What’s been your approach as a CSO to developing a novel approach to CAR T?
Whenever you’re coming up with something truly novel, the natural reaction is to ask, “Is this too complex to work?” My job is to work out how much is too much. We’ve tried the simple ideas – a CAR T against a target – but any solution has to really work on multiple aspects at once. We have to think through, “How do we bring just enough sophistication to this product to solve these problems?” We don’t want a combination of six things; we want to know if it’s possible to do it with two components.
We know that specific viruses can express cytokines, but it’s not enough without a T-cell target. We know CAR T can get into solid tumors, but they need more help to be functional. A cancer-specific virus can seed the ground in the tumor for a CAR T to enter and function effectively. We’re taking two well-characterized modalities and combining them in a fundamentally different way.
How different is this approach from what’s been done before with tumor-specific viruses?
There had been previous work done with tumor-specific viruses to change the immune environment, but they weren’t also incorporating targeting. When you add these agents together, it’s synergistic versus additive. When you put a tumor-specific virus into a tumor, even if it propagates and expresses the target, it stops after a while. So when the CAR Ts come in and recognize the infected cells, they lyse them and release another wave of virus. The two in combination feed off each other and keep going.
What we’ve seen in preclinical models has been a slow burn from the inside out. We just keep marking more tumor cells with the target, which essentially get chewed away and cause another wave of viral infection.
"Whenever you’re coming up with something truly novel, the natural reaction is to ask, “Is this too complex to work?” My job is to work out how much is too much."
How are you engineering to overcome the consistent challenges of CAR-T cell therapy?
The first challenge is the targets themselves. With heme malignancies, you can take a specific target and knock out the whole lineage. There isn’t that ability in solid tumors, and you have to find a specific target that’s present on the tumor and not present on normal tissue. It’s a very small list of such targets. Our approach was to use a tumor-specific virus that has been evolved to replicate only in tumors. We used that to express a target, which we call “Flare.” Flare acts as a synthetic beacon that CAR T cells are already equipped to recognize.
The second challenge is the tumor microenvironment. In solid tumors, it’s designed to exclude T cells, and it’s difficult to solve that challenge from the T-cell side of the equation. So in the virus we use to express the target Flare protein, we also express CXCL9. That draws the T-cells into the tumor. And because the solid tumor microenvironment is so hostile to T-cells, there’s not a supportive cytokine niche there. We create that from the virus, too. In our lead program, we express IL-18 from the virus.
How are you validating the approach?
In our lead program, we’re partnering with a clinically validated BCMA CAR. The Flare in this case is engineered BCMA, and we’ve partnered with BMS to use Abecma. We can take that CAR T, which was designed for myeloma, and use it to make something we can use in all epithelial solid tumors.
We started with known biology so that we can move very fast. By working with a partner to use a validated CAR T, we just had to engineer the virus. BCMA is an excellent target because it is expressed only on myeloma cells and plasma cells, but it’s also a shed target. A lot of BCMA gets shed from the cell surface, so we’ve actually engineered a version of it where we’ve taken out the gamma-secretase cleavage domain so it doesn’t get cleaved, as well as any internal domains which could induce signaling, so that it’s just a tag to allow CAR T can see all epithelial tumors as myeloma.
"A cancer-specific virus can seed the ground in the tumor for a CAR T to enter and function effectively. We’re taking two well-characterized modalities and combining them in a fundamentally different way."
How are you leveraging this virus to overcome the suppressive nature of solid tumors?
Solid tumors start as inflammatory, which T-cells can recognize and attack, before becoming more suppressive over time as myeloid cells, T-regs, etc., are recruited to calm things down and get rid of that inflammation. Viruses are inflammatory agents. When the virus we use first goes in and starts infecting, there is a cascade of inflammatory factors produced. One result is that the myeloid environment, which is normally hard to overcome, gets changed.
We looked at the gene signatures that happen in vivo when we put the virus into mouse models, and there is a change to the inflammatory phenotype. There is less of a cancer-associated fibroblasts signature and more pro-inflammatory macrophages. You see secretion of other factors that draw T cells in. Then the cytokine has to be expressed from the virus, too, so that when the T-cell gets in, it doesn’t undergo activation-induced cell death or get exhausted. Together, these changes create a more permissive environment for CAR T function.
Do you think this would work across multiple types of solid tumors?
We think what we've designed has the potential to work in all of those situations. In our first trial, we will have a spectrum of indications, from non-IO responsive to responsive, so we can see where we are on the spectrum and what level of inhibition we're able to overcome with this therapy.
Different solid tumors have different things that happen in the microenvironment. Some solid tumors are just deserts, and the T cells just don't get pulled in; therefore, if you can pull them in by expressing CXCL9, that might be all you need to do. Prostate cancer would be a good example of that. Some have physical barriers like a rim of fibroblasts around the outside of them, so anything that discourages fibroblast accumulation would be good. And then some of them have negative factors such as TGF-beta being secreted that the T cells have to overcome.
How are you applying your past experience towards your work at Dispatch?
At Amgen and Pfizer, I learned the nuts and bolts of drug discovery. At Allogene, I learned a lot from talking to the regulatory agencies, seeing what the machine of drug discovery looks like, etc. It enabled me to become a little more creative, to take those rules and say, “How can we cut a few steps without multiplying risk? How can we do things in parallel, and balance speed and resources?” And now, at a smaller company, I have a front-row seat for patients. We can see what changes patient to patient, and learn how decisions we make impact patients and motivate the team to keep going.
Also, changing fields is generally a good thing (e.g. from one field of biology to another- I moved from normal Hematology, to malignant Hematology to Oncology). Don’t be afraid to shift from one speciality to another. You have to be more humble and open to learning, but you can also find things that you can bring over from your previous experience that people in your current field might not know.
What advice would you give young people entering life sciences careers?
Pick something that you think is exciting and that motivates you. Pick something that you think is needle-moving and try as hard as you can to do it. It takes long hours and a lot of determination to do anything like this, so you have to love it.
One of the hallmarks of a small company, although it doesn't have to just be small companies, is that everyone has a common vision that they believe they can achieve. If you focus on all the reasons why something won’t work, you’ll never get there. You can spend your career sitting around and saying, “That will never work,” and be right a lot of the time, but what about the things you didn’t give a chance to that could’ve made a difference? Be proactive and bring the energy you need to do something impactful.
