Creating a Playbook for Bespoke Gene Therapies

While there is a steady stream of new gene therapies expected to be approved in the next decade, there are hundreds of diseases that could benefit from gene therapies but are not pursued by drug developers because they affect too small a population to be considered commercially viable. In an effort to change the economics of gene therapy for ultra-rare diseases, the Foundation for the National Institutes of Health is establishing the Bespoke Gene Therapy Consortium under its Accelerating Medicines Partnership program. The proposed five-year, $102.5 million program involves the National Institutes of Health’s National Center for Advancing Translational Sciences, the U.S. Food and Drug Administration’s Center for Biologics Evaluation and Research, and a group of commercial gene therapy developers. We spoke to P.J. Brooks, deputy director of the Office for Rare Diseases Research at NCATS and one of the architects of the program, about the need it is trying to address, why it is looking beyond translational science to issues including manufacturing and regulation, and how it hopes to accelerate the development of gene therapies for rare diseases. This episode is part of our ongoing Platforms of Hope series that explores advances in gene therapy and gene editing.

Daniel Levine: PJ. Thanks for joining us.

PJ Brooks: Thanks Danny. Look forward to the conversation.

Daniel Levine: We’re going to talk about the need for platform approaches to gene therapies for rare disease, the bespoke gene therapy consortium, and how this extends beyond the work you’ve been doing through platform vector gene therapy. I’d like to start by asking you what is meant by a platform approach and the case for taking this type of approach with regard to rare diseases.

PJ Brooks: Sure. So I think people use platform approaches in different ways, but here for gene therapy there’s kind of a general approach, a therapeutic platform. And specifically, we’re talking about AAV, or adeno-associated virus, gene therapy, and this is a way to deliver genes into cells. Different types of AAVs go into different cells. And it’s a platform because you can swap in different therapeutic genes fairly easily. I don’t mean to oversimplify it, but it’s a fairly straightforward thing to do to put in different genes using the same sort of AAV platform and thereby create, in principle at least, treatments for different diseases in a fairly straightforward manner. And you’d contrast this with traditional drug development, where you really have to carry out a lot of studies to identify first the biochemical basis of each disease. And then once you’ve identified the biochemical basis, you kind of have to carry out screening studies, lots and lots of screening to find small molecules that will impact those biochemical abnormalities. So you have to do everything from scratch, whereas with a platform like AAV gene therapy for monogenic diseases, the problem is always the same, right? You have two copies of a particular gene, both of which have variants that cause the gene to not work. And so you don’t have that gene function and the solution is always the same. You need to deliver a working copy of that mutated gene to a specific cell type. And the AAV is a platform for doing that.

Daniel Levine: When you start thinking about the extent of the challenge for developing rare disease therapies, why does that kind of an approach make sense?

PJ Brooks: Well, I think, again it goes back to the fact that mostly we’re talking about monogenic diseases and typically these are recessive diseases. So, you have two mutant copies and you need to put a working copy in, and it’s attractive because it’s been quite successful in terms of both preclinical studies and there are now two approved gene therapies using AAV in the United States, a lot of preclinical success stories. So, it’s a platform that’s shown to be effective in multiple diseases. I think the goal of PaVe-GT [Platform Vector Gene Therapy] and also the bespoke gene therapy consortium is to try to build on that success to extend it to as many diseases as possible, particularly diseases that are so small, that they won’t be of commercial interest.

Daniel Levine: Any guesses as to what extent this might speed up the process of delivering a therapeutic to a patient.

PJ Brooks: I wouldn’t want to put a number on it. I do certainly hope though, that it’s an explicit goal of both PaVe-GT and the BGTC [Bespoke Gene Therapy Consortium] to speed up the process and to make as many templates as possible if you will, available to people. So, when a new group is going to come in and try to develop a gene therapy for a new disease, they aren’t starting from scratch. They have a place to start, they have templates available that they could use. They don’t have to guess at the kind of studies that will be needed to get approval to start up their clinical trial, there’ll be a lot more clarity. And we think by doing that, we can certainly speed up the process, but by how much I wouldn’t necessarily want to guess at that.

Daniel Levine: Before we talk about the bespoke gene therapy consortium, I want to touch on PaVe-GT. This is an effort to develop gene therapies for four different conditions at once using the same vector. What’s the ultimate goal of the program?

PJ Brooks: Yeah. The ultimate goal of this one really is to try to streamline the process and see how much, or let’s say another way: every time you develop a single gene therapy, there’s multiple steps you have to go through, you have to do the animal model proof-of-concept. You have to make the vector. Often you have to ask, where does the vector go within the body? What’s the so called the biodistribution. You have to do toxicity testing and various analyses of the vector. And what we hope is that because we’re using the same vector over and over again, the same serotype, in our case AAV9, that it may be possible to sort of streamline if not skip some of those steps. For example, the bio- distribution question, like where does this AAV vector go within the body. That really is determined by what’s on the outside of the vector and should be independent of what’s inside, which is the DNA and coding your therapeutic genes.

