Inside Out Quality

TGN1412 and PMOA –Story of a monoclonal antibody that did the unexpected

March 23, 2021 Aaron & Diane Season 1 Episode 9
Inside Out Quality
TGN1412 and PMOA –Story of a monoclonal antibody that did the unexpected
Show Notes Transcript

In this episode, Diane and I are joined by Husain Attarwala, Head of Pharmacometrics and Clinical Pharmacology at Moderna and author of a paper diving into the 2006 failed clinical trial of TGN1412.  We discuss how things went wrong, and what to consider when planning clinical trials to help keep study participants safe.

Kelly Creighton, VP for Regulatory and Strategic Development at Clinipace also joins us to discuss the importance of understanding the Primary Mechanism of Action (PMOA) of biologics and FDA expectations.  

Aaron Harmon:

Hi, I'm Aaron Harmon.

Diane Cox:

And I'm Diane Cox Welcome to Inside Out quality.

Aaron Harmon:

both Dan and I build an implement quality systems in the biotech and medical device industry. But we often get asked, Is this really necessary? Do we know if we are doing too much too early? Or do we even need a quality system?

Diane Cox:

Our goal is to explore questions like these through real life events and experiences shared by our guests from various regulated industries. We will show you why quality is not just about compliance and how when it's done right, it can help your product and company improve lives and make a difference.

Aaron Harmon:

Doug Mofle also on Episode one told me a story of a monoclonal antibody that caused such severe adverse events during its first clinical trial that had stuck with him for years. When this study happened. The reactions were so severe that all six trial participants required hospitalization and for ended up in organ failure. The antibody that caused this was named TGN 1412. The TGN in TGN 1412 stood for T general, the company which brought this antibody to clinical safety trials, known in the pharma world as a phase one trial. You may wonder, what is the monoclonal antibody, our immune systems have tools to attack bacteria, viruses and even cancer cells. One of these tools is called an antibody, the B cells of our immune system with the help from T cells can make a Y shaped protein which sticks to the very specific spot another protein. Now for protein in the body is misbehaving. We can potentially stop it with an antibody. As an example some breast cancers grow fueled by a protein called her too. If you stop her too from being seen by the cancer cells, you can slow the cancerous growth, an antibody is perfect for that job, and the one that does is called Herceptin. Now, I mentioned the term monoclonal, each B cell makes a unique antibody. If you take the antibody from only one B cell, then it's called a monoclonal. If you take antibodies from multiple different B cells, you get what's called a polyclonal antibody mixture. Now back to TGN 1412. If you do a Google search for this monoclonal antibody, the first scholarly paper you will find is written by Dr. Hussain Attarwala, who studied this incident during his graduate work. If you follow any of the COVID vaccine work that's currently happening, you will come across a company called Maderna. That is where Hussein is currently the head of pharmacometrics and is here to tell us about TGN 1412. And phase one clinical trials which went very wrong. So with us today is Dr. Hussein Attarwala. Hussain, welcome to the show.

Husain Attarwala:

Thanks for having me.

Aaron Harmon:

So the first question I wanted to ask, the paper I had referred to was from your thesis work while you were studying at Northeastern University, what drew you to study and write about TGN 1412.

Unknown:

So in 2009, when I was a graduate student at Northeastern University, I was studying drug design, clinical drug development, and ethics and clinical trial. And this was as part of our pharma Pharmaceutical Sciences graduate program. So, as part of our drug design course, we were asked to work on a review article describing a story that we thought significantly impacted our understanding of translational science and best practices related to conduct of clinical trials, especially first in human clinical studies. And but then TGN 1412 was a very recent and very significant incident that had happened. And it jumped out as a very obvious spec, due to unprecedent unprecedented nature of events that happened after that. The events were unlike anything that the medical community had seen in the entire history of conducting first in human clinical trials. And this trial forever changed how clinical trials are conducted and how non clinical results are interpreted when it comes to selecting a first inhuman dose. So based on all of this special features that were related to this incident, and how it changed the course of history as to how clinical trials are conducted. It really I kind of dived in and thought that hey, let's work on it. Let's put together a nice review paper and learn as I go along during the course of that exercise. Yeah,

Aaron Harmon:

yeah, when I first heard about this, it was pretty crazy because you normally do not hear about these events happening with a phase one trial.

