Significant industry milestones have been met this year in the chimeric antigen receptor (CAR)-T space, with Novartis’ and Gilead’s treatments receiving marketing approval in Europe.
Both products received regulatory approval in the US in 2017, as did the first directly administered gene therapy, Luxturna (voretigene neparvovec-rzyl), indicated to treat an inherited form of vision loss.
Developments at bluebird bio are moving equally rapidly, as demonstrated by the firm’s collaboration with Celgene to advance CAR-T therapy, bb2121, through to commercialisation.
In addition, bluebird bio’s gene therapy, LentiGlobin, secured an accelerated assessment from the European Medicines Agency (EMA) and received FDA breakthrough designation for its Lenti-D treatment.
BioPharma-Reporter (BPR) spoke to Derek Adams (DA), chief technology and manufacturing officer at bluebird bio, about how his team is managing the manufacturing of three treatments edging close to the market.
BPR: What is bluebird bio currently working on?
DA: We are on the cusp of commercialising gene therapies for three severe genetic diseases, which are: LentiGlobin for transfusion-dependent β-thalassemia and sickle cell disease, and Lenti-D for cerebral adrenoleukodystrophy (CALD). In addition, we have also partnered with Celgene on the CAR-T therapy we are developing, bb2121, for multiple myeloma that we hope to launch in the US, in Europe and around the world. Of course, these are currently investigational, but we are going strong right now, preparing for multiple launches, so it is an exciting time.
BPR: What are the challenges of trying to work on the manufacturing process for multiple drug candidates?
DA: As chief technology and manufacturing officer, a lot of those challenges fall to my team to try to address issues associated with bringing through an autologous cell therapy. A lot of the challenges posed are due to the personalised nature of the approach – it's the complexity of the one match, one patient approach. This leads to difficulties in achieving economies of scale, particularly for the piece of the manufacturing process that uses the patient's own cells.
We are extending enormous effort trying to do that efficiently and effectively, but also provide the patients with as best of an experience as we can. Clearly, they're being treated for a very serious disease, and their experience is not one we can control completely, but our best efforts are aimed towards ensuring the manufacturing process does not get in the way of their treatment – the aim is for it to actually serve their treatment paradigm as best as we can.
BPR: In what ways can treatment can be an issue for patients?
DA: There are logistical components involved, particularly in the severe genetic disease area, where the patients provide the cells that start the manufacturing process. This means they have to travel to a particular centre to do the apheresis and we have some time limits regarding when we need to start manufacturing. Essentially, we need to line up all the manufacturing schedules to when the patient is available. Trying to line up the manufacturing with the patient's needs, and all other stakeholders involved, is a very complex challenge and is a new model for me, personally.
It means that our manufacturing teams are working very closely with our clinical teams and, ultimately, with our commercial operations to make sure everyone is aligned. Everyone has to be speaking with the same voice so that when patients are wanting information, they can feel confident when they give us their cells.
The backend, which you hear a lot about from the CAR-T area, concerns how fast can we return the cells to the patients and that's the next challenge, in terms of doing all of the release testing and having the treatment ready to send back to an infusion centre, for the patient to receive the transduced cells.
BPR: What kind of timeframes are involved in getting these therapies back to patients?
DA: For CAR-T therapies and, in fact, even in a fairly rapidly progressing disease such as CALD, we want to turn them around as fast as we can, so we're working on a two to four-week timeframe, as early as possible. For β-thalassemia and sickle cell disease, there are existing treatment paradigms that they can be fit in, but we don't have full control over the treatment centres. Turnaround time can then vary, but we can still make it available within the same two to four-week timeframe.
BPR: Is it difficult for the manufacturing process to keep the same pace with the speed of these new developments, such as CAR-T?
