November 14, 2024
Host: Oliver Ball, Director of Business Development, Dark Horse
Presenters:
Opening Remarks
Oliver Ball: Okay. Hello, and welcome everybody to the next unbridled excellence webinar. I'm your series host, Oliver Ball, Director of Business Development for Dark Horse.
Particularly special welcome to you if this is your first time joining one of the Dark Horse Webinars. This is now the 10th in the series, and we established this series about a year ago, to really try and disseminate some of the experience that Dark Horse has generated now over our 10 years of working in the cell and gene space.
Over that time, I think some of you might have heard we've now had, I think, over 450 clients across in the field. So we really just wanted to set up this webinar series as an opportunity to kind of share some of that experience and disseminate best practices and give you guys a little bit of a taste about the kind of things that we work on in the practice.
So today we're lucky to be joined by 2 of our senior consultants, Christina Fuentes and Elizabeth Figueroa, and they are going to be talking to you about a topic which I think has been of increasing interest in the field, and I know is going to be something that many of you are working on or interested in, which is gene editing across non-viral and viral methods.
And so in this talk Christina and Elizabeth are going to be giving a bit of an overview of the field, the current state of the art, and where things stand in clinical development, commercialization of gene editing platforms, and then go into a little bit more of the specifics of what a sort of typical development pathway looks like for these modalities, and some of the sort of unique and special considerations that apply to gene editing.
And before we get into that, I just want to highlight that we will be having another webinar in January which will be focused on optimizing facility operations as a means of reducing cost of goods. I know cost of goods reduction is something that everyone is kind of having at the forefront of their minds at the moment, and optimizing facility operations is a really powerful way of doing that. So keep your eyes out for that webinar coming up soon.
We'll also have potentially some time later in today's webinar for a Q&A. So if you have any questions that you want to submit, you can do that through the Zoom interface and through the chat function. And when we get time at the end of today's talk, Christina Elizabeth will be happy to field your questions. So put those into the chat as we go through the webinar, and we can get to those.
Finally a reminder that the content today will be available for on-demand viewing. So if you want to share the content with your colleagues, or you want to watch it again later, you can go to our website in the days after this webinar to access it on demand.
So with that I will invite Elizabeth and Christina to start the main content.
Presenter Introductions
Christina Fuentes: Fantastic. So thank you everyone for joining. I wanted to start by giving you a little bit of background. Really a key motivator for this session was the fact that we are increasingly seeing the key role that gene editing is playing in creating the next generation of cell and gene therapies.
Today's talk is going to cover the development of these next gen therapies from both a viral and non-viral perspective. And we'll give a focus, particularly on in vivo applications as that continues to be an emerging space.
Now, before we kick off, we want to briefly introduce ourselves.
Elizabeth Figueroa: Hi, everyone! Good morning. My name is Elizabeth Figueroa. I am a senior consultant who joined the practice in 2021, 3 years ago. My background is as a bioengineer by training. I defended a PhD thesis at Rice University on a novel nanoparticle based gene therapy vector for gene modified cell therapy and cancer vaccine applications.
And prior to joining Dark Horse, my most recent industry role was as associate director of process development at a transposon, or rather transposase gene modified CAR-T and TCR-T cell therapy company, called Zyopharm Oncology.
And since joining Dark Horse, I've had the privilege of applying my expertise as it relates to non-viral nanoparticle, mediated gene therapy as well as gene modified cell therapy, process development, manufacturing and regulatory strategy.
Christina Fuentes: And my name is Christina Fuentes. I'm a senior consultant. I've been with Dark Horse for 4 years now. I'm also a bioengineer by training. I received my PhD at UC Berkeley, where I worked on a variety of projects, including engineering a self inactivating AAV CRISPR-Cas9 system for viral mediated, transient expression of Cas9 in vivo.
Since joining Dark Horse, I've supported a range of product types, including a specialty in viral vector gene therapies for gene augmentation introduction as well as gene editing. I've supported a range of areas across process development, manufacturing and regulatory strategy of which we'll highlight some of that in today's talk.
Session Overview
Christina Fuentes: So this session will be in 3 main segments. We're going to go through gene editing technologies, clinical progress of these technologies, and finally, talk about how you create a modular roadmap to IND for gene editing products and be able to then understand what are some of those unique considerations when developing a gene edited based product.
Gene Editing Technologies
Disease Pathology and Gene Editing Potential
Christina Fuentes: So the underlying genetics of a disease, as you can see from this pie chart vary and can range from insertions, deletions as well as point mutations. Gene editing has the potential to expand on the number of types and types of diseases that can be treated because of its ability to precisely target regions and specifically induce a particular type of edit, as we'll cover shortly.
Now with traditional cell and gene therapies, such as AAV that introduces transgenes for augmentation, and focuses primarily on autosomal dominant disorders. Gene editing has the potential to address both autosomal dominant and autosomal recessive disorders by being able to edit both alleles.