If that’s the case, it makes sense that you wouldn’t have to repeat the biodistribution study four times. But we don’t know if that’s true, we have to ask the FDA and they will respond to us and we’ll make that information available. Similarly, some of the toxicology studies that have to be done, perhaps there are those that can be reduced in some ways based on the fact that we’re using the same platform, the same vector that’s made in the same facility in the same reproducible way. And you know, that’s really the goal and also to create these templates for people to use. Because one thing, for example, is we will be applying for INDs. These are investigational new drug applications that have to be granted by the FDA for us to go forward with a trial. And those are quite costly and complicated to prepare. Our idea is that once we’ve submitted one and it’s been approved, we can then make that available to everybody as a template that they can use for their studies. And this would not only benefit the new groups trying to develop these gene therapies, but I think actually it would be appreciated by the FDA. I can’t speak for the FDA, I want to be very clear, but it seems logical to me that it would make their life easier because they would see a standardized type of IND and one that’s basically one that they’ve approved and that would facilitate the process. So, it’s really those kinds of goals.

Daniel Levine: I think it was about a year ago that we spoke on this program about PaVe-GT. What progress have you made?

PJ Brooks: We’ve made a lot of progress in terms of the animal proof-of-concept studies for one of the diseases, propionic acidemia. And we’ve had our first meeting with the FDA. This is an interactive meeting, which used to be called a pre-IND meeting. That’s a good way to think about it where we could ask them some questions about our proposed approach. We prepared this document that we submitted to them. They reviewed it, got back to us, and then we had an hour-long telephone call. We discussed some of these issues. And I think what we’re working on now is how to make that information, the document and those discussions, available to the public. That will be one of our first milestones for PaVe-GT—to follow through on what we said we were going to do, which is making this information public. We’re actually in the process of writing up that material. We didn’t feel like we could just take it and just dump it on the website without any kind of information to give it some context. So we’re working on how much context to provide to make it understandable to people. That’s the kind of progress that’s been made as well as substantial progress in producing the clinical grade vector that we’re going to use for the first trial.

Daniel Levine: Have you learned anything from that work that’s helped to inform the bespoke gene therapy consortium?

PJ Brooks: Well I can say just to backup—I’ve certainly learned a lot from it, which is how difficult and complicated this process is. I’m not a clinician. I’ve never run a clinical trial on the PhD, and I’ve never actually done this before and seeing the amount of work and effort that we’ve already gone into, it’s really kind of striking. I can clearly see the need for streamlining, let’s put it that way. In terms of forming the bespoke gene therapy, I think to some extent, but the bespoke gene therapy effort really came about partly through PaVe-GT, but very much through the ideas and interest of Peter Marks at the FDA Center for Biologics. And it turns out that many of the same ideas were independently arrived at by both of us. In some ways they may seem kind of obvious, and nobody understands the need for streamlining better than our colleagues at the FDA. So, I can certainly see potential opportunities for overlap and cross-fertilization, if you will, between the two programs, I think that would be a wonderful thing if it could happen. And one of them that comes up a lot is what sort of tests do you have to do on a new batch of AAV vector to ensure that it’s safe before you administer it to human beings? It turns out we’ve learned from a meeting we had at NIH several years ago that different companies do different sets of tests on their vectors to demonstrate safety. I might’ve thought that there would be one standard set, but there isn’t, and the number of tests different groups do can vary. An idea that potentially would resonate with both efforts and I think even more broadly is if you could standardize what the minimal set of tests that would be needed for a new vector to pass before it could go into human beings.

In saying that, and when I talk about streamlining making things faster, I do want to emphasize the goal is not to do anything that is going to risk patient safety. The goal is to be sure that we aren’t doing that. Everything that is being done is what’s necessary to be done. The sense is that there may be certain steps in the process of the IND process where we could actually make improvements. We obviously want to do these, make everything faster and more efficient as much as it can, while maintaining the same level of safety that we would expect for a clinical trial that has been granted, and an IND by the FDA. So I do want to make that clear. We don’t want to cut corners that will impact safety. It’s just trying to make things as efficient as possible.

Daniel Levine: The Bespoke Gene Therapy Consortium, which is under the Foundation for the National Institutes of Health Accelerating Medicine Partnership program strikes me as a more comprehensive and ambitious program than PaVe-GT. This was proposed as a five-year $102.5 million effort. Do we know what the funding is yet? What’s the source?

