The Biotech Pulse S1E1 - The Manufacturing of Complex Biologics

An edited transcript of the podcast follows below.

Rogier Rooswinkel: Hello, my name is Rogier Rooswinkel, and I am a General Partner at Forbion. My role at Forbion is to source, evaluate and manage investment opportunities with a special emphasis on oncology. Our first podcast is a conversation with a biotech entrepreneur we back, to hear directly from him about challenges he has faced and the smart solutions he has come up with, as he has grown his business. Today, it is a real pleasure for me to introduce our first guest on The Biotech Pulse, Philip Astley-Sparke. Phillip is the CEO of Replimune, a NASDAQ listed Forbion portfolio company working at the cutting edge of cancer immunotherapies. Before starting Replimune, Philip was president of the US at uniQure, as well as CEO of BioVex. Phillip is here with us today to talk about the manufacturing of complex biologics and to shed some light on his approach to manufacturing.

Philip Astley-Sparke: Replimune is a publicly listed company. It is the leader in oncolytic immunotherapy. Oncolytic immunotherapy is the use of viruses that selectively replicate in and kill tumor cells, but do not replicate in healthy tissue. Viruses are injected into tumour deposits, and end up engendering a full systemic immune response, which hopefully also vaccinate the patients against relapse.

Rogier Rooswinkel: 2023 and 2024 promise to be exciting years for Replimune. Let us focus on the topic of manufacturing - the topic of our podcast. All biotechs that are manufacturing complex biologics at some point in their lifetime are faced with the decision to either work with a CDMO or to set up manufacturing capabilities in house or a mixture of both. Obviously, you have gone through this multiple times. So can you tell us about some of the considerations that come into play with making this decision?

Philip Astley-Sparke: There is a central premise that underpins this whole conversation which is that the material you use in your pivotal study, the study upon which you base an approval, and the process used to produce that material, has to all intents and purposes be the same that is used for launch. That is kind of intuitive. You are testing products to show what its efficacy and safety profile is, and that has to be reproducible when the product is prescribed around the world to patients. If you are developing a relatively simple molecule, there may well be contract manufacturers that have the capacity, not only to bring your product through clinical trials, but to actually launch on a global scale. If a company is developing a more complex biologic, you may not have a contract manufacturer that actually can produce at scale to meet worldwide demand. Then you are going to have to look to bring manufacturing in house. At least for the launch you could use the contract manufacturer all the way through the clinical development process. The conundrum is that it takes two years lead time to bring manufacturing in house. So you must have made that decision ahead of knowing whether your drug is going to be successful in a pivotal study. This is the approach of Replimune. We have built out a facility with the use contract manufacturing for all of our clinical studies. What we have done is show comparability from the material that we have used in the pivotal study, to what we plan, to the material being produced in our facility which we would not use a launch. We did not feel that we could rely on contact manufacturers at Replimune to meet worldwide demand. Hence, we have bought manufacturing in house while using contract manufacturers for the development process.

Rogier Rooswinkel: So it is clear that there is for every business its own unique model or hybrid model that you might want to stick to. So say that you have decided to go with with a CDMO. What are some of the criteria that you look at for selecting the right partner?

Philip Astley-Sparke: If you are working with complex biologics, and I have worked with oncolytic viruses and gene therapy, then first of all, you can look at the technical ability of the contract manufacturer. Have they got experience working with viral vectors? Have they got the technical ability to scale up? You may have a process that kind of works for phase one study. In the old days in gene therapy we used to use something called roller bottles, but it is very difficult to use that methodology when you are serving the worldwide market. Do they have the ability to upscale something called hyper stacks larger bio reactors, or even the state of the art bio reactors based on iCELLis? Do they have filling capabilities, and also, over and above, the technical side - capacity? I am not talking about capacity whether they can ever launch a drug. I mean, in terms of your own timelines. Can they produce multiple batches for you with redundancy? What I mean by redundancy is that when you manufacture complex biologics, you have to assume failure. Batches are going to go wrong. If you have employed a contract manufacturer, whereby they have squeezed you in as a third or fourth most important client, and your batch goes down, and so I need to run another one then that is all well and good, but I cannot fit you in till next year, then you have a real problem. Obviously biotech companies are cash burning companies until they become profitable. You cannot just sit on the sidelines for one year. You have to think very carefully about whether the manufacturer is not too busy, and has the time for you, and you can establish a very important relationship. I think in some way your business has to be important to that contract manufacturer in that regard as well.

