CAR-T: What’s next? Part 2
In my previous blog post, I suggested that the market size for gene-modified T-cell therapies would be roughly $9 Bn in 10 years, with a range of around $5 - $20 Bn. In this post I will provide some structure for this wide range by breaking out five key factors that will help guide you in interpreting what happens over the next few years. Three of these factors are clinical and scientific: the products and their indications (this one is the most obvious); performance against solid tumors and a broader set of hematological malignancies (i.e., will they work beyond the B-cell malignancies where they’ve performed best so far); and the success of allogeneic ‘off-the-shelf’ treatments (versus the current autologous, one patient one drug approach). The other two factors are business-related: manufacturing/supply-chain, and pricing/payment models.
Looking at the products in the pipeline, and their associated indications being tested in clinical trials, provides the first inklings of how the market will evolve. Making a base forecast involves adding up every product forecast for each indication and weighting each one by an estimate of the probability of success. Tedious, but a necessary first step. But some events – candidate successes and failures in clinical trials – have more impact than others depending on how they affect the other factors we need to consider. Which brings us to the other four factors.
The second area to watch is how well gene-modified cell therapies work for other hematological malignancies and – the really big prize – for solid tumors. Although clinical evidence best supports efficacy against certain B-cell hematological conditions using CD-19 targeted CAR-Ts, we don’t know how broad this will extend. So far, the signs are good that B-cell malignancies like DLBCL and other non-Hodgkin’s lymphomas will be addressable, both of which have larger addressable populations than ALL. But for others such as CLL the benefit/risk ratio is less clear. There are many developments and variations in the CAR-Ts currently being tested, both in their design for safety as well as in the selection of targets, so it will be at least a few more years, even with the rapid pace of development in the field, before we know how broadly CAR-Ts will be effective. The next 2 years or so will give us the first hints, as new products and indication expansions come up for approval.
What about solid tumors? Products targeting these indications won’t be ready for approval for another 3-4 years, but so far, we can make two general inferences: first, CAR-Ts in solid tumors will only be applicable in special situations, such as glioblastoma; second, if solid tumors take off as a market, the products will be mainly TCRs, not CAR-Ts. There are two fundamental reasons CAR-Ts have been so well suited for hematological malignancies: (1) they target cell-surface receptors, so can access all the cells in a ‘liquid’ tumor (i.e., blood), but can’t get at cells inside a solid tumor; and (2) CD-19, the cell-surface receptor that has worked so spectacularly well, is specifically expressed on B-cells, avoiding potentially toxic, non-specific killing of other cells. Glioblastoma is a special case because treatment can be delivered locally to the brain via intra-cerebral injection, thus avoiding non-specific interactions with cells elsewhere in the body. Mustang Bio’s MB-101, a CAR-T targeted against IL13Rα2 in Glioblastoma Multiforme, is a promising candidate. There may be other such special cases as well as exceptions for CAR-Ts in solid tumors, but it’s too early to know.
Currently the best bet for the more general case targeting solid tumors lies with TCRs. These are genetically engineered T-cell Receptors designed to target intracellular antigens bound to the Major Histocompatibility Complex (MHC) on the cell's surface. MHC proteins present intracellular antigens on the cell’s surface as a way of identifying the cells as ‘self’ rather than foreign. Cancer cells tend to express different and often novel intracellular proteins (neo-antigens) and so are often identified as foreign and killed by T-cells. ‘Tumor-infiltrating Lymphocytes’, or TIL, are a manifestation of this process, which the engineered TCR-expressing T-cells exploit. Although theoretical and pre-clinical data support the TCR approach, estimated product approvals are not expected before 2020. Near-term clinical trial findings, such as GSK and Adaptimmune’s phase 2/3 trial of NY-ESO-1 in synovial sarcoma, should provide some early inklings to how well this approach works.
The third area to look at is whether autologous therapies will give way to allogeneic off-the-shelf treatments. Autologous therapy – using a patient’s own cells to create a personalized therapy just for them – is the ultimate personalized medicine. The approach is not new – autologous hematopoietic stem cell transplants have been around for some time now – but the addition of a manufacturing process to genetically manipulate the cells and expand them as well as supply chain logistics to ship cells back and forth while maintaining quality and integrity, greatly increases complexity, risk, and cost. And it is not something most pharmaceutical companies are used to managing. As an individualized therapy, reducing costs by increasing scale doesn’t work, at least not nearly as well as for a batch process.
The holy grail in cell therapy is to create one large batch of cells than can be used to treat everyone – i.e. an off-the-shelf therapy. Cellectis is leading the challenge of developing a donor-derived (allogeneic) off-the-shelf CAR-T therapy, which they call Universal CAR-T, or UCART. Although Cellectis has announced advances in their approach, analysts don’t forecast any UCART products to come out before 2021-2022, and experts we spoke to think that’s optimistic. Cellectis’ recent announcement of a clinical hold on UCART123 is a reminder of the risks in any clinical development program.
The two other key areas that will play out over the next few years will affect the entire field: Manufacturing & supply-chain costs; and pricing and payment models. As mentioned above, these therapies are hugely expensive to manufacture commercially, so reducing COGS related to manufacturing as well as supply chain logistics will be critical. Industry experts believe this is possible and will happen, but it will take time and require innovations in process engineering and in tools[1].
Workable pricing and payment models will also be critical to the industry’s success. Kymriah was launched with a list price of $475,000 and will be charged only if the treatment works after one month (ALL is aggressive: patients recover or die quickly). Novartis argues that this is a fair price for the value delivered, while some argue that these therapies are simply too expensive and another case of pharmaceutical companies charging what they can. Both statements are true, but Kymriah’s pricing and payment model is only the first stab at the issue. Only when several treatments are on the market and contracts with multiple payers negotiated, will we have an idea of where this market is headed with respect to both pricing and payment. Differential pricing, competition, and manufacturing cost reductions will certainly play their parts.
We look forward to watching – and commenting on – events in this field as they unfold. Please subscribe to our blog if you would like to be notified of future blog posts from us.
[1] Tools include equipment, consumables/disposables, media & reagents, and vectors
