah, the micronucleus essentially trumps everything else. Now, if you have, if you end up with, if you have invested already so much into this program, with the micronucleus assay showing positive, then some people might choose to actually go and do additional assays to demonstrate human relevancy and say, is this only happening in rodents or can it happen in humans? Then you have to go and repeat that in human cells. Then you have to demonstrate it doesn't happen in human cells. Then if you do that, then they wonder that you probably have to do multiple other assays to show that maybe somewhere else you missed it. Because again, you only use three assays to go to index. They might take you back and say, do additional in vivo assays. So then you have to do all of those. It's costly. If I could, if I were to advise my company, dump the one with the micronucleus, go with the backup compound. If you don't have a backup compound and that's your only baby, then you have to decide you want to spend all that money or not. And most often, if you have micronucleus in rodents, you're probably going to have micronucleus in human. It's very sensitive in that way. So in this particular assay, you can see that CPA is positive, which is expected, and then the drug itself doesn't have any micronucleus formation. What is the experiment like essentially, in terms of in vitro micronucleus assay? It is very similar to the in vivo micronucleus assay. It is very sensitive. You can use various cell types. Before even you go to animals, I would Recommend Doing this one. The top dose is 1mg meal 3 dose levels. You grow the cells to about 1.5 to 2 normal cell cycles and then you expose them the standard way, 3 to 6 hours with and without S9 fraction and 24 hour with S9 without S9. Then you stop the cytokinesis, which is essentially you stop the cell division at the time that the two nucleus haven't separated yet or the cells haven't separated there, but the nucleus has separated by adding cytokalosin B. Then if there is any micronuclease, you will not see it in the binucleated cells. You stain them with crest and you visualize it on the microscope. Here is an example of a binucleated cell that hasn't gone through a complete division. You can see that it says two normal nucleus or nuclei, I should say. Then here you can see a binucleated cell that has two normal nuclei plus a micronuclei. Here. Then here is how you graphically present the data. Again you have compound A that we're evaluating at multiple dose levels. And then the positive control is mitomycin C. You can see that mitomycin C is pretty positive in this assay. But in this case the compound A was also positive at any dose levels we've given. So you can see this compound is practically probably going to die. I remember this compound. Actually, we ended up killing it. Yeah. In V2. And in vivo you used a different standard. Different standard. So no, the standard. So for the in vivo assays, cyclophosphamonide you use. We use cyclophosphamide for you using mito. Mitomycin. What's the reason behind that? I think it's just because it's easier. They had probably at this, for this stage they had slides that were already prepared before. But sometimes you can do these assays, but you don't have to do your positive control at the same time. You just have slides. Sometimes they have that material in house. You could potentially also use cyclophosphamide. It's just that we use this at that time. What is the relevancy between genotoxicity and carcinogenicity? Well, as I mentioned, not all genotoxic compounds are carcinogenic and not all carcinogens are mutagenic or genotoxic. There is a single positive results in genotoxicity assay does not necessarily mean that you're going to kill your product, but you have to then do additional assay to show human relevancy. You have to make sure that you do cancer risk assessment for your drugs before you continue into your clinical trials. There is a guideline on how to assess cancer risk. That guideline was put together recently by FDA because they Wanted to help reduce the development timeline, reduce the animal use and also to have a better understanding of the carcinogenicity in animals before they go into patients. For phase one. Okay, sorry, not phase one, phase three. What considerations are you going to have for carcinogenicity assessment? I'm not going to go through this entire table because we're probably going to run out of time. But I'm going to tell you that you have to make sure that you have the cause for concern factors evaluated. You're going to make sure that your product class, if it has carcinogenic potential, you're going to look at that. You're going to make sure that there is no evidence of for example hyperplasia in your assays. If you have any of those, then you will have to do carcinogenicity study. Drugs that are used for long term and you have had unequivocal genotoxicity. You have to use carcinogenicity studies. Drugs that are used for treatment of certain serious life threatening disease such as Ebola, Zika or anything that is life threat cancer. You don't have to do carcinogenicity study before you go to phase two IND or even to phase three. Sometimes drugs that are used for lifetime of the patient such as hiv, those you have to do carcinogenicity study and drugs that are used frequently but in an intermittent fashion. For example, you might take, I don't know, a drug every six weeks for the rest of your life, then that drug for that you have to do carcinogenicity assessment. Then drugs that are used infrequently such as antibiotics or anesthetics or imaging agents, they don't need carcinogenicity study question.

