On June 2, 2021 Timunity reported that the multifaceted campaign to create a CAR-T for solid tumors has run into a lethal problem (Endpoints, Tmunity Therapeutics, JUN 2, 2021, View Source [SID1234583392]).
Schedule your 30 min Free 1stOncology Demo!
Discover why more than 1,500 members use 1stOncology™ to excel in:
Early/Late Stage Pipeline Development - Target Scouting - Clinical Biomarkers - Indication Selection & Expansion - BD&L Contacts - Conference Reports - Combinatorial Drug Settings - Companion Diagnostics - Drug Repositioning - First-in-class Analysis - Competitive Analysis - Deals & Licensing
Schedule Your 30 min Free Demo!
In exclusive interviews, Tmunity founders Carl June and Oz Azam tell me that they have had to scrap their lead program for prostate cancer following the deaths of 2 patients earlier in the year — the last a little more than a month ago — after they experienced lethal bouts of neurotoxicity. What started as a voluntary halt by Tmunity in the face of these deaths flipped to an FDA-enforced red light for the clinical trial, while the biotech scrambled to rework its CAR-T design with a plan to get "version two" back into the clinic as soon as possible, with a new IND planned for the second half of the year.
In the meantime, the company is restructuring and laying off an undisclosed number of staffers while reevaluating a timeline for clinical development which has temporarily iced plans to go out with an IPO.
But that internal reorganization, the founders say, isn’t nearly as important as laying out a description of what happened — first in this interview and later in a peer-reviewed journal, as soon as that can be arranged.
The critical mission now, says Azam, is to get the word out about a safety issue which he believes may be characteristic of other programs in the field, a hidden landmine threatening more patient deaths as others pursue the same mission of cracking solid tumors with an engineered T cell.
"In our lead program in prostate cancer we saw profound clinical responses like things that made me and Carl June sit up in our chairs and say, ‘Is this possible?’" Azam tells me. "When we give them CAR-Ts we see phenomenal drops in their PSA levels within days of getting the treatment. To the point where I was shocked because I never thought we’d see this in a solid tumor setting."
But those gains were swept aside with the deaths of 2 patients in the small study.
"What we are discovering is that the cytokine profiles we see in solid tumors are completely different from hematologic cancers," says Azam. "We observed immune effector cell-associated neurotoxicity — ICANS. And we had 2 patient deaths as a result of that. We navigated the first event and obviously saw the second event, and as a result of that we have shut down the version one of that program and pivoted quickly to our second generation."
June, the famous Penn professor who led the way to the first approved CAR-T at Novartis, and Azam, who once led that Novartis charge before he jumped over to launch the Penn spinout, have both grappled with much the same situation during the first generation of CAR-Ts for blood cancers. Originally it was cytokine release syndrome that bedeviled investigators and triggered patient deaths. But the field largely mastered that threat. Now it’s a similar neurotoxic threat — ICANS, explains Azam, or macrophage activation, in June’s description — that’s causing deadly effects.
"We didn’t see this coming until it happened," says June. "But I think we’ll engineer around just like we did with tocilizumab back in 2012." And once past, he adds, they can go on to developing safer off-the-shelf therapies, which is the longer term goal of the company.
First, they have a big challenge ahead.
This particular drug is designed to use engineered patient cells to target PSMA while also using a dominant negative TGFbeta receptor to block a key checkpoint involved in cancer, allowing persistence required in solid tumors. There were 24 patients recruited for the dose-escalating study with plans to release data from high-dose cohorts later in the year. And the study is in the process of being wound down.
"We are going to present all of this in a peer-reviewed publication because we want to share this with the field," says Azam. "Because everything we’ve encountered, no matter what, whether you’re an NK style company, an iPSC company, whatever platform you’ve got, people are going to encounter this when they get into the clinic and I don’t think they’ve really understood yet because so many are preclinical companies that are not in the clinic with solid tumors. And the rubber meets the road when you get in the clinic, because the ultimate in vivo model is the human model."