Unknown:

In general, like when a drug is considered to be taken for first in human trials, a battery of non clinical tests are conducted. And those are those are called a non clinical toxicology studies. So the aim of those studies is to de risk exposure into humans. So when The new chemical entity is given to a human is it safe to be given at a dose that is safe that it can be given that similar testings were done for TGN 1412. So there are a number of toxicology studies were conducted some in vitro experiments were conducted, and there, they came up with those that was considered to be safe in humans. So, the dose in monkeys that they found to be safe was 50 milligram per kilogram. And at that time, based on the guidances, that were available for selection of first inhuman dose scaled, that goes down to point one mg per kilo, which was 504 lower than the highest dose, lowest dose form safe in animals. But this TGN 1412 is a very unique molecule. So, the way it works is it binds to the CD 28 receptors on T cells, and without any CO stimulatory action on the T cell receptor, it can activate T cells. And because of this property of this antibody, it is called a super agonist. So TGN 1412 is called a CD 28 super gun is because of this reason. So at this low dose when it was given to help the volunteers on and this happened on 13th, march in 2006. All six participants who received this first dose, they experienced life threatening reactions, which was later identified as a sidecar kind release syndrome. And they all had to be moved into intensive care unit. And they were at a near there was fighting for their life. And luckily, all of them survived. But it was an eye opener for the entire medical, regulatory, and even public community that how and what should be done to conduct these trials safely as we go along.

Diane Cox:

Yeah, I bet. So how can an antibody make it through animal testing and still cause such a reaction in the human volunteers? How does it get so far?

Unknown:

So I will say this is a special case. So if you think about monoclonal antibodies, each monoclonal antibody is specific against an antigen that it targets. So if you talk about, let's say her to receptor for breast cancer, that monoclonal antibody that targets that receptor binds to only that target, so its activity is limited to and it is very specific activity, but TGN 1412 because of the super agonist activity is a unique case that was found there. Despite all of this non clinical testing that happened, what was missed in those trials is that the receptor expression expression in humans, that was completely different than in monkeys. So in monkeys, there is a cell type in both humans and monkeys, there is a cell type called memory T cells. But in monkeys, the memory T cells do not express the CD 28. While in humans, the CD 28 receptor is expressed on memory T cells, when they tested in monkeys, because of an absence of that expression of CD 28. In monkeys, a higher dose still was found to be safe, quiet when we went into humans, because memory T cells also express this. And those T cells then induce that pi cytokine release syndrome was very unfortunate, a sad day for those six young men who participated in those trials with an aim of hey, let's do something let's progress the science but they ended up struggling for their life.

Diane Cox:

So could have this been predicted. I know that you mentioned that, you know, there's a difference between the monkeys and then the humans and the expression of of the receptors. And so now that this is known, looking back, could this have been predicted?

Unknown:

Yeah, so there were some science if we carefully look at the toxicity results that they found in monkeys. So if we just thought about the mechanism of action that ESA activist he says by the city 28 agonistic activity, when we would think that it would potentially only lead to regulatory T cell activation, but not the effector T cell activation But in monkeys, low level of those cytokine release was still seen. If someone would have thought more than a, this is a flag that we shouldn't see pro inflammatory cytokine release with this kind of mechanism, then some that food probe some more, say, experiments on PBMCs, human PBMCs ex vivo cells. Yeah. But this is easy to say after the fact, when this was new, and it was extremely exciting because in non clinical models that were there in mouse models of 40 immunities such as arthritis, this antibody was looking fantastic. I mean, it would activate regulatory T cells, and it would cure diseases. So it was really thought to be a miracle drug for treating autoimmune diseases, there was this little hint that was missed. And if thought more carefully, you're like questioning each and every data point that we say in non clinical tests, then you would end up having a more conservative approach the world's first inhuman dose. So let's say if someone saw that low level of cytokine, in non clinical studies, then we would not follow that standard lowering of those of say 500 fold or something, we would go even more lower, just to be on a more conservative side of things. Yeah. But this has led to that kind of thinking this event has triggered that thinking approach across regulatory agencies and pharmaceutical companies that they are paying much more attention to all signs and signals and then how they can conduct trials as safe as possible, especially for such mechanisms where there could be over stimulation of immune system.