Rogier Rooswinkel: So in the end it is a puzzle that you need to solve for biotechs. For uniQure and Replimune, it was decided to do some of the manufacturing in house, and you were instrumental in all of these processes. So can you tell us some of the challenges you faced and maybe some of the learnings that you had over that time?

Philip Astley-Sparke: Yes, I think that actually is very useful illustrating the points I have just made. So let us start with BioVex, my first biotech venture, which was also in the oncolytic space. BioVex ended up getting the first oncolytic approved by the FDA. So there we started with contract manufacturers and we had delays. The first thing I did when I moved BioVex across the Atlantic to the United States was to establish in house manufacturing. We actually established in house manufacturing, to produce material even for our pivotal study. So that by the time that we got that result, we were using the same process and the same facilities we were going to use for launch. So roll forward to uniQure, where I was appointed to really move the center of gravity of uniQure from Europe to the United States, and again the first thing I did because of these experiences was establish manufacturing in Lexington, which I think was the largest gene therapy manufacturing facility at the time. This is going back over 10 years ago, and there was limited viral vector manufacturing expertise. So that really had the decision to bring the process in house or have it running pivotal studies again, such that the processes worked out and we had the facility. We were going to have the change facility ahead of launch and look in the rearview mirror. That all worked out very well. The lead haemophilia B product from uniQure is now a marketed drug. At Replimune, things are getting better. By the time we founded Replimue in terms of there were now more contract manufacturers, we were experienced with viral vectors, gene therapy was trying to get into vogue. So here we actually took a more of a middle approach. We have used a contract manufacturer all the way through clinical development through pivotal studies, but we were going to use our own manufacturing facility for the launch. Because we used contract manufacturing for our pivotal study, and we had to show the process is identical in comparability to the material being released from our in house facility. We have now done that. We cleared that with the FDA. Now, as I mentioned in the beginning, we are building inventory. So in each of these examples, we did things slightly differently.

Rogier Rooswinkel: That is very exciting to hear actually Philip, all your stories actually indeed, outlining slightly specific or slightly different ways of getting to the same result. So when we talk about advanced therapies, I often hear you say "the process is the product." Can you elaborate a little bit on what you mean?

Philip Astley-Sparke: Absolutely. So let us use the example of oncolytic immunotherapy which essentially is a virus that selectively replicates and kills tumor cells. So the product is a virus and our virus is the herpes virus. What I mean by the process is the product is how you grow that virus up becomes integral to what the product essentially is. So if you think about the manufacturing process for an oncolytic virus, it has to be grown in a growth medium. Viruses can't just grow by themselves. With something that is essentially a live product, the way you manufacture it becomes integral to the product. Hence any changes in the process are essentially changing, the product coming right back to the central premise from the beginning, which is why it is essential to have for all intents and purposes an identical process you are using in your late stage clinical trials, and you - not the agency - is going to insist on that. Otherwise, it is going to determine that your product has changed, because the process has changed.

Rogier Rooswinkel: That makes a lot of sense. Obviously also that rigidity is what any pharma party, if you want to collaborate or at some point be acquired, will look into very carefully. Maybe switching slightly to the topic of scaling things, because obviously a biotech at its infancy does not know whether its product actually will reach the market, and spending a lot of money immediately on the right process and being ready for the market might turn out to not be needed if you hit a downturn. So how do you balance that capital expenditure at the beginning and also being able to scale such that you can actually handle the marketing? You already gave us the example with with Replimune. How are you doing it now? Maybe you could also see a general premise here?

Philip Astley-Sparke: I can answer that in two ways: one is a technical point and one is a statistical capacity point. In terms of the technical point, regardless of whether or not you are going to have a physical footprint yourself from early on in the development cycle, it is essential as a company, even if you are not going to have a footprint, that you develop very strong process development and analytical development capabilities. Let us break down what I mean by that. Essentially, if you are not starting to bring manufacturing in house, and not using a contract manufacturer, you know you are going to have to scale. You are going to have to work with your contract partner, and you have to get to the right scale at some point in the future. You cannot rely on a contract manufacturer to do that. There may be some clever people there one year, they may have left the second year. It is important you have control over your own destiny so you need to build your own process development department that is working on scalability and process development. Essentially, it is looking at increasing yields. So that when you do get to that final step, that pivotal study, you have got a reproducible process that can be bought through to launch. The same with analytical development. The way you actually prove that your product is that same within a suite of studies, released studies or release assays, that basically show that your product does what it says on the tin, i.e. efficacy and safety. When you run comparability between an earlier stage process and a more scalable process, then essentially those assays have to come up with the same results showing that the process you used early in the development course and the process you use for launch is one and the same thing. Because those assays are all coming up with a consistent set of results. That is also very important. Even if you are not building your own manufacturing footprint from the get go. You have your analytical development capabilities in house working on that, so that you can direct the contract manufacturer, and if a contract manufacturer has turned over or lets you down, you do not have to go back to the drawing board. You have that skill set in house or when you are actually ready to start building your own own physical footprint. And then in terms of scaling, in terms of capacity, and it is always good to have a plan for the future. So at Replimune we have a site which we constructed and we did make sure we had a fallow area. So if we need additional capacity and everything goes incredibly right, we need to basically produce products beyond our lead indications, and if we can fit that without having to scramble around and find a second sight.