So what about some of the older drugs that are in use, they were invented in the 80s and such as lisinopril, I call it lifestyle drugs, Lipitor, some of them may have been done genotoxic studies and carcinogenesis, some of them have not. Are we going back to do those things? What's the guidance from the fda?

So no, they're not going back to do those because for those drugs you have already patient populations that have a lot of information for and they can go back to the epidemiology of those drugs and look and see if there was an incidence of increase in cancer formation then they would decide how to advise the patients who are taking these. So if for example if you put aspirin into some of these assays it probably shows positive so they're not going to take it off the Market. But what they're going to do is they're going to look at the population and see if there is anything, any issues. If there is a major issue that they see, if they see increase in incidence of carcinogenicity in the population, who's taking for example Lipitor, then they're going to recall and start looking at it very similar to some of the things that you guys have done, like the metformin for example, that had the nitrosamine or some of the, or the ones that you were looking at Zantac, that had it. So if they see that then. Or if there is a reason for a cause to evaluate. So for example, again with nitrosamine you saw a lot of compounds, ended up having a lot of drugs, ended up having it as a. So they had to go back and look at those again. They don't take them off the market. Metformin is still in the market, but they're going to look at the process and see what can they do in the manufacturing process to change. I don't know if I answered that question. In terms of carcinogenicity assessment and timing. You really do the carcinogenicity two year carcinogenicity studies on phase two. And for some drugs you could do it after approval, but for most drugs you do it before new drug application to get it ready for approval. Then the, the regulatory requirement for doing carcinogenicity studies in United States is to actually go through FDA process that is called special protocol assessment. You essentially interact with FDA through this to gain approval from what is called the Executive Carcinogenicity Assessment Committee or ecac. In order to determine what those levels, you have to use how many number of animals you have to use, whether the design of your study is accurate. So FDA will participate in that. You cannot start your carcinogenicity studies without FDA reviewing all of these documents. Usually, FDA spa assessment is not just for carcinogenesis. For many other things that requires going to patients and for approval, carcinogenicity is one of them. These studies are two years long, plus the fact that you have to do a reporting process that would take about a year for these studies, you really have to start this at the beginning of your phase one so that three years later you have the results to submit. Now what will happen during this carcinogenicity and weight of evidence assessment? Essentially, FDA expects that you take all of your data, the entirety of your data, and you evaluate it and you conduct this integrated analysis of weight of evidence. Once you do that, then you fall into Potentially three buckets. One of the three buckets you could fall on the bucket, for example. That's shown here as carcinogenic potential in human is likely. If you have that, then what you want to do, if your data shows that it is possible that it's going to be carcinogenic, you want to tell fda, there is no point for me to do this two year expensive rat study we know already is carcinogen put on the label. You might fall into the second category, which is the carcinogenic potential in human is unlikely. In which case you argue with fda. You don't want to do the two year carcinogenicity study because it's unlikely to be carcinogenic. Fda, however, might ask you to do a mouse carcinogenicity study in transgenic mice, but you might be able to also avert that if you have the proper evidence, and I'll get to that later. Or you might say, okay, I looked at the data and the potential in carcinogenic potential in human is uncertain. In which case you have no choice but doing the two year carcinogenicity study in both rodents, in both rats and mice. Now, what do you need for your carcinogenicity assembly in terms of weight of evidence? This table gives you all the documentation that you need to put into this package. I'm not going to go through every one of them, but just know that there are seven categories. The data from your pharmacology, the data on your class of compounds and class effects, data from secondary pharmacology, histopathology data from your animal studies. You need to include evidence of any hormonal changes that could occur in these animals. Then of course, genetic toxicology studies, whether they were equivocal or unequivocal, and then evidence of immune modulation, whether you cause any kind of immune system effect, then plus that any additional mechanistic studies that you've done, you want to incorporate into this weight of evidence assessment, then what FDA does is that they're going to take all of these and they're going to look at this and say, let's see in what spectrum you fall. Are you going to fall on the side that you have to do the two year carcinogenicity study or to decide that you don't have to do the carcinogenicity study. And the way they look at that at is by looking at these five parameters, they're going to say, well, let's look at the target biology. If you have poorly characterized biologic pathways and you don't know exactly your Class effect, you're more likely to fall on this category. If you have secondary pharmacology that shows there is an off target activity somewhere that could potentially lead to some toxicity, you will fall on the left side in terms of histopathology of chronic. If in your chronic studies you