"It’s a really important dialogue to have with the field," says Azam. He explains:
It’s complex, it’s elaborate, there are different cytokines that are involved that we are seeing. What we are also seeing is this picture of this ICANS, this neurotoxicity picture, this immune effector cell-associated neurotoxicity syndrome, which is kind of like a spectrum of CRS.
You have the classic cytokines that get elevated and then a delayed picture occurs after a few days, typically between 5 to 7 to 10 days in some patients. Where there is an endothelial receptor activation in the brain, the blood-brain barrier starts to break down and you get this massive inflammatory response in the brain, which leads to this neurotoxic syndrome and in our case we had 2 patients where it was fatal.
It’s caused really between this imbalance between macrophages and T cells cross talking to each other, and in a way in which the cytokines really are quite elevated which leads to this picture. We’re decoding it, we’ve decoded it in the past but there’s a lot more to decode here. As importantly we actually think we have the solutions to it, because we’ve been working on it for 4-and-a-half years.
The solution, he adds, will require pharmacotherapeutic interventions to prevent the crosstalk and additional engineering of the CAR-T.
We have these proprietary modular engineered vector systems and we’ve developed construct banks. We’ve been building this steadily over the past 4.5 years, where with different costimulators, armors, enhancers and switches we think we actually have the solutions in place.
June is taking a wait-and-see approach on resolving this challenge, but he believes it can be done.
"I think the bottom line is we’re finding that toxicity is different in liquid tumors like leukemia and myeloma from solid tumors like pancreatic cancer or prostate cancer," he adds. "I think it’s what it is called macrophage activation syndrome, which is related to cytokine storms or CRS, it’s probably caused by different cytokines."
They’ve seen it in bad viral infections, or where kids with genetic defects would suffer from macrophage activation syndrome. The macrophages make different cytokines and "what’s really scary" is that triggers neurotoxicity in the brain.
Says June: "We didn’t really see this in blood cancers."
In blood cancers, they found that IL-6 was the "smoking gun," in June’s words, which was tamed by tocilizumab. But in these new cases, says June, tocilizumab doesn’t work. That may be simply explained by the fact that tocilizumab doesn’t get into the brain well. Or some other explanation may apply. They still need to find the smoking gun with solid tumors.
With the first wave of cytokine storms in CAR-T, adds June, clinicians were often blindsided by the severe reactions that killed patients. In these select trial sites for Tmunity, he adds, everyone is acutely aware of what that is. But they couldn’t do anything about these cases — both deaths and a third case that occurred in the group as well.
"It was clearly refractory to the standard management," says the Penn scientist. "We need to find a way to work around this and probably dial down the potency of the CAR as we currently have it configured."
In the deeper analysis to come, adds June, investigators will examine what distinguished these patients, why they reacted the way they did, and how that can be prevented again.
For now, sums up Azam, the plan is to stick to the drawing board and produce a new IND for version 2. A follow-up approach "that we think will produce a safer and kinder CAR."
Azam’s backed by one of the most impressive boards in biotech, with ex-Vertex CEO Jeff Leiden as chairman, billionaire Sean Parker and VC vets Beth Seidenberg and Jorge Conde on as directors. They were all drawn by the promise of mapping a CAR-T breakthrough in solid tumors, where the rewards can be much greater than the blood cancer pioneers have found.
After early setbacks dating back to the 1990s, the field has been experiencing a renaissance of sorts. Researchers have mapped out the safety landmines and efficacy hurdles keeping solid tumors out of reach for CAR-T cells, and are just starting to put next-generation constructs into humans. But many of them are also acutely aware of biology’s tendency to surprise and have likely been anxiously anticipating jolts like Tmunity’s announcement as part of the "iterative cycles of learning" — not that it makes patient deaths any easier to stomach.
Notably, Poseida Therapeutics reported last year that a patient had died of hepatic failure after receiving its PSMA-targeting CAR-T and showing symptoms of macrophage activation syndrome, triggering a clinical hold. The Phase I trial resumed after the biotech amended the trial design, including the patient selection criteria as well as the frequency of monitoring and laboratory testing.