Rogier Rooswinkel: Very interesting and maybe slightly switching gears. If you find yourself in such a position and you can you have the option to choose your VC partner, is there anything that comes to mind that you find incredibly important besides overly deep pockets?

Philip Astley-Sparke: So, when it came to selection for Replimune and the seed financing, we looked at Forbion and one other party Amiga who also invested in BioVex. It was the understanding that we had between management and Forbion that this is a potentially long game and success is not linear. You have experience with Forbion in both uniQure and BioVex, in their backing us through the inevitable ups and downs. I had a lot of respect for the Forbion team, and as much as I felt that they have gone into technical areas where others have feared to tread, like oncolytic immunotherapy, like gene therapy before everyone else was on the bandwagon and actually made it work. So I was very comfortable working with Forbion. Again, knowing that they understood some of these more complex issues including manufacturing, and would support management in making the right judgment calls in terms of when to invest in CapEx, and if we did not hit any bumps on the road, would be supportive through them. I think what was also important is brand name and a critical mass of size. You do you need a brand name that is going to attract other investors, you build that syndicate and look credible in an IPO scenario. So I think the relationship side of things between your investors, as you look to grow syndicates with subsequent rounds is also very important. It is remarkable that Forbion has been at the dawn of not just oncolytic immunotherapy, but also the dawn of gene therapy. I have actually often wondered what it is about Forbion that they actually have the mindset to actually go where others fear to tread. So maybe I could throw that question back to you, Rogier.

Rogier Rooswinkel: Thanks Philip, it's a great question, actually. I think it is a couple of things. First of all, we go where the data leads us. So in that sense, if there is strong disease biology, linearity, understanding of modes of actions, we are not afraid to take risks on either the disease side or modality side. I think it has also been paying off for us over the years. I think it is been like you mentioned, we were indeed early believers in oncolytic viruses, which I think are now becoming more and more an established modality. Gene therapy obviously we were both in bluebird and in uniQure. I think the other point is people, and that is on two levels. First of all, I think there are people in house, we have a lot of expertise in certain areas where others might not. I am thinking in this case of Sander van Deventer who founded uniQure, then also come back as the CSO while it was listed. He was one of the first to really pioneer gene therapy in many ways. Secondly, hopefully it is about the people you work with. I think it is a lot of "can people execute in those uncharted territory territories?" I think obviously with you we have a strong link and a good connection, and I think that we trust that you are able to solve some of the puzzles that are in the unknown, mindful that this might take a little bit more time than we anticipated, mindful that there will be unknown obstacles that we need to overcome. But if we believe in management and the data together, I think that is why we are not afraid to go in uncharted territory, like you said,

Philip Astley-Sparke: I think it really shows you the courage of your convictions as well because sitting here today, gene therapy has been established to a degree with oncolytic immunotherapy. But if we go back 15 to 20 years to a time when we were starting to do work together and develop these products, they were wildly unfashionable. So I think the other point is having having the courage your conviction, and it has been an honor to work with a group of like minded individuals.

Rogier Rooswinkel: Thanks very much and the same for us, Philip. It has been a pleasure and I am very glad we're sitting here today. Again, thank you very much for being on the podcast today.

Philip Astley-Sparke: Pleasure to be here and share some insights, particularly in relation to CMC, you know I think in conclusion it is really about controlling your own destiny and you have to look forward. If you are nervous about the technical capabilities or capacity for manufacturing, you probably at that point you do need to take the leap and get control of your own destiny. At a minimum, establish your own quality process development and in house development groups, and moreover, potentially looking at your own physical manufacturing footprint as well.

Rogier Rooswinkel: Very useful. Okay, it's been a pleasure, and for our listeners, I hope you have enjoyed listening to the Biotech Pulse. Please join us for our next episode, which lands in November, where we will be focusing on mergers and acquisitions, and how to set up yourself for success by being able to demonstrate market potential.