How Many CAR-Ts Do We Need to Test Drive? Save

Hi, I'm Dr. Janet Pope reporting at RheumNow at EULAR 2026 in London, England. I want to talk about how many CAR T cars do we need to test drive, and the reason I'm talking about this is there was a lot going on at this meeting on CAR T therapy. And in fact, I searched and there were 93 abstracts at EULAR 2026 that had CAR T in the title. I also went to a wonderful session on June 5th on the what is new, or WIN session, by Dr. Hector Chinoy, and he talked about CAR T and inflammatory myositis, and many of the concepts that he talked about I've tried to internalize to update all of us.

So with this many abstracts I think we know lots, but we have more questions than answers. So one of the first questions is who should we select for treatment for CAR T? I think in general the consensus would be certain diseases might do better than others, such as lupus — maybe myositis would do better than maybe RA or systemic sclerosis — but I'm just giving you a gestalt. We really don't know within an individual.

The other thing on selection that's important is we would like someone — in general the perfect candidate would be younger and not having comorbidity, high disease activity, low damage, and they haven't failed everything but they have poor prognosis. So we know the writing's on the wall, say, in a lupus nephritis patient who hasn't responded well at all in three months of triple therapy for lupus nephritis and is adherent.

Then the next question is well, what do we want to inhibit? So there's a lot of CARs out there. So CD19, CD20, BCMA, NK cells, dual targets, even a triple target. I learned something called an armored CAR, which is a CD19 CAR T therapy but also giving cytokine inhibition, which is supposed to increase the durability of the response. So which target to choose? I don't actually know, and I don't think that in these trials we know who will respond to what.

Better then — do we prime with chemo or not, and do we give leukapheresis? And leukapheresis will depend on the type of CAR, so if you're giving the traditional one that was around the longest — autologous — you have to get the cells, cell-sort, so you get the CARs made against, say, CD19 if that's the target. And then you give chemo to get rid of all their cells, at least in the circulating compartment and some within organs as well. So that's autologous, and there's no graft-versus-host or anything like that because it's your own cells. But there's a time frame where you wait. If it's not made right and the patient's off therapy, they might be flaring more, or not, because they've been primed with chemotherapy.

There's allogeneic, and it's kind of attractive because it's off the shelf. So we would assume in something like that — say, again, the target is CD19 — that it would be a CAR that would look like most people's CD19, that they could basically be distributed and destroy the CD19s within the patient.

Then there's this new concept of endogenous, so in vivo, where you're putting it into the patient. It could be in vivo with mRNA vector or even with viral delivery.

The next issue really is safety — the short term and the long term. So it's important to know the timing of reactions. So cytokine release syndrome, or CRS, takes a few days to a week or two to occur in general. Then there's ICANS — that's immune effector cell-associated neurotoxicity syndrome — that's more severe; the patient can be obtunded, having other neuro side effects. And that problem might actually need tocilizumab. Either CRS or ICANS might or might not need glucocorticoids.

Then there's a new problem that's been described called LICATS — that's local immune effector cell-associated toxicity syndrome. LICATS is a mouthful to think about, but it's more in our autoimmune connective tissue disease patients, where they might have an organ wake up for a while that is transient. And the thought is it's when cells are coming up and then more normalizing over time, but I'm not exactly sure why it occurs.

Then what do we do if there's a recurrence? Do we have frozen cells if it's autologous? Is the patient going to be resistant? How long will it work for? And what will it work for? Will it work for every manifestation or just some? Will it work for lupus nephritis and not maybe for pain and fatigue? We don't know all that.

Then what to select in different diseases? Is there one optimal route of administration, like the autologous versus the allogeneic? Is it whatever you can get your hands on? Are we priming?

And with that comes the new expectations. So when I naively heard about CAR T therapy when it first came in oncology more than a decade ago, I thought, "Wow, this is amazing. It's a cure. It's quite a neat immune way to mop up a lymphoma or leukemia that has been recalcitrant." And so the idea in CAR T back then was immune reset. Now they're talking about a new buzzword phrase — I hadn't heard it till this meeting — called immune dimming. And
what that means is the immune system has improved, but you're not getting a patient say to remission and certainly not in drug-free remission that you might have to restart something.

Then a final long-term concern, not for everybody, but there have been reports mostly again coming from the oncology literature of an increased risk of cancer um years after getting CAR T therapy. But in oncology literature, the biggest risk of cancer is having had cancer. The next biggest risk is having had um huge cocktails of chemotherapy.

So I think we have a lot to think about, but if there's 93 abstracts just at this meeting, this is still a really exciting field and right now I don't know how many CAR T's we need to actually test drive. Please follow me at Janet Berdо and follow at RheumNow. Thank you.