James Heath, PhD., is president and professor at the Institute for Systems Biology in Seattle. He also has the position of Professor of Molecular and Medical Pharmacology at UCLA, and he has directed the NCI-funded NSB Cancer Center since 2005.
The Heath group works on translating P4 medicine from benchtop to bedside, with a focus on oncology. They are particularly interested in developing and applying new molecular, imaging, computational, and microchip-based single-cell techniques for quantitatively interrogating tumor/immune interactions. The team uses single-cell methods, coupled with large scale molecular dynamics calculations to develop a molecular-level understanding of patient responses to engineered T-cell and drug-based cancer immunotherapies, and to engineer novel therapies for “the most challenging patients,” according to Dr. Heath.
GEN recently spoke with him on current issues and future trends regarding CAR T-cell therapeutics.
GEN: How does T-cell function play a role in strategizing the development of CAR-T cell therapies?
Dr. Heath: Basically, you want to optimize the T-cell killing function. Whether CAR-T cells or any other T-cells, you have a host of different functions that the cells can exhibit in the form of secreted proteins. Any given cells can secrete some 100 different functional proteins. Having a handle on all those different functional proteins can provide a pretty powerful feedback on CAR-T cell performance.
GEN: Are there existing challenges with the way CAR-T cell products are being characterized for potency or quality?
Dr. Heath: There are standard potency or functional assays, such as interferon-gamma, that were used in some of the early CAR-T cell work. I thought that they were incomplete. As CAR-T cell therapies have emerged, there has been a much deeper understanding of exactly what is needed and what you do not want to do in terms of highlighting T-cell functions. A lot of those advances in understanding are reflected in new generations of engineered CAR-T cells. In fact, some of these new generations now represent the pinnacle of cellular engineering.
GEN: Are all the CAR T-cells in the cell therapy product created equally? What methods are currently being used to determine the editing of these cells so that they lead to the expected outcome?
Dr. Heath: Well, this issue of creating CAR-T cells equally is something that has been refined only in the last year or two. Very precisely engineered CAR-T cells are just now entering the clinic.
Previously, maybe one used different retroviral vectors and things like this to insert the CAR-T cell receptor, and it led to a pretty high variability in CAR-T cells. One did not always have great control where that receptor was inserted, and so you might have ended up with a given CAR-T cell batch that actually contained a host of different products from the genetic point of view.
But now, more precise genome editing methods, such as CRISPR, and the ability to insert genes as well as to remove them, are narrowing that spectrum. Although they are not in the clinic yet there are also a lot of efforts to try to develop off-the-shelf CAR-T cells, in which one designs a CAR-T cell product that can be used for almost any patient. These kinds of things have evolved over the past few years so that the unintended diversity in CAR-T cell manufacture is gradually disappearing. Intended diversity, through engineering, will continue to expand as scientists continue to invent.
GEN: At the other end of the spectrum, are there ways to predict adverse responses to the CAR-T therapies?
Dr. Heath: There are some, but this is an area that is still evolving. A key challenge is that there are many different types of CAR-T cells, and for any given type, we do not have enough patients in clinical trials who exhibit specific adverse responses to allow one to resolve the causes of those responses. The diversity of CAR-T cell products that are being introduced makes it challenging to do an apples-to-apples comparison.
Some of the early, serious neurological events may have been attributable to the variable properties of the CAR-T cells, coupled with an incomplete knowledge of how diverse patient populations might respond. But even then, the number of those events was pretty low in frequency. For example, when we looked at CAR-T cell products from Kite Pharma a while back we saw that there were certain cytokines that tended to precipitate neurological adverse events. Not serious ones, but nevertheless they were real. We were able to identify certain signatures in some of the CAR-T cell products that seemed to correlate with those events, and we saw those signatures before the CAR-T cells were given to the patient.
What you are trying to do is engineer a patient’s immune system to attack their cancer, and there are inevitably going to be consequences because of that. A tremendous amount of research is focused on trying to understand how to avoid autoimmune or neurological adverse events. Those types of results are just now being physically resolved or will be over the next few years.
GEN: How might single-cell functional analyses accelerate development of CAR-T therapies?
Dr. Heath: These single-cell functional analyses have been pretty good across the board at differentiating responders from non-responders, both during the course of therapy and, in some published cases, in terms of analyzing CAR-T-cell products before they are even infused into the patient. There are not many other metrics that have yet emerged that give you that type of feedback.
The analysis of CAR-T cells prior to infusion can guide and help you anticipate patient response. This can provide feedback on manufacturing because you want to control the manufacturing process to optimize the tumor killing properties of the T-cells, while also minimizing factors such as T cell exhaustion. Those are two separate biological processes and so separately optimizing them for therapeutic performance is a reasonable near-term goal. Poly-functional single-cell analyses can provide powerful tools for such optimizations.
One thing that is a bit of a challenge is that as CAR-T cells have become more and more effective you still will have adverse events. You have patients who do not respond well. But you do not always see the sort of detailed molecular workup of the patients that you would need to understand if there is something about a patient that is limiting the positive functional characteristics of the T-cells. To take a related example, almost no patients with melanoma tumors get biopsies anymore, because of the success of checkpoint inhibitors. There are still patients who don’t respond, or who might respond badly to such immunotherapies, but there really aren’t opportunities to study such patients.
It is my hope that as we continue to push new CAR-T cells forward, these participating patients will be deeply studied over the treatment course, and even beyond. That type of patient participation is key to helping us aggressively advance the science in a way that allows it to benefit