CRISPR-Cas9 Mechanism
Christina Fuentes: The most well-recognized gene editing tool is CRISPR-Cas9. It's a nuclease or Cas protein that cuts DNA paired with a guide which tells it where to go within the genome.
So how does gene editing work? Well, it introduces a break, and when that happens, the cell repairs a break mediated through 3 different pathways. Here we're showing 2 of the primary pathways NHEJ and HDR.
NHEJ is error prone, so that continuous double-stranded breaks can allow for insertions deletions which ultimately can disrupt a protein's reading frame and knock out a gene, whereas in HDR in dividing cells, if you introduce a template with your editing tool you're able to knock in a gene.
Because CRISPR-Cas relies on breaks in the DNA, this poses a particular safety concern about genome integrity. This can be due to chromosomal rearrangements or knockout of critical genes that are a result of off-target editing. This is why there are emerging tools that do not rely on double-stranded breaks as we'll discuss in the next slide.
Gene Editing Tool Landscape
Christina Fuentes: So here we're showing that obviously, Cas9 is not the only tool available for gene editing. And, in fact, when we look at double stranded break mediated editing tools on the left hand side, there are several types of nucleases that have preceded Cas9: meganucleases, zinc fingers and TALENs. The main difference of these nucleases, compared to CRISPR-Cas9 is that they require protein engineering to target specific sequences.
This is why Cas9 has become widely known and utilized because it does not rely on protein engineering, and instead can rely on being readily targeted through changes in the guide sequence or the targeting component.
Now, as we talked about, given the safety considerations and concerns about introducing double-stranded breaks, we're also seeing other editing technologies, some of which rely on modified CRISPR-Cas9 systems to edit the genome without relying on these breaks.
There's base editing, prime editing for nucleotide substitutions for insertion of small sequences respectively. They both rely on modified forms of Cas9 that have another protein fused for it to give the desired function.
There's also transposons. We've known about transposons since the 1940s. It's been around a long time. But we're seeing a recent interest in this technology, particularly because it has the potential to enable insertion of large sequences.
And then we have epigenetic editing, for example, being able to methylate promoter regions to impact the transcription levels, to modify what gets transcribed, and proteins get expressed without inducing any breaks or edits to the actual genome sequence.
Tool Selection Criteria
Christina Fuentes: Now, with all of the tools available in our toolbox, there are many potential approaches you can take. And there's typically several criteria that can use to help guide that selection.
The first is understanding the underlying pathology of the intended population which tool can achieve the desired edit.
There's specificity. How well does the tool target the intended sequence? What's the risk of off target edits?
There's also fidelity. So how accurate is the gene edit, for example, for gene insertion. What is the likelihood that the insertion is within frame to have a functional gene and activity following that.
And then, given the existing delivery vectors out there, do these systems fit within the size constraints? The image below shown here just shows some example technologies for genome modification and its size relative to the capacity of the viral vector AAV. And you can see some of these are beyond the capacity of AAV, which would then require either an alternative delivery platform, a dual AAV approach where you split the editing tool, if at all possible or potential redesign.
The important piece here is to understand that you can have the best tool available, but it has limited utility if you're unable to deliver it to the correct site, as Elizabeth will cover next.
Delivery Considerations
Elizabeth Figueroa: Thanks, Christina. So as you've so nicely laid out, the design space for gene editing tools is really large, and it continues to grow. But any gene editing tool is going to depend upon delivery to bring it into proximity with its target site in the genome.
And so when we think about delivery, we can think about 2 key elements, and these are broken out in the slide on the left of the slide in the boxes. The first key element is whether the tool, the gene editing tool will be delivered to cells in vivo or ex vivo.
And ex vivo gene editing represents the stepping stone that has brought gene editing technologies into the clinic. With the December 2023 approval of Vertex's CRISPR-Cas9 edited hematopoietic stem cell product for sickle cell and beta thalassemia.
This was obviously a huge milestone for the field of gene editing, but it is associated, as are all ex vivo modified approaches with high cost, manufacturing complexity as well as clinical challenges, such as the need to perform myeloablative conditioning of the patient prior to infusion.
And so, for these reasons, in vivo gene editing really represents the future state of the field, and then how we'll extend the reach of the current gene editing tools. But in order to accomplish in vivo gene editing, we need to find a precise, efficient, and safe delivery platform. And so that brings us to the second key element of delivery which is the identification of a gene delivery vector.
And so, traditionally, we see these broken into 2 categories of viral and non-viral. You may also consider physical methods which we won't cover for today. But in this image on the right you'll see illustrated in this schematic a lipid nanoparticle containing a Cas9 mRNA and your single guide RNA, as well as an adeno-associated virus with a CRISPR-Cas9 system. And these 2 images kind of represent the representative viral and non-viral approaches that we'll be highlighting today.
