Wayne A. Marasco, MD, PhD, discusses the intricacy of engineering CAR T cells and the early data he has observed with CAR T-cell therapy in renal cell carcinoma.
Wayne A. Marasco, MD, PhD, professor of medicine at Harvard Medical School, and principal investigator of cancer immunology and virology at Dana-Farber Cancer Institute
Wayne A. Marasco, MD, PhD
The explosion of CAR T-cell therapy in hematologic malignancies has revolutionized treatment for select patients, but their application in solid tumors, particularly in renal cell carcinoma (RCC), still poses significant challenges, said Wayne A. Marasco, MD, PhD.
However, as developments continue to advance, Marasco is confident that CAR T-cell therapy is here to stay.
"My goal is to go for a cure," said Marasco, who has been involved in preclinical research with CAR T-cell therapy in this space. "I wouldn't put my time into this if I didn't think we could do it. To say that you can cure a cancer is a bold statement that I don't take lightly, but [we can] with the tools we have developed and the knowledge that we have gained across the field."
In an interview with OncLive during the 2019 Kidney Cancer Research Summit, Marasco, professor of medicine at Harvard Medical School, and principal investigator of cancer immunology and virology at Dana-Farber Cancer Institute, discussed the intricacy of engineering CAR T cells and the early data he has observed with the approach in RCC.
OncLive: Could you share some background on how you became involved with this research on CAR T cells?
Marasco: In 2000, I was diagnosed with RCC. While I was in the hospital, the father of five young daughters at that time, I decided I had to do something about it. I work in the field of therapeutic-antibody engineering. I wrote my first CAR T-cell grant 18 years ago, before it was fashionable to do this sort of work. Since that time, I've been working on this to be able to develop a cure that will [help] people.
What application does CAR T-cell therapy have in RCC?
CAR T cells are a way of engineering your T cells to target a particular cancer. In the case of kidney cancer, there are certain proteins on the surface of those cancer cells. We put antibodies on the surface of T cells to get the T cells to be activated in their natural way, but with an artificial moiety to direct them directly to where you want them to go. When you do that, you can get pretty dramatic responses.
The engineering is relatively straight forward. We take white blood cells out of a patient, generally by leukapheresis, and we put them in tissue culture. Then, we treat them with viruses that contain the encoded moiety. Ultimately, the end result of that is a population of CAR T cells that could be expanded and be given back to the patient.
To test the quality of these, we do a number of different characterizations of them. Impressively, if you take these now-patient cells that have become CAR T cells and you treat the tumors with them, they are markedly enhanced in their ability to kill tumors, as well as their ability to secrete pro-inflammatory cytokines, such as interleukin-2 and interferon-gamma. They have all of the hallmarks of a potent killer cell.
Why has there been more hesitation in using CAR T-cell therapy in solid tumors until recently?
There are a number of reasons for it. Success with CAR T-cell therapy has almost exclusively been in hematologic malignancies that have a unique protein on their surface called CD19. That is what makes it somewhat unique, and it is a liquid tumor. That molecule, CD19, is not expressed on normal tissue.
One main problem with solid tumors is on-target, off-tumor adverse events. Solid tumors overexpress proteins that are also expressed in healthy tissue. Because of that, and the fact that CAR T cells are so potent, CAR T cells can kill healthy cells.
Now, the field has matured, and we have been trying to figure out how to get past that major barrier.
How do you combat that barrier?
We've had to learn to almost reverse engineer the way we normally did monoclonal antibody therapies. I've spent my career making human monoclonal antibodies as potent as they can get—with high affinity and high specificity—and as it turns out, that is generally not what you want to do on a CAR T cell.
You have to down-regulate the affinity; you have to reverse engineer them so that they are in this "sweet spot." If you get it just right, the affinity of those CAR T cells will recognize a highly expressed, overexpressed protein on a tumor cell, but it will not recognize that same, normally expressed protein on a healthy cell because it is expressed at lower levels.
Another way is to actually target two different antigens. A tumor cell may overexpress one antigen that is expressed on a healthy cell, but the likelihood of a healthy cell expressing two antigens that are overexpressed in a tumor cell is significantly less likely. Therefore, we try to find two proteins that are overexpressed on tumor cells, but are not expressed in any healthy cell. Then, the same engineering has to take place to find the sweet spot.
Finally, we have to engineer them so that they are both expressed in that CAR T cell. This involves getting everything right: the topography, the angle of approach, and how far the target is from the membrane.
What other obstacles exist with CAR T-cell therapy in solid tumors?
The other problem is the tumor microenvironment. Basically, the tumors have commandeered the immune system. When you look at a tumor bed, they are often surrounded by white blood cells. They know there is a tumor there, but they cannot do anything about it; they are exhausted and are nonfunctional. It is the ability of tumor cells to either express proteins on their surface or secrete molecules that commandeer the immune system and suppress it.
The key is changing the tumor microenvironment. It is not enough to get the CAR T cells to the tumor, even if you could do it safely by dual targeting. You have to change the tumor microenvironment. If you don't, those CAR T cells will get exhausted like every other T cell.
Our approach to do this is to create CAR T-cell “factories.” These secrete monoclonal antibodies at the tumor site directed to these molecules, which we know are involved in shutting down the immune system.
In the first published work we have, we did this with CAR T-cell secreting with PD-L1 inhibitors. We all known that checkpoint blockade is central to immunotherapy.
Of the data we have thus far, what seems promising?
It takes a lot of data to prove your point that this is a viable therapy. In animal models, we have shown that the generation of CAR T-cell factories has a profound effect on white blood cells reverse exhaustion markers like PD-1, TIM3, and LAG3.
We have been able to show that these antibodies can localize and stay local at the tumor site. You don't get waning doses like you do with a monoclonal antibody injection or a drug injection; CAR T cells secrete 24/7.
Lastly, in the animal models, we showed that tumors treated with CAR T-cell factories were smaller or nonexistent [after treatment].
You covered a lot of ground, is there anything else that you would like to discuss?
Some of the technologies we have available now are able to dissect the immune system at the tumor site. Remember, you didn't make it to adulthood without an intact immune system, so you are not going to cure a tumor by just treating it with targeted drugs. You need the immune system involved.
As a cellular approach, if you use CAR T cells right, these cells are going to stay around for life. It is not a “hit and run,” and kill the tumor and disappear. These cells have engineered long-term durability in patients, so we are creating an artificial but evaluable immuno-surveillance system.