Living Drugs and Rewriting Immunity: A Conversation with CAR T-Cell Pioneer Michel Sadelain

Michel Sadelain, MD, PhD, is widely regarded as one of the founding architects of CAR (chimeric antigen receptor) T-cell immunotherapy—an approach that reprograms a patient’s own immune cells to hunt down disease. Now at the helm of the Columbia Initiative in Cell Engineering and Therapy (CICET), Dr. Sadelain is building a future in which genetically engineered cells could help treat not only cancer, but also autoimmune diseases, transplant rejection, and beyond.

In this wide-ranging conversation, he breaks down the fundamentals of cell therapy, reflects on the dramatic shift in public perception around living drugs, and explains why Columbia’s interdisciplinary strength makes it uniquely positioned to lead this next era.

Your work in cell therapy, can you explain what that is?

We take human cells and then program those cells to acquire new properties that are increasingly unnatural.

In classic gene therapy, the intent is to provide a gene that's missing in someone's genome or that's mutated and inactive, or in some instances, to repair a defective gene. However, in the field I'm currently working in, which is often referred to as cell therapy or CAR T-cell therapy, the genetic modification typically does not aim to repair an inherited defect, but rather to reprogram the cells to perform new tasks.

Wow. What are these new tasks, and what types of cells are being reprogrammed?

The most advanced aspect of this field is based on the genetic modification of immune cells, particularly T cells. T-cells are part of what we call the adaptive immune response, and they're very important for fighting infections. And we also know that they contribute to fighting cancer, and sometimes they prevail, although most of the time they lose the battle, which is why so many people ultimately die from their cancer despite attempts made by their T-cells and their immune system.

Therefore, the notion that maybe we should confer new properties upon those T-cells to better equip them to fight cancer was born. And the idea is straightforward, but it took the last 30 years to reach a point of working in some cancers. The first two of these genetically modified cells were approved in 2017, and they are harvested from a patient brought into a laboratory, genetically modified, and reinfused into the same patient. So, they're autologous.

Will you explain the purpose of the genetic modification itself? How does that work?

The purpose is to target those T-cells so that they will recognize an antigen of our choosing. And when those genetically modified T-cells see that antigen, they're supposed to go and kill those cells. The first target approved by the FDA is a molecule called CD19, which is present in many lymphomas and leukemias. That works rather well in patients with relapsed, refractory disease. These are patients who had exhausted the standard of care, various chemotherapies, and often a transplant.

There were remarkable, complete responses in a majority of patients with these lymphomas. And this led to the rapid approval of these drugs. In fact, when describing how these receptors are designed, how the genes are introduced into these T cells, I refer to them as living drugs.

So, it's an immune effector cell, a T-cell; now, sometimes other cells are repurposed and targeted to perform a specific task. It turns out that these T cells, which target CD19, are not only useful in B cell malignancies (i.e., mostly lymphomas and leukemias) but also in autoimmune disorders. Today, there is a mad rush to run clinical trials in autoimmune disorders, starting with systemic lupus. There's a group in Germany that, on a compassionate basis, produces these T-cells, which you can do now in hospitals.

Does the cell repurposing process work similarly in autoimmune disorders?

Select hospitals produce them in the exact same way they produce for cancer patients, and they want complete remission. It's still a limited number of patients, but it has triggered a tsunami in the rheumatology and autoimmunity fields.

On a bigger scale, it's not just for cancer. We still need long-term studies, and although they're not perfectly safe by any means, they're certainly acceptable and accepted as safe. And this really unleashes the imagination of people, including scientists and physician-scientists, to take advantage of this approach and remodel cells to perform different tasks.

In the world of immunotherapies, the cells that are modified are referred to as immune effector cells, and there are also what are called natural killer cells. Additionally, there are other subsets of immune cells that may be useful in treating various diseases.

There are other cells that may also be amenable to genetic modification and could be of use in a number of fields, including the field of transplantation.

Let’s talk about transplantation. How do you see the potential applications of these cell therapies evolving in this field?

I'm going to learn more about transplant from my colleagues here at Columbia, but to limit rejection of a transplant, you have to establish immunological tolerance. And transplanters have been perfecting that for many years, especially when it comes to xenotransplants, which are quite different from the recipients.

What I think the world of cell therapy and genetic engineering brings to this field are additional modalities to tackle this overarching challenge. There are immune cells that are immunosuppressive and protect our normal tissues. They're naturally occurring cells. We all have them. Sometimes these cells protect the tumor, which is what we don't want. Cancers astutely recruit these cells to prevent being rejected by the patient's own immune system. So, they play this protective role, and of course, they can play this protective role in a desirable way now.

I know that Megan Sykes and some of her colleagues are interested in mastering this technology to control these cells to protect organ grafts. That is obviously one area where I would be more than delighted to actually work with these teams and bring our knowledge of T-cell engineering. However, this is another type of cell, known as the immunosuppressive cell, specifically the regulatory T-cell.

How close do you think we are to engineering immune cells that can selectively suppress transplant rejection, while maintaining broader immune function?

There is progress in this area, but it has not been a slam dunk. Let's just say that we are at the beginning of some experimental work in clinical trials. The field is very advanced in terms of genetically modifying cells.

Does that mean these clinical trials are designing CAR-T cell therapies to target all different kinds of cells across the board?