PJ Brooks: I think we’re getting to a pretty good idea of what that funding is, where we’re finalizing it, and hope to announce when we officially launch the program in mid-October. I can’t give you a specific number now because in part, the actual process of raising those funds is done by the Foundation for the NIH, and as someone who works for the NIH, I can’t be out soliciting and raising funds from private entities. So that’s where the Foundation for the NIH comes in because that’s what they do and that’s what they were set up to do. There’s really two parts to the bespoke gene therapy consortium. One part is focused on getting a better understanding of the basic biology of AAV and so that we can improve and increase the efficiency of making AAV vectors in these production facilities and also perhaps improve the clinical efficacy of these drugs when they go into people. That’s what you might think of as a more basic biology component. And then the larger part, which is what I think most people think about of the BGTC is the set of clinical trials that will be done. We anticipate somewhere between three and six clinical trials, different diseases using different AAV serotypes, but all of them being ones that have been used in clinical trials before and have been gotten approved INDs, and to really see how much we can make a standardized regulatory process for doing all three to six of these clinical trials.

Daniel Levine: There seems to be a really good partnership that’s developed between you and Peter Marks here. Both of you were co-proposers of this along with Seng Cheng, the chief scientific officer of rare disease for Pfizer. What, why does having both regulator and industry involved at this critical level mean for accomplishing what you’re trying to do?

PJ Brooks: I think it means it’s hugely important. To be very honest, I don’t know. I don’t think we could have done this without the FDA’s participation, because I’m not sure that industry would have the same interest, as they do with FDA’s active participation. So I think that’s hugely important. And I think also having the participation of some of these companies that make AAV vectors and their willingness to participate is really going to greatly increase the impact of the whole effort. It’s a very exciting and sometimes daunting effort that we’re about to embark upon here, but I’m very excited to get started because I think the potential impact is quite huge. And it’s really a pleasure to work, particularly with Peter. I’ve, I’ve gotten to know Seng pretty well, but Peter and I have really worked very hard on this together, along with our colleagues at the FNIH and to see the level of interest that he has in this, I think is really quite an inspiration to everybody. I think that’s what makes it as potentially impactful as we hope it will be.

Daniel Levine: As you think about the challenges of what you’re trying to do, they extend well beyond translational science and include both manufacturing and regulatory challenges. I wanted to walk through three of the challenges and have you spell them out in a little detail and talk about the consortium and what it’s doing to address this. The first is understanding vector biology. What’s the issue and what are you doing in this area?

PJ Brooks: In terms of the understanding vector biology, there’s really two parts to that: there’s understanding and really optimizing the process of making these vectors for clinical trial use. I mean, people have been making AAVs for a long time, and they’ve come to understand quite a bit about how to do it, but a lot of the processes that are used have been based, to some extent, on trial and error and going with what works. And you can understand that once you’ve got something that works, you don’t really want to mess around with it too much. But I think what we really appreciate is that if you dig down into the mechanistic steps that are necessary from when you put in all the materials to make an AAV vector into human cells and start growing up those cells and getting the vector made, there’s several of these mechanistic steps that could potentially be optimized and even modified by things like regulating genes in the host cell. And if we optimize each one of those steps, it might be that we could increase the efficiency of making vectors in these large production facilities. And if it could change and make that more efficient and safer, perhaps in order of magnitude, that would greatly reduce the cost of making these materials. And so that’s certainly one goal. Another goal is that once AAVs become an AAV gene therapy given to patients, we have a sense of what happens when the AAV gets into the cell and how ultimately the therapeutic gene is produced. And there are specific mechanistic steps that take place. Here, again, I think the idea is that we could optimize and find ways to make those most efficient, and what’s particularly exciting is the idea that we can screen already approved drugs to see if they might impact some of these mechanistic steps with the idea being that you could give a drug, you know, a pill to a patient at the same time, or just before they get the gene therapy and the gene therapy becomes all the more efficient. So those are some of the things we’re trying to get at with the basic biology component of the BGTC.

Daniel Levine: The second concern is manufacturing. You touched on this a little just now, but what are the challenges as they pertain to manufacturing and how will they be addressed through the consortium? I take it having industry here and the regulators are a big plus.

PJ Brooks: Yeah, I think there’s a couple answers to that. One is, as I mentioned, coming up with a standard set of analytics that get applied to all the vectors that are made in the consortium. And that would, I think, be a real, huge benefit because right now you don’t have that. Individual manufacturers analyze their own product, but here we’ll have a chance to put a common set of analytics and all the vectors made by the different entities in the program. And that will provide, I think, very valuable information, particularly to the extent that we have adverse events and things like that, or maybe [we’ll be] in a better position to understand the causes of some of those adverse events, which again, could I think, benefit the whole community. Also when we’ll be working with some of these companies that have made AAV vectors for clinical trials and approved products, we’ll be asking them to make a batch of vectors for one or more of the clinical trials in the BGTC, and really trying to take advantage and utilize as much as possible. This is where the FDA will come in, you know, the work that these companies have done on other products and can that work be leveraged in making a new product. Those are the kinds of things that we really want to find out. And that’s really where the participation of the regulators will be essential because I can’t see how we could do it any other way, honestly.