Aaron Harmon:

If you were advising somebody who is developing an antibody, or there were some other lessons learned you would apply that trial.

Unknown:

After this event, there were quite a few things that changed forever, as to how clinical trials are conducted. So one, one of the biggest thing that was done during to 2006. And before is if a drug is one safe and non clinical testing, they will give them they will enroll a cohort, a cohort could be six participants who are on active drug and two on placebo. And they will give that drug to all participants all at once. But after this event, something called Sentinel dosing strategy is implemented. So what is So, if a new chemical entity whether it is a small molecule or a monoclonal antibody, if it is given for the very first time to a human participant, then you can only give that new chemical entity to one individual at a time. You cannot give that chemical entity to all participants in a cohort all at once the way it was done for TGN. For one for sure, at least, if the if there was something some exaggerated unpredicted toxicity that would happen, it would limit to at least one participant and then you observe that participant for a period of time, see if everything was fine, and then those another one, and then those rest of the subjects, and then those escalate while monitoring safety at all times. And this has slowed down clinical trials. I mean, if you if you were waiting for let's say two days or a week in between dosing participants, and you had 30 participants to those in a clinical trial, then that phase one trial would take quite a long time. And there were many many discussions across regulatory agencies that what is the right balance that we do? We do not want to slow down clinical trials so much that you know takes forever to conduct them while we also want to be safe. There are new guidances that lay out the what kind of mechanisms are there, if there is if it is a new entity mimicking M carnism that is already known, then, you know, just one Sentinel and one placebo observe and then you can do as rest of the subjects. So those kinds of concessions are made. However, Sentinel dosing is a standard practice now across different new chemical entities that go in first human trials. Yeah. Think about some other things then. Choice of styling those. If we think about the CGI Infolink well incident, they went with a 500 fold lower those that was formed safe are those 504 lower dose led you to the lowest safe dose in my case that was chosen as the starting dose for TGN 1412. And that was based on just non clinical and oh al now what is anyway, it is no observed adverse event cloud. If we take this approach for something that has a mechanism that can cause super organism of immune system, then this approach is of obviously not conservative enough. So what is done that, first we find an NF E L in non clinical toxicity studies, then we conduct in vitro experiments to see, what is the receptor expression in human cells? What is the binding affinity to humans? Or is the binding affinity different is the receptor expression different? Are the cell numbers expressing those receptors in humans different compared to primates. And all of this is taken together. And then a starting dose is selected. So, a much more scientific approach towards selecting a starting dose is now being taken across pharmaceutical industry and regulators ask for it now, specifically, so they will not just approve something with a mechanism that is unknown or novel based on just No, yeah, so those guidances are no

Aaron Harmon:

frame, the antibodies are becoming such a mainstay in treatment, that it's good that though TGN 1412, had the problems that had, I think it's really kind of helped pave the way for a lot of other antibodies to get to market and do so a lot safer.

Unknown:

So similar, if we think about this Regeneron antibody that has gone through this, for COVID, that antibody has an potential low, very low risk of actually making the disease worse. So if the binds to that spike protein of the virus, and if that antibody is not neutralizing, but just binding it, then it can actually cause enhancement of viral replication by promoting uptake of the virus cell, because antibody is bound to the virus. And that antibody can easily get internalized by other cell types in the body, if it is not neutralized. So there are a bunch of battery of studies that needs to be done to mitigate that risk.

Aaron Harmon:

Sure, so you're essentially making the case that you have to use good science as well, that you can't just follow, regimented regulatory strategy, but also apply the biology of what's going on.

Unknown:

Yeah, each target has its own biology. So viruses, enhancement of viral infection kind of risk potential for if you're inhibiting a protein, which is in a strict balance. So let's say if it is something like antithrombin for hemophilia, now, we know that clotting mechanism is a piece of balancing guard, you would need to inhibit only to a certain level, if you have it more you will get coagulation across the body. If you have a class we see bleeding disorders. So things like that. So that balance, where do we strike that balance and how translatable that biology is, from nonclinical species to humans,

Aaron Harmon:

modern is currently leading the development of the vaccine for SARS, cov. Two, which is very exciting, as Glad you're able to make time to join us cuz I know you're a very busy man. Your work with them is in pharmacometrics. I'm pretty new to the term, but I hear it's pretty helpful for developing drugs and vaccine safety. Could you tell us a little more about that?

