FDA hits Cellectis’ off-the-shelf CAR-T program with clinical hold after first patient treated in phase 1 trial dies

The FDA has placed a clinical hold on two phase 1 trials of Cellectis’ UCART123 after learning of the death of one patient. Development of the off-the-shelf CAR-T therapy is now in limbo while Cellectis works with the FDA to redesign the protocol to mitigate the risks identified in the first weeks of the trials.

Doctors at the MD Anderson Cancer Center dosed the first blastic plasmacytoid dendritic cell neoplasm (BPDCN) patient with CD123-targeting CAR-T UCART123 on August 16. The patient died nine days later. 

Initially, the 78-year-old man responded to the lowest dose of UCART123 without complication. On day five the patient suffered a grade 2 cytokine release syndrome (CRS) and grade 3 lung infection. On day eight he experienced a grade 4 capillary leak syndrome (CLS) and a CRS that, despite treatment with corticosteroids and tociluzumab, played a central role in his death the next day.

The only patient treated with UCART123 in the other phase 1 trial experienced similar, albeit less severe, reactions. That patient, a 58-year-old woman with acute myeloid leukemia (AML), suffered a grade 3 CRS and grade 4 CLS nine days after treatment with UCART123. The adverse events put the patient in intensive care but had cleared up by day 12.

The list of adverse events suffered by the patients suggests UCART123 may be affected by safety issues both general to CAR-Ts and specific to its targeting of CD123. CRS is a known and, in the case of autologous products, generally manageable side effect of CAR-Ts. The process for managing the events is well established enough for Roche to have won approval for tocilizumab—also known as Actemra—as a treatment for CAR-T-induced CRS.  

It is conceivable Cellectis can prevent further patient deaths by lowering the dose of UCART123 and follow the example set by other CAR-T trials by intervening earlier and more aggressively to treat CRS.

“We think one of the key learnings from the CD19 CAR-T trials is the early administration of steroids; however, we think early administration of steroids was withheld due to potential negative impact on cell persistence. In our view, [Cellectis] may need to re-evaluate the timing of steroid administration and be more aggressive in treating CRS,” analysts at Jefferies wrote.

The way to manage the adverse events that may be tied the candidate’s targeting of CD123 is less clear. 

Stemline Therapeutics’ SL-401 is the precedent in this case. Three patients in clinical trials of CD123-directed therapy SL-401 have died after experiencing severe cases of CLS, a syndrome characterized by the leaking of blood plasma through capillary walls and into the surrounding tissue. Stemline added dosing and safety parameters to the lead-in stage of the study following the first two grade 5 cases of CSL. But those precautions failed to prevent the third death.

Cellectis now needs to figure out its own precautions before it can resume enrolling the 71 BPDCN patients and 155 AML patients it plans to treat across the two phase 1 trials. The data safety monitoring board has already proposed lowering the dose of UCART123 and capping the amount of the chemotherapy cyclophosphamide patients receive during the preconditioning stage. The two patients treated to date received cyclophosphamide alongside fludarabine.

The one bright spot for Cellectis in the otherwise grim safety data is the lack of reports of graft-versus-host disease (GvHD), a complication that arises when the immune system rejects an allogeneic transplant. Cellectis’ use of T cells from donors, rather than patients themselves, means GvHD is a potential concern. Cases of the condition would scuttle Cellectis’ ambition to capture the CAR-T market from Novartis and soon-to-be Gilead unit Kite Pharma by industrializing the CAR-T production process. 

Shares in Cellectis opened down about 30% in Paris.