We had an exciting session today talking really about the boundaries of immunotherapy now in cancer and its very rapid pace of development. My specific topic was on CAR T cells that have just recently become FDA approved and their uses in leukaemia and lymphoma. Then I spent some time discussing the use in experimental settings such as solid tumours like prostate cancer.
What did you mean by, “it’s a process, not a drug?”
This is a concept that drugs are purchased and are given one time and then they metabolise and leave really no trace in the patient. CAR T cells, what I mean by process is similar to a bone marrow transplant. In a bone marrow transplant one finds a donor and then basically exchanges the immune system and that requires a whole team of physicians and a specialised treatment. So CAR T cells, the patient themselves is a donor but it requires manufacturing and then when the cells are given they stay there, we now know, for up to seven years in the patient. So it’s a living drug unlike the standard drugs that are metabolised.
Can you give some patient examples?
The first patient we treated was in 2012 and that was Emily Whitehead and she was reported in the scientific literature a year later. She’s now more than five years out without any further therapy so she was a striking result. When that happened, though, she almost died of cytokine release syndrome. In the publication in The New England Journal of Medicine we showed that her fever was 106o for three days so she had a very severe reaction but that correlated with the elimination of her tumour. So we didn’t know when we first treated her if that was a fluke and she was going to be a rarity or not. It turned out our first twelve patients we had a 92% complete remission rate so it was striking how well it worked. That became FDA approved for people with acute leukaemia ages 3-25 in the United States last summer.
What are the chances of getting Cytokine Release Syndrome?
Cytokine release syndrome is the fever that occurs when the tumour is being attacked by CAR T cells. We do know that the chance of getting cytokine release syndrome depends on how much tumour you have. So if you have a packed bone marrow with lots of tumour then about 50% of the patients get cytokine and end up in the intensive care unit. But it also correlates with efficacy so that if the patient doesn’t have cytokine release syndrome they’re usually the ones that don’t respond. The primary cytokine that causes the fever and the cytokine release syndrome which can lead to other organ system failure if not treated is called interleukin-6, a cytokine that now can be blocked by a drug called tocilizumab which is an IL-6 receptor antagonist. That was co-labelled with CAR T cells that are now marketed both by Kite and Novartis.
Is there potential for use in combination?
Yes. CAR T cells appear to combine well with many other modalities such as checkpoint therapies, oncolytic viral therapies and some targeted therapeutics such as ibrutinib. So we think there will be many combinations. One limitation right now is that the preclinical models are not very good at telling us what is going to be the best combination or whether sometimes there might be antagonistic effects.
What are some of the challenges involved when developing CAR-T cells for solid cancers?
At this point the big scientific challenge to the field is developing CAR cells for solid cancer. So far the results have been generally disappointing and we now know why. One issue is that the CAR T cells targeting solid cancers have had a paucity, really, of good targets where we could not have lethal on-target effects but off the tumour. So, for instance, causing encephalitis at the same time the tumour goes away, or lung reactions at the same time as the tumour if there’s shared antigens. So the optimistic aspect, though, is that if we take tumour biopsies from a number of adenocarcinomas, and in our case we’ve done studies in ovarian cancer, pancreatic cancer and lung cancer, we find that in a petri dish the CAR T cells do kill the cancer cells and in a very similar way that they do in leukaemia. So the optimistic aspect is the CAR T cells can kill and it’s going to be an issue now of solving the targeting issue which can be done with CARs that have multiple targeting warheads, we call them, as well as being able to shut them on or off. All those things are in the laboratory at this point.
Is there a potential for the patient to become a donor of CAR-T cells?
When we infuse CARs they actually expand, in some patients more than 100,000-fold from what we infuse. So there’s the manufacturing we do in the laboratory and then that in the patient. It will obviously become cheaper if we can make most of the manufacturing, meaning growth of the CAR T cells, happen in the patient rather than the laboratory. It’s very different than other drugs where there’s a fixed dose and then they’re metabolised; in this case we actually need less CAR T cells with the patients who have the higher tumour burdens. It’s very counterintuitive but it turns out the higher the tumour burden then that drives more proliferation of the CAR T cells so we need less. So it’s very different than standard pharmacology when you have a living drug that reproduces in the patient.
That, in fact, is quite likely, that because we know when you do a bone marrow transplant there’s both stem cells and then the passenger T-cells they’re called. So if you have someone who has been a CAR T cell recipient and then they become a bone marrow donor, potentially the CAR T cells could be transferred from the original patient into the recipient and then they would take root and that patient would then also have CAR T cells without ever having gone through the process.
So that would be an unusual case but that’s more than theoretically possible, I think it would happen. But what is a more severe issue is we use engineered viruses such as lentivirus or gamma retroviruses to introduce the CAR T cells. In the past with less sophisticated methods it was possible that those viruses could become replication competent and then that way infect other cells in the body. If that would happen then that could be very serious if the patient had replication competent virus that’s used to make the CAR T cells. That has never happened at this point but it is being watched by all the FDA and other health authorities.