Unknown:

pharmaco metrics is a science of quantitative pharmacology. Now. What before that? What is clinical pharmacology according to clinical so clinical pharmacology is the science of drugs in humans under optimal clinical use and patients. It is underpinned by the basic science of pharmacology with an added focus on the application of pharmacological principles and quantitative methods in the real patient population. It has broad scope from discovery of new target molecule to effects of drug using in whole populations. Clinical Pharmacology also connects the gap between medical practice and laboratory science like what we saw with TGN 1412 n vitro, what is the receptor xpression, what is the binding ffinity number of cells, which ells are expressing all of that inking to those level xperiments to the expected linical outcome? The main bjective is to promote the afety of prescription maximize rug effects and minimize side ffects. So now, when we come to harmaco metrics, it is applying ll of these principles in do a uantitative framework. When I ay quantitative it is eveloping mathematical models sing differential equations, sing well defined concepts in athematics that scale from non linical species to humans, such s physiological scaling. So ike what is the blood flow rate n animals versus humans, what s the metabolic rate. So, in nimals versus humans, 50 go rom a mouse to a monkey blood low rate goes up, metabolism lows down as we go in a larger nimal or a human. So it helps redict a dose that is fficacious in humans at the ame time being safe. So scaling ifferent metabolic rates, blood low rates, between non clinical pecies and humans, and then oming up with a dose and egimen that is predicted to ive us that optimal window here the drug is expected to ork in

Aaron Harmon:

humans. It sounds like a really fascinating field, it's like you're taking guesswork out of scale up, it's really easy to do things in mice and then trying to predict how it's going to go up into the larger species. And then in the humans, if you can apply other sciences to make a much better prediction.

Unknown:

Yeah. And now regulators are asking for it. So if what is the strategy for first new window selection and modeling and simulation pharmacometrics is one of the major guiding method to come up with that dose range. Let's think about rare diseases where there are only say, a few 1000 patients in whole entire world. And we might, in our clinical trial, we may be able to enroll, say 50 participants. And with those many participants, and given the duration, we have to come up with a dose that is efficacious, and also safe. So there, it becomes very important to pick the right dose not only for the sake of finding a right dose for the drug, but also because those patients are suffering from diseases that many times lead to that in a certain period of time. So as soon as they receive a therapy that is efficacious, better for them. So from an ethical standpoint, we have to try our best to come up with a dose that is expected to be efficacious in such populations. Yeah,

Aaron Harmon:

without thinking a lot about it. It sounds like a lot of work, especially when you say the regulators are asking or expecting it now. But I feel like if you do it, well, it will actually speed up the work and help you get to patients much quicker.

Unknown:

It does. Yeah, it does. It tremendously improves the time to reach efficacious those and our ability to have a successful phase three trial. So prior to like, if you look back in literature a few years ago, and try to see the attrition rate in phase three clinical trials, a small percentage of drugs that entered phase two trial would make it into a drug. Now that attrition rate has reduced because we are getting better and better at predicting what dose and what regimen is expected to be safe and efficacious in phase three clinical trial. So the major reason why drugs fail in phase three trials is either the drug is not efficacious, or it is not safe. Now, if we are able to predict that invalid in advance using this mathematical modeling, then we can have our bets. And our guess not just based on the prior data, but some kind of predictive algorithm that would say, Hey, if you went in a larger population, if you conducted this trial, backed by the simulations that we have, this is the predicted outcome. This is the probability of having safety events in this much percent of population. Is there a fraction of population that is more susceptible to safety events? And can we enrich or can we label that is selective, that could have a slightly different dose recommendation for that population where there is adverse events expected. So all those have really come very useful when it comes to getting a drug for a disease. So many potential drugs that might have not gotten approved previously. In absence of the science now, they have a chance. So

Diane Cox:

like Aaron pharmaco metrics is kind of a new term to me. So is this a growing field? Is it pretty small at the moment, I guess, you know, I'm wondering if all vaccine companies have someone like you? I would hope so.

