As safety looms larger in CAR-T research, study highlights how small tweaks can make it easier to use
For all the promises of CAR-T therapy, significant side effects — notably life-threatening episodes of cytokine release syndrome and neurotoxicity — have limited its use in closely monitored settings. As researchers seek new approaches to tamp down safety concerns, a small clinical trial published in Nature Medicine has offered the first glimpse of hope that small tweaks on existing molecules can make CAR-T safer while maintaining its cancer killing ability.
The hypothesis here, University of Southern California professor and lead study author Si-Yi Chen tells me, is that CAR-T doesn’t need to excrete so much cytokines to work. So they took Novartis’ FDA-approved Kymriah and made a variant with slower cytokine production, observing significant difference in how the therapy behaved.
Among 25 patients enrolled in the Phase I study — which took place in China in collaboration with Marino Biotechnology — none of them experienced cytokine release syndrome or neurotoxicity. That compares to a 57% CRS rate and 39% neurotoxicity rate in Novartis’ Phase I trial for its product by Chen’s count.
“All the patients don’t need supporting care, basically,” he said, hinting at the potential for CAR-T administration as an outpatient procedure.
Meanwhile, the tweaked construct appeared to “achieve efficacy comparable to other clinical trials” of established CAR-Ts, Chen told me, with the longest response tracked for about two years. At the highest dose (2–4×10^8), 6 of 11 patients saw complete remission.
The improvements can all be attributed to a switch up in amino acid sequence, Chen said. Structurally, chimeric antigen receptors (the CAR that gets engineered into T cells) consist of an extracellular antigen recognition domain, an intracellular activation domain, and a hinge and transmembrane region that connects the two. By adding a few amino acids in the hinge region, Chen’s team was able to create a CAR-T that proliferated more slowly, giving the patient’s body more time to clear cytokines in the blood.
Whether the remissions induced by this supposedly safer CAR-T will hold up in the long term will be crucial. And Chen wants to pursue a global, multicenter Phase II with 80 to 100 patients to prove that the effects were not by chance.
Having begun the project six years ago — at a time when the focus was squarely on proving efficacy — Chen counts himself an early laborer on a topic that Fred Hutchinson and the National Cancer Institute are now working on. Earlier this year at ASH, a team of NCI scientists reported “strikingly” lower neurotoxicity rate using a CAR construct similarly comprising a hinge region from a CD8α molecule.
“Altering CAR hinge and TM domains can affect CAR T‐cell function and is a promising approach to improve the efficacy to toxicity ratio of CAR T‐cells,” they wrote in an abstract.
Chen admits that he has yet to raise the funds needed for the Phase II, or sort through the patent issues surrounding the tweaked compound, but he’s open to any options that will move the work forward. “Scientifically I feel it’s justified to take a look,” he said.