Historically, the field of cell and gene therapy through pure play gene therapy as well as gene modified cell therapies has really advanced through the application of viral vectors, especially AAV in the clinic. However, with the advent of gene editing tools, we've also seen in parallel the advancement of non-viral vectors like lipid nanoparticles entering into the clinic as well.
Viral Vectors (AAV)
Elizabeth Figueroa: And so first let's discuss AAV vectors a bit more as they represent the most clinically advanced in vivo viral vector.
The image on the right represents something that Christina alluded to earlier, which is a dual AAV approach. And here one of the viral vectors is being used to deliver the meganuclease cargo. And the other vector is being used to deliver a donor DNA template to facilitate a gene insertion.
When we think about the advantages of viral vectors, one of the key benefits is that viruses have evolved to have the ability to target specific cells. So by selecting your specific viral capsid variant, you're able to confer tissue specific targeting attributes, which is really important when we think about in vivo delivery.
In addition, the non-clinical and clinical safety profile of AAV vectors is relatively well characterized, due to their long history of use in traditional pure play gene and gene modified cell therapy applications.
And so they represent a more well understood development and regulatory pathway from the delivery perspective given their clinical and commercial use in cell and gene therapy.
However, a concern that's really relevant specifically in the context of gene editing is that the AAV vector is able to persistently express genes in cells. So you can imagine this presents safety concerns when we consider the potential impact of persistent expression of a nuclease.
AAV's expression can also be limited by an immune response to the capsid. And so this has implications, both in terms of redoseability of this vector as well as posing challenges when we think about identifying patients that do not have pre-existing immunity against the AAV vector.
And additionally, as you may recall from a couple of slides back, the cargo capacity of AAV can be limited in the context of some of the larger gene editing tools, such as base editors and prime editors.
Finally, as many of you are likely well aware, the manufacturing complexity and costs associated with GMP production of viral vectors, including AAV can be quite burdensome.
Non-Viral Vectors (LNPs)
Elizabeth Figueroa: For non-viral vectors, as I mentioned early, we'll specifically focus on lipid nanoparticles or LNPs. These are essentially spherical fat particles that encapsulate cargo within their inner core. They're typically comprised of 4 types of lipids. And the secret sauce of these formulations is really the cationic or ionizable lipid which really influences the safety and efficacy profile of these vectors.
As shown in the figure on the right, LNPs afford a great degree of flexibility, both in terms of the cargo types that they're capable of delivering. We see demonstrated here at plasmid DNA, as well as mRNA cargo, as well as a ribonucleoprotein complex, so protein cargo, but they also confer benefits in terms of their cargo size capacity, especially relative to an AAV vector.
They're generally associated with a favorable safety profile that facilitates redosing. We've seen this demonstrated with the Onpattro siRNA LNP particle as well as recently, Intellia demonstrated this in their late stage clinical trial.
And importantly, the manufacturing complexity and cost associated with LNPs is significantly better than with their viral vector counterparts.
However, looking at the cons box on the bottom left, LNP biodistribution and efficacy in vitro and in animal models is not always a true recapitulation of what their behavior will be in humans. This can really make formulation screening quite laborious at early stages of development and can be a challenge.
Additionally, LNPs are really really good at getting to the liver in vivo. But if you want to go somewhere else, if you want to target extra hepatic tissues in vivo, there's going to be a lot of formulation development and optimization that's needed, for instance, modification with targeting ligands to alter their biodistribution.
And they typically have lower DNA expression efficiency relative to AAV vectors.
And finally, there can be supply chain complexities that can hinder product development due to a complex intellectual property landscape, both for the ionizable or cationic lipid components themselves, as well as the resulting LNP formulation.
Combined Considerations
Elizabeth Figueroa: So kind of coming full circle at the end of this background section of our presentation, I just want to emphasize that a genome editing drug product profile is going to be dictated by the combined impact of both the gene editing tool as well as the delivery vector.
And just to highlight that for any given application, multiple tools and delivery approaches will exist, and they just each come with their own pros and cons. So, taking a look at the figure on this slide, which was generated by my illustrious co-presenter, Dr. Fuentes.
We can see that some of the ways that we approach selection of the tools can be driven by the type of genome modification that's desired, and the delivery approach can be guided by the desired expression profile, the cargo size, the target tissue, and route of administration, as well as many other parameters. So the take home message here is really your gene editing tool and your delivery vector profiles should be considered really carefully early in development, as they will really inform your non-clinical CMC and clinical aspects of your program.
Clinical Progress
CRISPR-Cas Gene Editing Progress
Elizabeth Figueroa: So now, having laid a foundation for the selection of your tool and your vector I'll spend a few minutes reviewing the preclinical and clinical progress.
I really like this graphic from the Innovative Genomics Institute. It was published in March of this year, and it depicts CRISPR-Cas gene editing tools across on the horizontal axis, early phases of clinical trials to commercialization, and along the vertical axis the disease indication for each category.
And one note for this graphic is that base editing was considered a version of CRISPR-Cas technology. So those products are included within this infographic.