Unknown:

Yeah. So if you went to around 2008 2009 timeframe, if someone there was, I don't think there were many programs. And even now, there are not many programs in academia that have specialized courses in pharmacometrics. But let's say if someone came out with a degree in pharmacometrics, in 2008 2009, there are not many jobs around for that individual. But now is completely different now, regulators are asking for it. So companies need to have this function built in every, every company at a clinical stage or nearing entry into clinical stage has this function, like many, many small companies would try to outsource it, but in small or medium sized companies, there is a function, which is called either pharmacometrics clinical pharmacology, or modeling and simulation.

Diane Cox:

I just I honestly, I have some curiosity. So what was the target disease for this particular vaccine? I'm not familiar with the name of it.

Unknown:

This DGN put in? Well, yeah, so it in non clinical studies, it activated direct cells. So T regulatory cells, if they're activated, then they can be used for treating a variety of autoimmune diseases. That's right. So that could could be arthritis, that could be diabetes, that could be a bunch of the glomerular, glomerular, nephritis, glomerulonephritis. So after this incident, a Russian company, the strategic neuron went out of business, and a Russian company took it over. But that molecule, and they quite a bit of work, and they did the right passes to find out what the efficacious and safe dose should be. And they came up with 1000 fold lower dose than what was tested in the first Newman trial. So those that these guys got the six individuals God was point one mg per kilo, the Russian company started with a dose of point one microgram per kilogram, they actually conducted a clinical trial phase one and now there is a phase two ongoing of the same model. At that low dose of point one to seven day, they could go up to seven micrograms per kilogram, that seven was still 15 fold lower than the those given to those six individuals. So and they are seeing the effects that were expected from this super agonist. So they say selective T reg activation, and that could be useful for treating autoimmune diseases. It's about those selections. So getting the right dose there, they went with a higher dose and this disastrous event happened. But if a lower dose of the same drugs drug is it is about how much so how much is good. How much is very important when we think about drug development? Kind of like a

Diane Cox:

happy ending? I'm Aaron told me that these six individuals ended up being okay. Is that? Is that your understanding as well?

Unknown:

Yes. Yeah. So okay, all of all of these six made it there, they're alive. Some of them have even kids, those who did not at the time of the seven. Yeah. And I think the doctors there did a great job in preventing this because they could identify there was a cytokine release syndrome based on the symptoms, which can be difficult given the novelty of the drug. And they treated these patients with very high doses of prednisone, or corticosteroids, and which actually make them blow up and become like, there was a term called Elephant Man that was that made it to the headlines. But that was because of those fluids that they gave to prevent any toxic wastes formed from the cytokine release syndrome.

Diane Cox:

Fascinating. Wow. Thank you so much for sharing that.

Unknown:

No problem. Pleasure.

Aaron Harmon:

Great to have you on the show.

Unknown:

Thank you.

Aaron Harmon:

Now we'll take a quick break to hear from one of our sponsors.

Unknown:

Today's startups become tomorrow's growth engines in South Dakota, we're entering a new stage of expansion for our biotech industry, and you'll want to be part of it. Hi, I'm Tony Johnson, Executive Director of South Dakota biotech, where the state affiliate of the International bio organization and we're proud to be leading a state that's driving innovation to feed, fuel and heal the world. South Dakota biotech is here to inform, to connect and to advocate for our critical industry. Whether you're directly involved in biotechnology, or looking to learn more about it, we want to hear from you. Find us at WWF via that SD bio.org. Now back to the show.

Aaron Harmon:

To continue the discussion on TGN 1412. We have Kelly Creighton, VP of global CMC with clinic pace with us. Kelly has years of experience and is an expert in biologics. Welcome to the show, Kelly. Thank you. Pleasure to be here. So first, can you tell us about yourself and how you ended up in the CMC world and maybe even what CMC

Unknown:

careers. So yeah, my foray into the CMC world as kind of a unique path after my postdoc, I decided academia was not in my future and decided I wanted to move in back into the pharma industry, I had experience with that through the animal health side, during grad school. And through a stroke of luck, I was able to sign on with a small group in Colorado that was doing technical writing for a drug application that involved two different recombinant proteins. And they need a technical writers to assist with that drug application. So because of my strong background in molecular biology, reached out to them, they signed me up, I worked on that drug application, which ultimately got approved. And then from there just continued to get involved in the manufacturing side primarily in the biologics around but I also work in small molecules. But then also worked in like the quality assurance side for a few years, before really getting back into CMC, regulatory, and of course, CMC stands for chemistry, manufacturing and controls. It's really the process of which we manufacture both drugs and biologics to ensure routine quality and safety of those products on a day to day basis. So that's how I got into this realm.