Daniel Levine: That brings us to the third challenge, the regulatory challenge. What issue do you see that playing in all of this, in terms of making bespoke gene therapies a reality, and how can they address the challenge of developing and producing these?

PJ Brooks: I think it often comes down to the standardization of the analytics you’d use for a new vector. If we can identify through the BGTC a set of say between 10 and 15 analytics that are all applied to every vector and come to some agreement that that set of analytics, if everything passes the quality controls, is then safe enough to go into humans. I think that would be a great accomplishment. I’m not saying we can do it. Obviously, the BGTC has not even officially started yet. So I don’t want to be making claims that we may not be able to attain, but I think that would be one example of something we might attain. Another one might be a standardized process for carrying out toxicology. Typically, this is done in mice and maybe come up with kind of a standard battery of toxicology assays, time courses, and things. So, new investigators won’t have to guess, and they won’t have to propose things that ultimately may or may not all be needed. And I think that benefits the whole field, if we could accomplish that. So those are a couple of examples.

Daniel Levine: What is the status of the consortium today? And what’s the timetable for really launching it?

PJ Brooks: We’ve got at least in principle enough agreements and have raised enough funds. And I say we, I always want to point out this is the Foundation for the NIH, not me and not the FDA, but the FNIH has raised enough funds. And also enough NIH institutes have contributed and pledged to contribute funds that we feel comfortable, that we will be able to launch. We have posted on the FNIH website two notices of intent to issue a funding opportunity. So let me explain what that is. Essentially what we do at NIH, in this case, the FNIH, is when we’re going to put out a funding opportunity, usually once that funding opportunity is put out, there’s a limited time, like 60 days or so, between the time that it’s announced and the time when the applications are due. And sometimes that’s not very much time, but by putting out an announcement of an intention to publish a funding opportunity, we can basically get people thinking about it. They can start preparing, they can start thinking about the materials they need. They can start making collaborations and contacts and setting up, getting together with collaborators so that when the actual funding opportunity does arise, they can look at the specifics and they will be able to hit the ground running, so to speak. And so having done that, putting that out in public, I mean, we obviously would not have done that if we didn’t feel like we’re in a pretty good position to be able to go forward with the program. So that’s basically where it is. And I think the official launch we’re anticipating now is sometime around the middle of October.

Daniel Levine: Ultimately, I know we’re just starting this, but what do you think this will do to change the landscape for gene therapies and for ultra-rare diseases and beyond?

PJ Brooks: I certainly hope that it will make it all more efficient and the net result will be that there will be more clinical trials of gene therapies for diseases of no commercial interest. Then that process of trying to get these clinical trials open, which seems so daunting, and frankly is pretty daunting, will become less daunting. And more than that, I think there may be additional benefits of this. Some of this regulatory streamlining could even impact gene therapy for diseases that are of commercial interest. I mean, I think the company partners would all like that. And I think that’s certainly a possibility. We’ve also now been seeing other efforts starting to begin doing the same kind of things, which I think is wonderful. It would be great if we could get, as those efforts go forward, not just PaVe-GT and the BGTC, but others, if we could get more, you know, data sharing and collaboration, and it’s the best case scenario. I may be a bit naive in that regard, but I think that would be a good outcome.

I also think that some of what will be achieved here, some of the learnings that we get from this effort, it’s not going to be bump free as people say. I mean, I think we’ll learn that sometimes you learn things by learning what didn’t work, but I think some of the learnings about how you can streamline this process may also be applicable to other types of genetic therapies, such as genome editing, for example, which is something that we’re also very interested in, which is not anticipated to be part of the BGTC. This is gene therapy and not gene editing, but some of the same principles of streamlining and a standardized toxicology package, I think could, could be applicable beyond AAV gene therapy, maybe also to lentiviral gene therapy, for example.

Daniel Levine: PJ Brooks, deputy director of the Office of Rare Diseases Research at the National Center for Advancing Translational Sciences at the National Institutes of Health. PJ, thanks as always.

PJ Brooks: Thanks, Danny. Always a pleasure.

This transcript has been edited for clarity and readability.

Thanks to Pfizer, Inc., Bluebird Bio, and Novartis Gene Therapies for their sponsorship of this initiative.