You'll note immediately we have one approved product which we mentioned earlier was the December 2023 Casgevy product.
And also interesting to note, and this is something that may not be readily visually apparent is that for all of the development candidates encompassed within this graphic, including the late stage developers Beam, Intellia, Editas, CRISPR Therapeutics.
These are either ex vivo modified or liver targeted in vivo gene editing products. And so this really highlights again, that point that in vivo targeting of extra hepatic organs is still in earlier stages of preclinical development. And that's really driven by delivery challenges.
Non-Double Strand Break Tools
Elizabeth Figueroa: This graphic in contrast with the previous graphic addresses preclinical and clinical progress of gene editing tools that do not induce double-stranded breaks which you'll recall from Christina's introduction may confer an improved safety profile. So these kind of represent a more emerging space within the gene editing tools.
The stage of development is demonstrated from left to right, and then we have information on the vertical axis about the number of developing therapies and indications, as well as a breakout of in vivo versus ex vivo.
And so you may note immediately off the bat that there aren't any later stage clinical products in development, and reflecting that a lot of these non-DSB inducing tools are earlier in their development lifecycle. But we do see that about half of the 22 or so products depicted in this graphic are in early stage clinical development, and that as well in vivo delivery actually represents a fair chunk of these products. So for these emerging gene editing tools, just an area of focus.
2024 In Vivo Progress Headlines
Elizabeth Figueroa: And so kind of on the note of in vivo progress. Here we've included a few representative headlines just from 2024 alone.
Of the slide, and AAV mediated delivery gene editing products on the right side of the slide. Kind of walking through this we have LNP based delivery by Beam Therapeutics looking at liver targeted delivery of base editors for α-1-AT deficiency.
We have Intellia. This is one representative headline. They're using LNP delivery of CRISPR-Cas to the liver for late stage clinical trials towards amyloidosis and angioedema, as well as early stage for α-1-AT deficiency as well.
And we have Verve Therapeutics using a modified LNP formulation for delivery of base editors. Interestingly, for cardiovascular disease or hypercholesterolemia specifically.
On the right side of the slide, we are highlighting the Arcus meganuclease technology, which is being leveraged by both Precision Biosciences and Eocure. Precision is targeting the liver for chronic hepatitis B indications and Eocure is using AAV to target the liver for OTC deficiency.
So with that kind of preclinical and clinical landscape overview, I will pass the baton back to Christina to walk through a high level drug development plan and roadmap to IND.
Development Pathway and IND Roadmap
High-Level Drug Development Lifecycle
Christina Fuentes: Fantastic. So before we talk about the roadmap to IND, I always like to start with a high level view of the drug development lifecycle and how that aligns with FDA engagement.
Preclinical, which is the focus of today's discussion, sits between discovery all the way to commercial. And what we're showing here is an idealized path, because cell and gene therapies, for example, typically have phase 1/2 studies combined and a lot of steps that are mixed in within each of these boxes.
There are several opportunities to engage with the FDA, and these are shown in the orange boxes. Here we'll talk a little bit about INTERACT, pre-IND, as those are some key early engagements with the agency before going to the IND stage.
What's important to understand here is that the preclinical package is really key to initiating clinical testing for many. This is the first meaningful assessment of your product from an external party, so that means other than your company or your investors.
And for the agency in this example, we're talking about working with the FDA. The FDA's objective in reviewing the IND is to assure the safety to patients, and that sound, scientifically driven evaluations are made.
Preclinical Objectives
Christina Fuentes: When we look at preclinical, really, at this stage, the objective is to establish the product's activity, profile, dosing safety to patients and relevant clinical monitoring. That's what these 7 objectives defined in the guidance from the agency are meant to provide.
And I just want to remind everyone that what we're showing here again with cell and gene products, they're complex. They're often at the cutting edge of the technologies, pushing the envelope of therapies. So the path isn't always straight, but I hope by the end of this talk you have some principles that can help you navigate the path.
Modular Roadmap Framework
Christina Fuentes: Here, what we've done is zoomed in and defined some of the sub stages within the preclinical stage of that pathway.
We're showing an exemplar path to IND covering key stage gates for a minimum viable product. This is meant as a framework from which you will be able to continue to build and tailor specifically for your own product.
Our focus today is going to be on non-clinical and CMC, and we're going to particularly emphasize some unique considerations for in vivo applications.
When we look at the non-clinical pathway, it's broken down into stages of identifying the target, selecting a lead candidate and starting to profile the drug candidate before ultimately running those key studies critical to your IND package.
Here we show in parallel the progressive maturation of your CMC that should allow the that should be performed with your non-clinical, so that you have what you need to create a successful IND package.
Above in the orange boxes, again, we've highlighted some key points and common ways to engage with the agency ahead of IND. We highly recommend this because early engagement reduces non-clinical CMC and clinical risk.