Yes, they are designed for an increasing number of targets, not just the CD19 that I mentioned. There are over 100 targets in the clinic today—more than 1,400 clinical trials of CAR T-cells listed on clinicaltrials.gov. Clearly, the basics for the methodology are there. And that's, of course, one of the first things I want to do here at Columbia, make that possible on this campus.

That's particularly relevant for immune effector cells in autoimmunity and cancer. That's not for these regulatory T cells, right? With regulatory T cells, I think some of the key challenges yet to be overcome are fairly well known. It's not so easy to expand these cells. You have to genetically modify them with receptors that may not be the exact same ones used in cancer because you have to respect or preserve those immunosuppressive properties. They have to remain to keep that function of protecting the graft for a long time. So, the stability of those properties still needs further investigation. The cells have to persist for a long time. We don't know how long—is it for the lifetime of the graft recipient? Perhaps not that long, but I don't think that answer is known.

What are some of the other challenges you face in designing these T-cells?

I mentioned producing these cells in large amounts while maintaining their properties over time. They can't flip into some other set of properties. They have to remain immunosuppressive. Their job is to protect the graft. Also, when you target these cells, you send them somewhere, you tell them where to go. There's a molecular target that you identify, and you tell your T-cell to find that molecule on those cells and do your job there. Well, what should those targets be in some of these different settings? There are many important questions, and I think there's every reason to believe that these challenges can be met.

In talking to Dr. Sykes, it sounds like there’s another interesting intersection between cell therapy and transplantation—stem cells. Can you tell me more about that?

That’s another setting where cell therapy intersects with the world of transplantation. You produce cells from induced pluripotent stem cells. Beta cells, islet cells not harvested from human donors, are instead produced in a bioreactor through the directed differentiation of pluripotent stem cells. Nobody at this point knows how to produce a kidney, for example, but people are coming up with the production of beta cells for diabetes or cardiomyocytes or retinal pigment epithelium for different forms of blindness, chondrocytes, cartilage, et cetera.

And they come in the form of cells because they're differentiated in a bioreactor. They're not an organ, but they can be used to replace or consolidate a failing organ. And so, because they are produced as cells, and assuming that this will not be autologous from the [transplant] patient themselves, we will run into the same issues of rejection by the host. Now, a key difference in this case is that because they're produced in culture and exist as single cells, they may be highly amenable to genetic engineering. It’s more about engineering the cell to protect it so that it doesn't get recognized by the host immune system.

Fascinating. How do you think academic medical centers and other institutions can best work together to advance these therapies?

Medical centers are ideally positioned to perform this work, which is not to say that there isn't a major role for the pharmaceutical industry, but the expertise in transplant models and genetic modifications, these cell culture systems, the frequent use of patient derived cells need to proceed very carefully in well-defined subsets of patients. Then, perform a very deep analysis on a limited number of patients who undergo these trials. All of that together points to embedding this in a large, comprehensive medical center.

That’s why we want to establish cell therapy and cell engineering on campus—close collaboration with both those who conduct research and, ultimately, the clinicians involved in the clinical trials. This nexus is critical.

How wide does that collaboration extend across academic disciplines?

It’s a very large group of investigators who can contribute to advancing these new medicines—deep expertise in transplant biology, transplant immunology, genetic engineering, systems biology, bioengineering. The proximity to bioengineering and a whole medical school is an enormous asset because there are a number of investigators developing chips, devices, matrices, better tools to direct the differentiation of cells in vitro to obtain the precise cell type that you want to give to your patient.

This creates an ecosystem that few institutions would have, and now we bring the possibility of actually implementing these ideas in the clinic. I hope that will serve as a clincher or a trigger or both for bringing together people who have an interest in taking advantage of cell engineering to develop new medicines.

I’m curious about the ethical considerations as these types of cell engineering advance. What should we be mindful of?

That's a great question. This actually brings us back to oncology, as the introduction of any new drug or technology is a very thoughtful process that involves scientific and ethical considerations. The science must be sound; safety measures should have been worked out before taking the risk of conducting a trial. Everything we do is FDA-reviewed in the field of cell therapy, and it involves a risk-benefit analysis. If the risk is greater, you are more willing to tolerate higher risks. And once you have de-risked some of these technologies, they can now be imported to other fields of medicine. The perfect textbook example is what just happened with these CAR T-cells targeting CD19, which started in cancer. Once people know more about their safety profile, it is now used in patients with autoimmune disorders.

We opened the first trial targeting CD19 in the US in 2007, but were unable to find patients. We had to produce a little video to explain because people were afraid of it. They thought it was just too weird to be engineering recombinant viruses, harboring a synthetic gene that encoded for a type of receptor never used in humans before. And it's an irony that today there are waiting lists for these therapies.

It must be thrilling to be part of such a paradigm shift.

We were fortunate to see that in our lifetime. It doesn't always happen. It's only every so often that there is what you could call a “paradigm shift” in science. You could just say that a new modality or tool is coming to the world of medicine, and now we have genetically modified immune effectors. And that's what I hope engineered graphs will do in the next 10 years as well.

Is there something you’re most excited about as you embark on this next phase?

Well, yes, that’s an easy one. It's to collaborate with my new colleagues. I want to learn more about what they are interested in, what they see as their next challenges, and barriers. And I want to know if what we have learned can be of any use to them. It’s very exciting.