Aaron Harmon:

So is it just seeing people go on their journey through their careers and end up where they are?

Unknown:

Yeah, when I was in grad school, I never thought I would end up in CMC regulatory, but it's, especially in the biologics realm, it's, you know, there's always emerging technologies and biologics. When I first got in, I worked on one of the very first cell therapies to ever be approved in the market. And 20 years ago, that was a very unique type of product, right. And now we see a ton of cell therapies from car keys to NK cells and all types of autologous and allogeneic stem cell programs. So, so it is, you know, there's always emerging technologies. Now you look at CRISPR, you look at like the mRNA vaccines, right? There's always these very unique emerging technologies, especially in the biologics realm. And that's an exciting area to work in

Aaron Harmon:

TGN 1412 shows us a few things. First, how important it is to know the mechanism of action of a drug or biologic. And then second, how important is to build safety into clinical trials? How have you seen the need for understanding mechanism of action spill into the CMC side? Yeah, it's

Unknown:

a great question, you know, from the CMC side, on the biologics, there's always been a requirement that you have some type of bio assay or functionality assay. As part of your routine release testing for Biologics projects. This is not a requirement for the small molecule side, only for biologics, projects, products. And so in reality, that need has always kind of been there. But what we're seeing a big change in over the last few years, is FDA really pushing that these bio assays or these functionality assays actually correlate now to clinical efficacy. There are a lot of bio functional assays out there that show that you can you know that the molecule or the antibody or, or whatever type of biologic product is, has a certain characteristic on a lot per lot basis. But a lot of times those assays are not potentially directly related to the primary mode of action that correlates to clinical efficacy. And so FDA is now really pushing that on the CMC side, when we're developing our CMC programs that we actually have bio assays now, that are correlates of clinical efficacy, that directly are supposed to directly show that from a lot per lot basis, that that law has the same characteristics, ergo should have the same efficacy profile, once you get a commercial setting, as it did when you're going through your phase one, two and three studies. So we are seeing this approach now from FDA, and it's gotten to the point to where now, if you cannot have a single assay that is a direct correlate to efficacy, FDA is requiring an authority orthogonal approach where you may have to want have two assays that together kind of demonstrate that the functionality of the assay or the molecule should have the same color clinical efficacy. So that's something that is coming or has been on horizon and actually just this past fall FDA told a company that is developing a tumor infiltrating lymphocyte product, they cell therapy product, that they could not file their biologics license application because of their bio assay was not adequate. So their bio assay was because it's the product is is really to elicit an immune response against cancer, their bio assay was to just show that product could really could cause a release of interferon gamma. Well, FDA said that's not good enough for a product of this type to ensure that it would demonstrate the same clinical efficacy as the products they saw through their phase three clinical study. So they pushed them back and said they could not file their VLA and had to develop additional bio assays that again, represented the primary mode of action that correlates to the clinical efficacy. So it is a big hot topic item within FBA. We've seen it for years. Now. What's very interesting, though, as you're developing biologic products, is unlike most small molecules, the mechanism of action may not be straightforward. biologic products can be very complex, and may just be the trigger of a cascade of effects within the body. And in those cases, right, it's very difficult to truly understand what are those key characteristics that trigger that cascade? To deliver that efficacy, if you consider different types of monoclonal antibodies, like the TGN 1412, right, that was directed at a very specific antigen on B cells. Once that binding occurred, that actually created a physiological cascade effect, which resulted ultimately right in this adverse event with the cytokine release storms. Again, because that's a very complex cascade of event. Truly understanding what in that cascade would deliver the actual primary mechanism of action for efficacy can be very difficult to tease out out of that cascade. That's different from like a monoclonal antibody that may be derived directed at a very specific antigen, you'd look at a lot of the migraine products that are out there, that are monoclonal antibodies directed at calcitonin, Gene related protein, basically, those monoclonal antibodies, or bind that protein, and neutralize it and prevent it from binding to its receptor that causes its cascade of events that causes migraines, it's a lot easier to show that your monoclonal antibody binds to an Amgen like that in those cases and prevents it binding of that antigen to x receptors, versus, you know, a stem cell therapy, where when you inject the stem cell therapy, again, right, typically, that those mechanisms actions result in a cascade of events. And it's very difficult to know, patient to patients of the subject, if those cascade of events are the same, or if there may, or there may be variations within patients. So really defining what the pmla for some of these biologics becomes very, very difficult. So when we're going through clinical development, even early on, right, because those are not well defined mechanisms. Typically, for some of these products, what we encourage or what I encourage a lot of my clients to do is to go ahead and take a very broad approach and look at several different characteristics and start collecting data on all of those bioassay or functional characteristics. And just start collecting that data kind of FAO as you're going through clinical development and build this database and then start looking at your efficacy signals and seeing if you have different efficacy in your trials based off different product lots that were used, and then trying to correlate different assay readouts to clinical efficacy, so that you can meet this new, you know, this higher FDA requirement that your bio functionality assays actually correlate to your clinical effect. So, it is a complex process, and most companies are not really in place to have these bio assays, really a good bio assays in place until you enter very late phase clinical development, typically phase three, study. So pmla is is a lot different on the biologics side. But and actually probably at this point, FDA has put in larger requirements for at least trying to understand that pmla On the biologic Your side than what we typically see on the small molecule side.