suddenly have liver hyperplasia, then you fall on the left side. If you have hormonal effects, for example, you're changing some endocrine pathways, you go on the left side. Genotoxicity. If you have positive genotoxicity or genotoxicity of which you are uncertain, you don't know whether it's relevant to human or not, you will be on a more likely doing the two year toxicology study. And then if your immune, if you impact the immune system, as I mentioned, you will end up doing this. If not, if you don't have any of these and everything is solid on your rate of assessment, you fall under likely less likely to do it. I don't know how I'm doing in time. So let me walk you through the carcinogenic cases study. I want to try to wrap this up as quickly as I can. Essentially this is an example that you can actually find in the guidelines is an example for a drug that is an antagonist of a neuronal G protein coupled receptors. Here the investigators, in terms of target biology, they've shown that there is a predominant receptor expression in the brain. They have demonstrated that this results in decrease in acth, which is the pharmacology of the drug that they have shown. For compounds that are like this compound that they've done previously two year rat studies, there was no carcinogenic effect. In terms of target, there is no leading to carcinogenicity. In terms of their secondary pharmacology, they've shown that there is an antagonistic effect of this drug on another receptor. But it happens at really large doses. The pharmacology of the off target is not associated with tumorigenesis. And they've also shown that in the two year rat carcinogenicity studies for other compounds similar to this that they see what is called the cell adenomal carcinoma with increase in thyroid hormones. Now this is very standard carcinogenicity that is observed in rodents. So this is not related to the pharmacology of the drug or even off target. This happens in rodents naturally. This is not an issue here. Then they looked at the histopathology of the chronic studies. In their chronic studies they saw no carcinogenicity. They did see some increase in liver size or liver hypertrophy, which essentially means that they're activating liver enzymes. They saw some increase in thyroid follicular hypertrophy. At very high doses they also saw some. In the non rodent study they saw some hypertrophy of other organs. That happens at much larger doses. These exposure levels of high dose means that at the doses that are relevant therapeutically, there is no toxicity. Then looking at the hormonal effect in the chronic rat studies, they did see the adrenal rate decrease in histopathology, which is consistent with decrease in acth, which is the pharmacology of the drug. In the chronic rat studies they saw no reproductive effect. In the rat fertility studies they saw lowering of the pregnancies. But that doesn't really mean. But that happened at high doses and again not at therapeutic doses. Their genotoxicity was normal and their immune modulation immune system was not impacted. When they did additional investigations, they looked at Cytochrome P450s again they showed that there is an increase in the level of these or induction in these particular Cytochrome P450, CYP1A2 and CYP3A1. They did see interestingly, bone and teeth fluorescence. And fluorescence is actually serious and it could lead to carcinogenicity. However, this was only observed in rodents and not in human. Again, they're showing the relevancy to human and there is no relevancy to human here. Based on this, the investigators essentially argued that they do not need to conduct a two year carcinogenicity study because it does not add value because it's not a potential carcinogen, it's not likely to be carcinogen in human. Based on that, they got the green light to go. Now two more slides and then we wrap up. Just talk about the two year rodent carcinogenicity study and the design of these studies with respect to the rodent toxicity. If you do rat two year study or mouse 18 to 24 months study for rats you need 50 male, 50 female. For rodents or mouse you need 60 male, 60 female per group. The evaluations include mortality, morbidity, clinical signs, body weights, food consumption, necropsy, organ waste, histology, histopathology and so forth. FDA may ask you to actually do control and high dose, but Europe may ask you to evaluate all those groups. Of course you have to do statistical analysis, but the key part of these studies is to interact with FDA about the top dose. You cannot select your top dose based on the maximum tolerated dose. You have to select it based on the criteria that are either, for example your dose is higher than 25 fold rodent human AUC that you usually determine from your toxic kinetic studies the dose that causes saturation in metabolism or absorption. You could select a dose that is a maximum feasible dose, a limit dose, or you could select dose at maximum pharmacodynamic response, but not any higher than that. The mid dose is geometrically selected and low dose. You should still have some sort of a rodent to human ratio of auc. FDA might ask you to do two controls instead of one. You might have a control for the vehicle and a positive control the species. If you use CD1 mice or spragdoli rats, you probably have to do a short pharmacokinetic studies, particularly if you have not used these particular animals in your toxic studies. Also the CD rats. If you use CD rats, you probably have to use more animals because their lifespan is shorter. So this 50 male, 50 female is for hand wisdom. You don't have to do toxicokinetic evaluation, you don't have to do ophthalmology evaluation. In terms of clinical pathology, FDA doesn't ask for it. European may ask for hematology. And then of course they want all of your carcinogenicity studies to be conducted in the same facility. There's a good reason for that. I won't get to it, but you can ask me later. In terms of the transgenic mouse studies, these studies are shorter, they're usually six to eight months. And it's preferred by FDA that you first use this type type of studies. Again, the number of animals is very limited. 