Importantly, INTERACT is not always granted. But in the case for novel technologies, where there is limited precedence, or where you have unusual or unique considerations in development which would otherwise delay the IND timeline if you didn't gain feedback. That's where you have a good chance of being granted an INTERACT. And I think this is pretty well suited, particularly when we think about in vivo gene editing applications.
Stage-by-Stage Development Considerations
Target Identification Stage
Christina Fuentes: So here, what we've done is we're going to break it down from each of the steps. This is the target ID stage. And really the key stage gates for minimum viable product is to demonstrate the desired modification can be achieved in vitro. And you want to do this in a human genome background, as that's the ultimate genome background that's intended for clinical application. So you represent your intended population.
And a nice to have is to show activity following an edit in a relevant cell type, whether that's wild type or disease.
The context for which you look at activity can be with your delivery vector or in the cases where the delivery vector does not work well in vitro, or you have trouble seeing your signal over the background noise alternative approaches using plasmids or mRNA transfection may be used as long as you understand what are the key takeaways and interpretations you can take out of that study, and how that relates to ultimately your representative product.
And on the CMC side, at this stage early research material is often used again, we're prioritizing time and cost. So, having a quick and relatively cheap process to get through your screening rather than having to have your ultimately intended clinical process at this stage.
In preparation for lead selection, there are some key CMC activities you can think about getting started that includes gene synthesis and plasmid production, as those are some of the key materials that will be used in the next CMC stage related to process and analytical development.
Viral Vector Considerations - Target ID Stage
Christina Fuentes: So what you'll see is that as we go through each stage, we're going to have an extra slide that talks about some of the unique considerations broken down by viral, non-viral and then the commonality with gene edited products.
So at the early stage of target ID, it's really critical to plan ahead as the decisions you make early on can have major implications later in development.
For example, if your payload is above the capacity of AAV, it can lead to truncations which impact the efficacy of your product, and it can take a major hit to your manufacturing yields, increasing the cost of goods. You'll want to assess at that point whether alternative delivery systems, a dual AAV process or an entirely different payload redesign needs to be considered.
Additionally when we think about your capsid selection for AAV, not only do you want to consider the transduction profile in your intended clinical population, but you'll want to be able to understand the transduction profile, the biodistribution in your potential model systems, and how that translates to your clinical population.
You might also want to consider the manufacturability, particularly for large indications and high dose indications where cost of goods is a big priority, and then the pre-existing neutralizing antibodies in target population should also be a consideration as that can impact the treatable patient population. Sometimes the pre-existing neutralizing antibody titer can be used as exclusion criteria clinically.
And finally, for your VG titer method, it's critical to understanding the concentration of product, especially as we move towards clinical use. So early development of this method is recommended so that you can trust you know how much product you're dosing in your experiments, and if done appropriately, you may be able to leverage this titer method both for product release as well as assessment of biodistribution in your model systems.
In some cases ITR titer based titers are used for early research so that you can get a single titer probe across product candidates. That's okay. But just understand, you will need a product specific titer method to get to IND. If you don't have it keep retained so that you're ready to test once you have that method in place.
Non-Viral Vector Considerations - Target ID Stage
Elizabeth Figueroa: Great and on the non-viral vector side for LNPs, this stage of research and discovery is all about gathering information to inform your LNP formulation, the challenge presented by LNPs. I alluded to this earlier. We don't have a great understanding of their structure-function relationship, and their in vivo behavior from both the functional and biodistribution perspective are not always recapitulated in vitro or even across animal models, including NHPs in some instances.
And so this makes formulation selection really challenging. If a developer is not licensing an existing optimized formulation as a starting point, especially when we consider how many variables there are to assess we have a composition that includes 4 lipids. We potentially have a targeting ligand. We have a nucleic acid drug substance. Maybe it's mRNA and guide RNA. All of these elements have molar ratios that can be adjusted within the LNP formulation.
So the first point that I want to emphasize at this stage of development is the importance of leveraging a high throughput in vitro and in vivo screening approach in order to help optimize the desired target product profile for your intended LNP formulation.
And kind of hand in hand with the challenges of LNP screening is the need to procure critical components of your formulation, which can include novel lipids, novel targeting ligands.
The LNP supply chain is still very nascent. It's far from commodified. So additional time is wise at this early stage of development to ensure that your novel excipients or ligands are manufacturable at scale, that you can attain a raw material with an acceptable purity profile, and with lipids specifically, this can be really challenging.
And then finally, from a manufacturing perspective, LNP process development, process scale up, GMP manufacturing, all of these stages there's less expertise in these processes across CDMOs. And so it's recommended to begin identifying and engaging with your potential CDMOs early in the development process.
