With one BLA at the FDA, Kite hustles along impressive new leukemia data and sounds a note of caution on safety
CHICAGO — With its lead program for non-Hodgkin lymphoma in front of FDA reviewers, Kite Pharma $KITE today updated their Phase I CAR-T study on relapsed cases of acute lymphoblastic leukemia, noting a strikingly high complete remission rate of 73% as they continue to work on improving the safety profile of their therapy.
The study included 11 patients in the Phase I portion of ZUMA-3. Complete remissions — with the responders all testing negative for minimal residual disease with few or no cancer cells in circulation — were being tracked that lasted from 2 to more than 7.4 months. And the Phase II trial will get underway as Kite looks to expand use of its drug KTE-C19.
“In many ways it sort of confirms what we thought we’d see in the clinical study; very rapid response in 2 to 4 weeks,” Kite CMO David Chang tells me.
Safety, though, remains a key concern. These CAR-T studies have been marred from the beginning by common cases of cytokine release syndrome. In this trial, there were three cases of grade three or higher CRS with one death and six cases of grade three or higher neurological reactions.
That left Kite looking at who should receive tocilizumab within 36 hours post‐KTE‐C19 infusion, and a lower dose of 0.5×10(6) CAR T cells/kg rather than the two doses of 2.0×10(6)/kg and 1.0×10(6)/kg used in the first round.
Kite recently reported that one patient died recently from cerebral edema, or brain swelling, in a separate trial, rattling investors who didn’t like the comparisons to the five cases of cerebral edema that forced Juno to scrap their led program, leaving Kite and Novartis alone in the race to a potential first approval later this year.
Kite has learned a lot about controlling safety, says Chang. Dosing, getting rid of infections before treatment, preparing to battle back CRS are all part of the game plan now.
Cases of cerebral edema have also sparked a debate over whether the CD28 costimulatory domain Kite uses in their CAR-Ts — instead of the 4-1BB domain Novartis uses — could be responsible. Juno’s initial lead therapy also used the CD28 costimulatory domain.
“In a way I can see why some people would say that, but there’s really no basis to draw that conclusion,” says Chang. He adds that there have been cases of cerebral edema with 4-1BB going back to the early academic studies.
Chang also isn’t keen on making cross-trial comparisons.
“There’s a lot of false conclusions that one can draw,” he says. But outcomes can be altered by the characteristics of the patients recruited, whether they were truly chemo refractory or simply relapsed, how many times they had relapsed on multiple lines of therapy and the distribution of disease types, all of which play into the data.
Kite has also been honing its manufacturing skills. In this latest Phase I the manufacturing team was able to get the time it takes to make the therapy from patient cells down to six days. Chang says that the company has the vein-to-vein time down to about 17 days, which he believes can still likely be improved by two or three days.