DA: It is a very big challenge and it's very early days for these platforms. I say a lot that most of the learning is still ahead of us, and we're still trying to define what the platforms really are. If you want to use some examples with monoclonal antibodies (mAbs) as a platform, we're where mAbs were in the early 1990s – low-yield, very complex processes, with highly manual parts of the processes and not a tremendously wonderful understanding of the biology. That is something that all of us, as an industry, has ahead of us – identifying the important elements and also learning what things we are worried about, right now, we don't need to worry about.
This was one of the interesting things that we learned from the development of mAbs: there were things that you might have worried about early on that, now, nobody worries about – there were other, bigger challenges that we didn't even recognise that were out there. All that learning is ahead of us is and that's exciting to me – it'll be keeping us busy into the future and knowing that we have a lot of new things to discover. This gives me great comfort because, currently, our processes are very complicated. They're expensive and are probably very inefficient, but they work. Knowing that it will get better keeps us going.
BPR: How far are you able to alter the processes while the products move through the clinic?
DA: Improvements you can make to a process, and how big these can be, tend to follow how much you know about the process. This is the interesting 'trap' – these products have shown remarkable efficacy in the clinic, such that they have a rapid development timeline, getting to commercialisation very quickly with processes that have not been worked on during standard, extended clinical development. You can be trapped into early-type manufacturing processes. On top of this, you have the dramatic clinical benefits that you don't want to disrupt: 'if it's not broken, don't fix it'.
However, knowing that you're getting to commercialisation very quickly with processes that are not ideal, as I said, all that learning is ahead of us, the question is how do you capture that learning and apply it? It depends on how well we understand the biology of what is happening. We still have a long way to go to learn all of the pieces involved. There's some great working going on but to translate that to the manufacturing areas, and understanding that a lever that you pull and what that will do to clinical effect or safety, right now we have only relatively crude measures. To your original question, the answer is: not well.
BPR: In regards to bluebird's gene therapies, what barriers are you coming up against in this space?
DA: One of them is the worldwide supply of the vector, used for transducing the cells. The number of products out there that are looking to use viral vectors for treatments is exploding and the capacity to make them has been constrained. We're addressing that by building our own manufacturing facility for vector production. We're also partnered with multiple companies in the US and Europe for the manufacturing of commercial viral vectors. Our facility isn't to replace external partners but simply to augment our supply – spreading the risk away from a fully external model to a mix of internal and external.
BPR: Why is the demand for viral vectors not being currently met?
DA: Well, it's almost inevitable that the capacity will lag behind the need. The field exploded so fast and the clinical development progressed from very early-stage to 'oh my goodness, we need to commercialise, now' so quickly that the capital couldn't be deployed fast enough. The manufacturers that are out there need long lead times to secure the actual customers and can be confident that they can have sustained business rather than throw money at something that might be a flash in the pan.
BPR: What changes do you expect to see going forward across cell and gene therapy manufacture?
DA: The whole industry is going to have to start worrying about the cost production and the cost of goods sold. This is already being talked about: the products are complex and expensive, requiring a lot of different components. This will require those working on gene therapies, including gene-editing companies, to be able to improve the manufacturing processes for the entire supply chain – that’s going to be paramount. I expect you're going to see a lot of partnering with innovative technology companies to help come up with technological solutions for that. You may also see other approaches, such as allogeneic processes, that may bear fruit.
BPR: How close do you feel an allogeneic CAR-T treatment is, in terms of commercial manufacturing feasibility?
DA: 'Off-the-shelf' CAR-T would be a wonderful thing for patients, but the technical challenges are real and there are companies working to address those challenges. It does present a challenge of how much effort do you put on trying to invent that off-the-shelf approach rather than try to improve the autologous approach. You have to weigh up where is the best investment. Do you put your money, on the process that is working right now, or do you try to solve some of those allogeneic processes? We think about this all the time.
Derek Adams joined bluebird bio in March 2017 as chief technology and manufacturing officer. Prior to joining bluebird, Derek was the senior vice president of CMC at Evelo Biosciences. At Evelo, Derek established the initial process development function and supply chain for clinical studies and drove strategy for product development.
