Cellectis interview – if at first you don't succeed...

A 28% share price fall in the month in which Ash conference abstracts were live is not what André Choulika, chief executive of Cellectis, wanted to see. And, given the earlier run-up, the dip must have been down to missed expectations.

Mr Choulika insists that the markets missed a trick by focusing on relapses in leukaemia studies of the Servier/Pfizer-controlled lead asset, UCART19, presented at Ash. Two key findings were largely overlooked: dosing is much lower than competitors’, and remission in a subject given a second UCART19 infusion points to what is now a central part of Cellectis’s allogeneic CAR idea.

The vast majority of autologous CAR-T therapies are given once only, and rely on cell engraftment and proliferation to give long-lasting persistence and efficacy. “But why should you dose a patient one time only?” asks Mr Choulika in an interview with EP Vantage on the sidelines of Ash.

Attack and consolidate?

“Maybe you should give – as with any therapy – two attack doses and then consolidation doses,” he suggests. Redosing an autologous CAR is possible too, but is complicated by the need for prior collection of sufficient cells, and growing these up in advance to make sufficient product to be infused several times. The advantage of an allogeneic product, available on demand, is obvious.

The redosed patient had had adult ALL, and after a CD19-positive relapse, allogeneic stem cell transplant and second UCART19 infusion went back into molecular remission, Ash heard. There were five responses in seven subjects in this trial, but only three are ongoing.

A similar pattern was seen in a childhood ALL trial, where all five subjects went into remission, but one died and two relapsed before six months – one with and one without antigen loss. So is Cellectis seeing a typical high rate of response, accompanied by unwelcome waning of efficacy?

Mr Choulika disagrees: “I consider this still as dose-finding,” he says, stressing that in adult ALL the current dose is just 100,000 cells/kg – versus 2-5 million cells/kg in Novartis’s Eliana trial of Kymriah. Moreover, CD52 knockout, to make UCART19 resistant to lymphodepletion with Campath, is effective in only 60% of cells, meaning that the UCART19 dose is really 60,000 cells/kg.

And he insists that UCART19 complete remission rates at 28 days – “the time when Novartis charges the patient” – are already comparable. Meanwhile, previously cited problems of allogeneic CAR-T therapy – severe graft-versus-host disease, lack of engraftment – have not materialised, he states.

Selected Cellectis-originated allogeneic CAR-T assets
Project Antigen/disease Ownership Status
UCART19 CD19/adult & childhood ALL Liencesed to Servier & Pfizer Phase I
UCART123 AML & PBDCN Cellectis Phase I off clinical hold, about to resume
UCART22 CD22/CD19-ve relapsed ALL Cellectis IND due Q1 2018
UCARTCS1 CS1/multiple myeloma Cellectis IND due Q1 2019
UCART38 CD38/multiple myeloma & T-cell lymphoma Cellectis Preclinical
UCARTCLL1 CLL1/AML Cellectis Preclinical
UCARTBCMA BCMA/multiple myeloma Pfizer* IND possible in 2018
UCARTEGFRvIII EGFRvIII Pfizer* Preclinical
Note: *part of 15-target collaboration.

But there have been deaths not due to disease progression: one adult ALL subject died of neutropenic sepsis, one ALL child of transplant complications. And recently trials of the wholly Cellectis-owned UCART123 were suspended after a death due to cytokine storm and capillary leak.

“The patient that died was very sick, and it would have been better not to enrol him,” Mr Choulika says. The UCART123 trial is now off clinical hold and about to resume infusion, at a lowered dose of 62,500 cells/kg.

One possible problem with UCART123 is off-tumour effects, since CD123 is expressed on healthy cells including myeloid progenitors, which give rise to red blood cells. The chief exec accepts this complication, but says it too could be overcome with repeat dosing.

“For me the only procedure with a target like this is a hit and run: inject the CAR, the CAR carves out everything ... and then bone marrow stem cells replenish the myeloid progenitors,” he says, adding that a persistent CAR makes no sense in this setting. “It’s better to give small injections and let the blood rebuild.”

No doubt investors will be paying close attention when UCART123 trials resume. In the meantime they can focus on the applicability of allogeneic CAR-T therapy: does it have broad market potential, or will it be limited only to patients who cannot yield sufficient T cells in apheresis, for instance?

There is no doubt in Mr Choulika’s mind: “Imagine a doctor at the hospital. A patient comes in. You just open a freezer, pull out a vial and inject. I don’t think that a decade from now a single autologous CAR-T therapy will survive; they will all disappear.”

To contact the writer of this story email Jacob Plieth in London at jacobp@epvantage.com or follow @JacobPlieth on Twitter

Share This Article