Lead Selection Stage
Christina Fuentes: Moving on to the next stage, which is lead selection. At this point you're looking to identify a lead candidate. And so for gene editing ideally at this stage, you're running assays for off target identification to understand the activity of your gene editing tool as that can drive the ultimate decision, not just on the activity in your on target site, but the overall potential, the safety profile of your tool. So that impacts lead selection many times. In these cases stage gates are showing no major patient safety risk due to the gene edit, the feasibility of your route of administration and dose, proof of concept efficacy, and also, should you desire to move forward with an INTERACT, you should have some pilot nonclinical data, as that's key to seeking alignment with the agency, especially if you are looking for model system selection where there's little or no precedence. So you want to have some data to be able to support and justify your stance on how to proceed forward.
Viral Considerations - Lead Selection
Christina Fuentes: Now for viral based products, you want to be able to confirm that the formulation and strength, in other words, the concentration is feasible from a CMC perspective. This is particularly important when you think about volume limited routes of administration, such as direct brain injection.
During process development, you also want to establish the scalability of process. If your process is not scalable for later stages, where the demand can be much higher, then you'll have to understand what are the implications for comparability later.
Non-Viral and Gene Editing Considerations - Lead Selection
Elizabeth Figueroa: And on the non-viral vector LNP side, a really similar theme to the point that Christina just made about AAV, and that's with respect to the challenges with manufacturing scale up, depending on the type of approach that's taken for LNP formulation. Some of these platform approaches are more scalable and others are less scalable. So, selecting a formulation, manufacturing approach that can be scaled is advised, and if that approach is not taken to be aware that the scale up can absolutely be non-trivial.
This is something to be really aware of and cognizant of, as you're engaging with your LNP CDMO, as we discussed a couple of slides ago.
I also want to highlight a few considerations for genome editing drug products that apply to both viral and non-viral delivery approaches, the first of which is the initiation of development of key safety assays.
This is really critical. These constitute probably some of the highest regulatory scrutiny that one might expect due to the potential for off target and genome altering effects from genome editing tools, especially when we're talking about in vivo applications.
We know from traditional cell and gene therapies that CMC related holds have historically taken the longest to resolve. It's unclear yet whether that trend will hold for genome editing drug products where we anticipate non-clinical data is going to be so paramount to demonstrating the safety of these approaches, and may actually present a larger hurdle for these complex therapies.
I would also highlight at this stage, the importance, in particular, for a multi-step or a multiplexed gene editing approach of developing several candidate potency assays ideally to probe each step of the genome editing mechanism of action.
And finally, while this is not a requirement, we recommend to begin transitioning key raw materials that may have potential impact on your drug product quality or safety earlier in the development lifecycle to ensure that your process development outputs are representative of your intended clinical product.
And so in the context of gene editing, you might specifically consider procurement of higher grade guide RNA, because we know that higher grade guide RNA can be associated with lower purity, which may be important to characterize the impact of earlier in process development.
Christina Fuentes: And I just want to say this contrasts with the thought, you know, traditionally, with viral vector mediated delivery. We often see this progression of research grade plasmids to high quality, to ultimately either sticking with a high quality, GMP-like plasmid or moving towards GMP.
In this case, for tox studies, you want to represent your intended clinical process and materials, and also in some way show the worst case scenario. So if your research grade guide has higher purity or higher purity, then it may not be as representative. So I think that's something a little bit contrasting to some of the parallels to viral vector.
Drug Candidate Profiling Stage
Christina Fuentes: Okay, so at this stage for drug candidate profiling, you want to define the therapeutic window durable response, lack of unacceptable toxicities within that therapeutic window, and how you define those unacceptable toxicities really is going to be driven by the intended clinical use as well as the benefit risk profile.
Non-clinically, you should be verifying your off target sites and designing IND enabling studies which can be shared at the pre-IND stage for alignment.
On the CMC side, you want to confirm CMC strategy, which includes the manufacturing process, your analytical strategy, as well as you know that can include your potency assay plans, and then to gear up for the next stage, it's having critical starting materials on hand, such as your guide, particularly if you want to use representative material for those IND enabling studies.
Viral Considerations - Drug Candidate Profiling
Christina Fuentes: On the viral side, if you're using large animals, you'll want to screen those cohorts for pre-existing neutralizing antibodies that's going to take you time. So this should be initiated well in advance of your IND enabling studies. You want to also start to plan for titer method qualification. It's best practice to use the same method in your IND enabling studies as what you intend to use clinically in order to be able to bridge the dose.
If you're not ready to qualify just yet, that's okay. Just keep retained so that you can test your non-clinical material with your intended titer method once ready.
Now, finally, viral mediated delivery raises concerns of persistent expression when you're targeting non-dividing and slowly dividing cells.
There's also a theoretical concern of AAV integration, especially when you have a gene editing tool that induces double-stranded breaks. It's analogous to thinking about the HDR pathway. We induce double stranded breaks to insert templates effectively here your AAV payload can act as a template for insertion at the cut site where you might be trying to knock out a gene not insert.
Additionally, germline editing also needs to be assessed, and this also applies to non-viral as well. The latest example being Verve that was put on that had one of their products placed on clinical hold for nearly a year, and part of that hold was the agency's ask for germline editing risk.