25 male, 25 female is a lot lower than the other one. The evaluations are identical to the other study, the rodent study that are not transgenic. In terms of recommendations by fda, they usually expect you to do a transgenic mouse study. They want you to prioritize this. However, if you can show in the weight of evidence that you are definitely sure that it's not going to be carcinogenic to human based on the data that you have so far, or if you show that the data indicates that only sub therapeutic level of the drug can be reached in these animals, then you don't have to do this transgenic mouse study. In terms of the top dose, very similar selection, except that for transgenic mouse studies or for the mouse studies in general, you're supposed to use greater than 50 fold rodent. Human AUC, remember for the other study was greater than 25 fold. This is greater than 50 fold. The rest of it is pretty much the same except that for species you use Ras H2 transgenic mice, but some people may use P53. In summary then you can use early screening to get rid of your unwanted drugs early and often. You can use QSAR and you can use high throughput aim. The battery of genotoxicity tests has not changed over the past 25 years with the exception of adding the in vitro micronucleus that wasn't there before and a PGA assay that is new. And also the top dose for the in vitro mammalian assay is now reduced to from 10 millimolar to 1 millimolar which would help with this stability issue. With this solubility issue, some genotoxicity assessment is still required before the phase one. In fact I would say all of it is required before the phase 1 in vitro genotoxicity results trump in vitro. If you have in vivo toxicity trumps the in vitrogen. Carcinogens can be genotoxic or non genotoxic. You need to do weight of evidence to demonstrate that you do not need to do two year carcinogenicity study. I would recommend that to everyone to do it anyway way for weight of evidence you have to go to FDA and use the SPA ECAC system. Then of course your top dose for the mouse studies is for the rodent studies is greater than 25, for the mouse transgenic studies is greater than 50 fold rodent to human AUC. I think with that I'll stop. Do I foresee that if the genotoxicity assays will change change with the modality of the drugs? Well, yes, they do. So for example, I mean it depends on what the modality is. They're still in the situation where if you're using small molecule you use these panels, but again for biologics you don't use these panels. There is a complete different way of assessing genotoxicity in biologics. You have to do integration assays. You have to look and see where in the DNA is integrating. You have to. There are several other assays that you look at in order to assess whether genotoxicity happens in biologics. Biologics didn't exist when the small molecules existed. They had to come up with new ones. But would these particular assays change 25 years here later? We only had PGA added. It is possible that someone comes with a new technology that is better, faster and easier to detect these genotoxicities. But the issue is that even if you come up with a new assay. These assays will have to be validated and we have to be reproducible. That's why they keep sticking to the same old assay, because they are reproducible. If you come up with a new assay and show reproducibility, then that could become a new one if it's better than all the other ones.

There is a push to get rid of animals from doing these kinds of SO assessments. How are we doing with that? Is that possible?

That's a one billion dollar question. I tell everybody that in my lifetime that's not possible and probably in my children's lifetime. And I tell people the reason for that is in vitro systems don't have the entire organism, right? So when you take a drug, it goes through your liver, it gets metabolized, it goes to your kidney, it gets metabolized further in the kidney with different enzymes that don't exist in your liver. As it goes through your guts, it gets metabolized more there. And there are gut bacteria that could also protect you from some of the toxicities because they will just metabolize and get rid of it and prevent the bioactivation. Well, if you look at the in vitro and you don't have the guts and the liver and the kidney, how do you mimic that? So, yeah, you could potentially have now a system, elaborate system. They have these microfluids and they have 3D cultures now that they try. But again, it's not always the same as the organism. Another example is, for example, receptors. Not all cell cultures that you have have all the receptors or the channels that you need for these drugs to get into the system. A lot of those are missing. And if they end up up developing something that has all of those, then you already have an organism that is maybe not natural organism, it's a synthetic organism. Then the ethic issues with that comes, I think when somebody asked me, is that possible? No, I don't think you can replace it. But what you can do is to minimize the number of animals used. That's what FDA came up with this particular weight of evidence. Can we use weight of evidence? Can we use the in vitro to screen majority of the bad ones? And if something passes the in vitro, then the chances are that it should be passing the in vivo, hopefully. So when you go to in vivo, you minimize the number of animals that you're using. And that is possible. And that's what they're doing. They're doing that for reproductive toxicology studies where they do a lot of in vitro assays to see if the drug has an impact on the embryo. And then based on that, then they select their drug. Then they take one drug only to the reproductive talks to check the box. I hope that answers your question.