Aaron Harmon:

So I've heard a number of times, we know it works, but we don't know exactly why. But we know it works. And understanding the why isn't really important for all these assays. But also if there's cascade effects, having a safety plan in the clinical trials, so that you know how to respond to it exactly. In the case of TGN 1412. The physicians knew how to handle cytokine storm that was in the safety plan. So there was some preparation for that, and they were able to treat the patients and yet recovered.

Kelly Creighton:

Yeah, as you're going through clinical development like that, for most products, there's always a theoretical mechanism of action. And that's what most people plan on and, and when you're going into like your first in human studies, you base your safety plans around what you the least propose that mechanism of action is. And so yes, a lot of these products right now we're getting more and more of them on the market, monoclonal antibodies have been around for a long time, a lot of the cell therapies have been long around for a long time. You know, we solve this kind of the same issues with some of the the first car T products that hit the market with regards to big cytokine, Storm adverse events, right, because basically, you're taking immune cells, T cells and trying to directly target them against tumor cells. And you think about that, one of the things that, you know, becomes an unknown with regards to the potential safety impact is, you can imagine, if one patient has one type of tumor burden, patient two has a much higher tumor burden, you give them that same product, well, that product may have a much higher risk of triggering a cytokine storm, in patient two, just because of the tumor burden in patient two vs. One, so you have to be aware of those kinds of differences. Even you know, even if you completely control the quality of the product from the CMC side, when going into different patient populations such as that you always have to be aware of these types of products that just because patient to patient differences. And depending on what that the target is for that biologic product, that you could have a different safety risk going in. So it's one thing and the industry is well aware of it. And they do put controls around that to try to understand right to have inclusion and exclusion criteria and not allow folks that may have a very high tumor burden that Ergo would be at a higher risk for a cytokine type storm adverse event. And they you know, not allow them to be enrolled in those studies. It's an evolving industry, from both, you know, the the manufacturing side, but also from the clinical side. And every year we learn more and more about how to appropriately approach these types of studies to try to minimize risk for the subjects involved.

Diane Cox:

So you mentioned a lot to lot variability, and what are potential causes for what could happen during manufacturing? I mean, I understand general manufacturing always has some variation, but what are some in particular for biologics? That would cause some variability?