To de-risk these studies, and to make sure you're on track and have the appropriate testing and strategy we recommend aligning with the agency at the pre-IND stage, and it's best practice to have some supporting pilot data in order to justify your plans for IND enabling studies, and beyond.
Non-Viral Considerations - Drug Candidate Profiling
Elizabeth Figueroa: And then on the non-viral side. So at this stage for LNP formulations that are containing novel excipients, such as a novel ionizable lipid, or a novel targeting ligand, it is recommended to perform excipient only, or vehicle only testing cohorts for your non-GLP pilot tox studies as a way of demonstrating the safety of the novel excipient. This is definitely an avenue or an area where pursuing the avenue of a pre-IND interaction can be helpful, because this is a topic that can be broached with the agency to get feedback on not performing excipient only, or vehicle only cohorts for your GLP tox studies.
And one point about a novel excipient. Novel excipients are excipients that haven't been introduced into a drug product before, or that are being introduced into a drug product by a new route of administration. So even if you're considering utilizing an ionizable lipid from Moderna's LNP formulation for the Covid vaccine, if you're considering an intravenous route of administration that would still constitute a novel excipient.
This is also the point in time where we would encourage developers to begin securing their LNP supply chain, especially identifying secondary suppliers and potentially entering into vendor agreements for safety stock of critical raw materials, such as those that are single sourced, or that have long lead times. I highlight this specifically in the context of LNPs given both historical precedent for shortages of some of these non-commodified raw materials, as well as for the potential of future market entrance that could increase demand on this LNP supply chain. If we see these types of products becoming commercialized and the continued development increasing.
And then, from the perspective of both viral and non-viral delivery, we would also consider whether a companion anti-drug antibody assay is needed for patient eligibility. This is something that you would consider in particular for a drug product profile that's novel, which may be the case for many genome editing drug products.
And on a similar note, the developer should also assess whether an immune response to the delivery vector or the gene editing components is elicited in large scale animal models that are utilized, such as NHPs, and this can be performed with an ELISpot assay.
IND Package Stage
Christina Fuentes: Now the last stage we have defined here really provides the key elements that go into an IND package. And so the stage gates here is driving the IND is driven by the IND enabling studies non-clinically as well as GMP batch production on the CMC side. So for in vivo applications, often we see the batch GMP, or planned clinical batch to be released prior to IND submission, although not always.
Viral Considerations - IND Package
Christina Fuentes: On the viral side, and for the payload itself, very generally speaking, changes in the product into the capsid variant, or the payload from the agency's view, will constitute a new product and risk requiring a new IND enabling study and possibly IND submission depending where you are in development.
And there can be exceptions, such as changes to the viral vector backbone, since it in theory, is not part of the final product. An example case here is, if you need to swap out your amp resistance gene with a kanamycin resistance gene to avoid the use of beta-lactam genes.
There's also some ongoing discussions about the ability to create gene editing platforms to streamline drug development. But this implementation still remains an open question.
Non-Viral Considerations - IND Package
Elizabeth Figueroa: And on a somewhat related note on the non-viral vector side, I want to highlight that for the LNP any significant change that's made to the formulation. This can be tweaks to the lipid molar ratio, for instance, can have profound impacts on the critical quality attributes, including the lipid nanoparticle size, the zeta potential, the encapsulation efficiency, even the biodistribution. So these types of changes can trigger the need to perform additional in vivo nonclinical studies to potentially generate bridging data or perform additional off target analysis.
This is a point where you know, anytime I'm talking about process changes or comparability studies, it's always a good idea to reiterate that the development of a robust analytical testing panel for release and also for characterization, is a critical tool to deploy at these times to be able to establish equivalence between a current and a future state process.
Summary and Dark Horse Support
Elizabeth Figueroa: And so with that, I hope that we were able to ignite plenty of interest and discussion around the continued development of novel genome editing tools and delivery vectors.
Ultimately, I think, the message that I continue to circle back to is the multifaceted selection criteria that goes into the development of a genome editing drug product from the mechanism of editing to the fidelity and efficiency of your gene editing tool to the cargo size, the intended route of administration, your target tissue, if you're performing in vivo administration the clinical indication and the desired safety profile, the list goes on.
We've also highlighted the clinical progress that's been made with ex vivo approaches as well as liver targeted in vivo approaches where continued momentum in the development of extra hepatic in vivo delivery approaches is critical to seeing the full breadth of gene editing tool applications.
And then, most importantly, we highlighted the unique challenges from both a viral and a non-viral perspective for non-clinical and CMC development of genome editing drug products from research and discovery through to IND submission.
And so, with all of these unique challenges in mind, I wanted to end by highlighting just a few areas where Dark Horse consulting has supported clients developing genome editing drug products spanning from strategic product development support, gap analysis, risk assessment to technical support, including non-clinical model selection, selection and off target identification data analysis as well as full spectrum regulatory support.
So with that, I think we have just a few minutes to take a few questions.
Christina Fuentes: Before we get into that, I did want to highlight at the bottom. We wanted to call out special thanks to a few of our colleagues that represent subject matter experts across the regulatory space analytical space as well as project management. So thank you to Kim, Brent and Heather.
Elizabeth Figueroa: Yes, thank you so much.
Q&A Session
Christina Fuentes: Okay. So I see we've had a few questions come up during the discussion. There was one question on our comment on the efficiency of expression in LNPs versus AAV, I think just to elaborate here.
The point being made here depending on the cargo for your LNP system, you may have. Not only do you need to get to the target tissue and the target cell, but you need to be able to transport your editing tool to the nucleus if you are editing the genome. And so in that case, thinking about mRNA specifically, you need, there's some additional hurdles. And so you'll want to design your payload with nuclear localization signals, for example, to ensure that you get your tool to the correct place.
Elizabeth Figueroa: Yeah, that's a great point. Thank you, Christina. And also another thing to consider that makes it more of a challenge on the non viral side than on the viral side, is that a lot of these non viral mediated methods of nuclear entry are dependent upon cell division. So you can have efficiency with some cell types that are dividing more frequently, and therefore the nuclear envelope is opening and allowing entry of your plasmid DNA, or whatever your cargo is, whereas in other contexts it's a lot more challenging and viral vectors definitely are more efficient in this context.
Sorry, Christina, I'm going to let you choose the questions because I can't see them.
Christina Fuentes: Sure there was a question about off target analysis, for in vivo gene editing products, understanding the cell types needed to assess off target analysis. I think that's a great question. I think your model system selection is largely driven by your delivery vector where it's basically, where does your product go?
And then, from your design perspective, do you have tissue specific promoters? Are there particular cell types where you expect the accumulation of your product of which that might be a cell type that you want to assess from a safety perspective, depending on where you expect it to go. Understanding what that profile is.
Elizabeth Figueroa: Yeah. And we would have loved to be able to include a lot of content on off target analytics, because we know it's such an important area in this specific field. But time was a limitation. So maybe in the future avenue, we'll be able to target that a little bit more I see a question about. Could you recommend any CDMOs that have LNP manufacturing knowledge and CQAs that define LNP from a purity profile assessment. I think this is a question that a lot of our clients have come to us for support, and we've absolutely have experience both firsthand and through client engagements and have some CDMOs that we're aware of that are able to perform process development of and scale up of and GMP manufacturing of lipid nanoparticle drug products. So I think that's definitely something that we would be happy to discuss further if you'd like to reach out to us directly.
Christina Fuentes: I see there's also a question about the INTERACT. And you know, do we need to talk about CMC and the INTERACT? I'll say, we've supported quite a few INTERACTs, whether that's authorship of the request or review of these materials.
Largely INTERACTs are driven by CMC or by non-clinical questions I could say, I should say, but inclusion of CMC questions, particularly if you anticipate any particular challenges, certainly can be included at the INTERACT stage, and when we think within CMC, you know, talking about your potency assay strategy, I think that is something that is never too early to talk about to gain alignment on. So we recommend those are considered at that stage.
Elizabeth Figueroa: Yeah, we've definitely seen instances where clients that may have more like a multi-step genome editing mechanism of action, needing to engage with the agency for guidance on, you know, are we needing to demonstrate potency at each stage of this process? Are we able to kind of show development data and then kind of refine the potency matrix into something more streamlined. So this is a hot area, a hot topic of discussion.
I see another question about could you compare the additional CQAs for bioanalysis, if any that are needed for LNP compared to viral vector delivery? And I'm not sure if I'm interpreting bioanalysis directly. I'm interpreting this to be kind of a question in terms of release testing and characterization of your drug product for a lipid nanoparticle versus a viral vector I think the key characteristics of a lipid nanoparticle are physiochemical properties. And so you'll see, kind of a handful off the top of my head. I would say of additional assays that you would leverage in order to characterize the LNP physiochemical properties as well as kind of the non-viral analog to some of your viral vector assays in terms of encapsulation efficiency and release efficiency.
Closing Remarks
Oliver Ball: Well, I think we are at the hour, Elizabeth. So I think I'm going to have to jump in and cut you off. I know there's a lot more questions in there. If you have a specific question that you'd like us to answer, that we didn't get time to address today, then please do reach out to us. We're very happy to talk to you further.
But just to wrap up a reminder that the content for today is available for on demand viewing in the days after the webinar, so you can always watch it again, share it with your colleagues. You can also access all of the previous 9 webinars on demand through the website as well. There's a lot of different topics covered. So have a look there and see if anything there would interest you, too.
Otherwise, look out for the next one in the New Year, and we look forward to seeing you again soon. Thank you everyone for attending today.
Elizabeth Figueroa: Thanks everyone.
Christina Fuentes: Thank you.