Unknown:

Yeah, so of course, biologics, normally the manufacturing process for the biologics are extremely complex, they're not what I consider the right the straightforward cookie cutter type of approaches with small molecules, which is normally synthetic type of chemistry, right, your organic chemistry, your actions and things that you can understand and control very well for biologics, right. Now, a lot of biologics are derived out of living systems themselves, either through from cell culture as are derived from animals or, or the product our cells themselves. And so because of the complex nature of the manufacturing process, and really more difficult ability to control the process itself, that is where lot to lot variability content can creep in. And, and there's always there's an old mantra in the CMC biologics realm, that is the process is the product, right? So whatever that process is, is going to result in what that final product is, and even having very small changes in a process somewhere, can result in a big difference in quality of a biologic product. And so that's why typically, the testing regimen for biologic products are a little more a yes. There's a higher bar Tick, tick layer with regards to to testing quality of biologic products. The flip side of that, though, is that most biologic products are not as easily characterized as small molecules. So small molecule, right, we can very clearly get down and show down at the bond, you know, single bond level that we have the right, the right molecule, biologics, we just we don't have that ability to do that very high level of characterization on each given lot. So therefore, we do have to rely more on controlling the process to ensure that we have batched batch quality. And then we have our subset of analyses like our bio assays or functionality assays, that again, then demonstrate that the molecule does have at least the characteristics that you expect that So, it is difficult from the biologic side, especially early on to ensure lot to lot quality. And then, of course, as you're going through phase three studies and the scale up that you normally have to do to prepare for your commercial supplies, there are a lot of requirements to demonstrate that the product from a small scale manufacturing to large scale manufacturing has the same level of quality and the exact same characteristics. And as you usually required very extensive comparability type studies, in order to demonstrate that your early phase one small scale material is equivalent to your Phase Three proposed commercial material.

Diane Cox:

This is why I'm in medical device.

Aaron Harmon:

So, Kelly, if someone came up to you, they're just starting out a small company on a biologic. Do you have any tips for them that'll make their lives easier when they get to the manufacturing stage?

Unknown:

Yeah, well, right. I mean, it's, I wish I could tell him, you know, Hey, make sure you spend adequate money doing your CMC development and analytical development. But right, the reality always isn't an industry is that for these early development programs, getting into the clinic is really the key milestone. But the thing we are kind of pushing and keep pushing folks that are in these early development stages, even if you don't set up some of these assays as your release assay, try to gain as much product knowledge as you can, throughout the process. So do as a much additional characterization as you possibly can on the product starting from day one. Because the more data you have, starting from your early phase, you're going to be in so much better a place, once you get to your phase three and go to then try to get your marketing approval to be able to support that you have the right controls that you have the right assays, if you wait and try to start gathering that data, you know, once you're in your phase three study, you're going to be behind the eight ball. And unfortunately, that's what we're kind of seeing with some of these companies like the company does pass ball that was told to not even file their VLA yet, because they just don't have the appropriate bioassay. So right now, it's kind of knowledge is power, again, right back to the old message that we all saw on TV as kids, because it really is, the more you understand your product and the characteristics of your product early on, the better off you're going to be. Once you get in to your later phase studies. And again, it also helps when you do have to do the potential bridges, because of scale up, right, the more you know, early on, it's easier to demonstrate comparability for a product you know a lot about, versus one that you have a smaller data set on. So that's kind of the key right now is it just test it to the nth degree the best you can and start building those databases.

Aaron Harmon:

I think that's especially rings true with TGN 1412 More data early before that phase one trial, but it also given them clues that the dosing may had been much lower and could have prevented that. Yes, which happened. So Right.

Unknown:

Hindsight is always 2020. Right? We're always learning things and, and that's the nice thing is that we can take those lessons from like TGN 1412, and apply them now to to not only monoclonal antibodies but other biologics as they emerge and understand, okay, where are our risk and what do we need to be doing now, so that we don't have those things same type of issues pop up.

Aaron Harmon:

Thank you for being on the show, Kelly. Yes, thank you. And thank you for listening. I look forward to bringing you another episode of Inside Out quality.

Diane Cox:

We hope you enjoyed this episode. This was brought to you thanks to South Dakota biotech Association. If you have a story you'd like us to explore and share, let us know by visiting www.sd bio.org.

Aaron Harmon:

Other resources for quality include the University of South Dakota's biomedical engineering department where you can find courses on quality systems, regulatory affairs, and medical product development. Also, if you live in the Sioux Falls area, check out quit a local Quality Assurance Professionals Network. You can find out more about pivot by clicking on the link on our website to the end and I would like to thank several people, but a few who stand out are Nate peple for a support with audio mixing Barbara Durrell, Christian or support with graphics design and web And lastly the support from South Dakota bio