Allogeneic CAR-Ts could serve an important niche – particularly for patients who can’t receive autologous CAR-Ts – but the earliness of the data makes conclusions tough to draw.
Since their transition from small, academic clinical trials to commercial production, CAR-T cell therapies have won recognition as a novel and highly effective therapeutic modality, particularly in blood cancers. But as one might expect, that journey has also had no shortage of speed bumps.
Many of the difficulties CAR-Ts have encountered are because of their inherently complex logistics and manufacturing as they are autologous products made from patients’ own T cells. Allogeneic or “off-the-shelf” CAR-Ts have been touted as offering a potential solution to some of those hurdles. But while doctors and executives from companies developing allogeneic CAR-Ts expressed excitement about their potential, they said it may be a while before those therapies are ready for widespread use.W
The two autologous CAR-Ts on the market – Novartis’s Kymriah (tisagenlecleucel) and Gilead Sciences’ Yescarta (axicabtagene ciloleucel) – respectively won Food and Drug Administration approval in 2017 for acute lymphoblastic leukemia (ALL) in children and young adults and diffuse large B-cell lymphoma (DLBCL), respectively. A DLBCL approval for Kymriah followed last May. Both therapies are autologous in that they involve apheresis — or extraction — of a patient’s own T cells, which are sent to a central manufacturing plant that engineers them to target the cancer cell-surface antigen CD19 before shipping them back for infusion. By contrast, allogeneic CAR-Ts are made from donor T cells and designed to be used “off the shelf,” thus obviating the cumbersome process to produce the autologous products.
The data so far for allogeneic CAR-Ts are early, but some of them suggest the cells may not last inside the body as long as their autologous counterparts.
At the December 2018 annual meeting of the American Society of Hematology in San Diego, Allogene Therapeutics presented preliminary Phase I data on its lead product, UCART19, an allogeneic CAR-T that the South San Francisco, California-based company is developing with France-based Servier. Among 16 evaluable pediatric and young adult patients with acute lymphoblastic leukemia, 88 percent achieved a complete response with complete or incomplete recovery of their blood counts, also known respectively as CR and CRi. That’s only 14 patients, but in percentage terms roughly in line with the CR/CRi rate of Kymriah.
Of the 12 of those patients who achieved minimal residual disease negativity, five remained in remission 4.5 to 16.4 months after receiving UCART19, while seven progressed. Median persistence of the CAR-T cells inside the body was 28 days, and one patient treated at a higher dose level showed a long persistence, with a very low level of detection after 120 days. Allogene says on its website that it may initiate registration-directed trials of UCART19 in ALL in the second half of this year.
By contrast, an analysis last year found that Kymriah persisted for up to 20 months, with a median persistence of 168 days. Of 75 patients with a median follow-up of more than a year, 81 percent remained in remission.
In an email, Allogene CFO Eric T. Schmidt wrote the company does not view UCART19’s persistence of response as being inferior to autologous CAR-Ts like Kymriah and Yescarta.
“For a variety of reasons, most of the ALL patients in our study who achieved a complete response went on to receive a stem cell transplant (ASCT),” he wrote. “At that point in time our AlloCAR T cells are wiped out, the experiment ends, and the (potential) benefits of ASCT take over.” He added that any comparison between UCART19 and autologous CAR-Ts will be easier in the setting of non-Hodgkin’s lymphoma, where ASCT is less of a confounding variable.
Other than the ASH abstract, there is insufficient information to say whether lack of persistence is the case, said Dr. Fred Locke, leader of the Immune Cell Therapy initiative at the Moffitt Cancer Center in Tampa, Florida, in a phone interview. It is possible that the cells do not persist, and the durability is thus lower than it would be with autologous CAR-Ts.
“I would point out, though, that there are many variables that are beyond just the fact that [cells] are allogeneic or autologous,” Locke said.
David Gilham, vice president of research and development of Belgian allogeneic CAR-T developer Celyad, echoed Locke’s analysis in an interview at last month’s J.P. Morgan Healthcare Conference in San Francisco.
“Based on the scant information available at the moment, response rates and durability [appear] slightly lower, but it is too early to say if that’s the case.” he said.
However, if lack of persistence is indeed the issue, it could be due to the immune system recognizing allogeneic CAR-Ts as foreign and eliminating them, said Rick Fair, CEO of Bellicum, another company developing allogeneic CAR-Ts.
“This is a very tricky problem – how do I create foreign T cells that can be effective but not be foreign,” he said in an interview at J.P. Morgan.
That’s where technologies like CRISPR/Cas9 will become important, in order to lower the chance of an immune response killing the CAR-Ts, said Dr Abhinav Deol, in a phone interview.
Deol, an oncologist at the Karmanos Cancer Center in Detroit, took part in the study that led to Yescarta’s FDA approval for lymphoma. Otherwise, he said, allogeneic CAR-Ts have to come from a genetically matched donor, like allogeneic stem cell transplants, in order to lessen the chance of the immune system killing them. But use of CRISPR in CAR-T therapies is still very early and will take some years to develop, Deol said. In UCART19, two genes are knocked out to allow its administration in non-genetically matched patients, according to the ASH abstract.
Companies using CRISPR/Cas9 in allogeneic CAR-Ts include Switzerland’s CRISPR Therapeutics and Cambridge, Massachusetts-based Intellia Therapeutics. Sangamo Therapeutics has a partnership with Gilead to use another genome editing technology, zinc finger nuclease, to develop cell autologous and allogeneic cell therapies for cancers, including CAR-Ts.
Even if allogeneic CAR-Ts have a ways to go, the drawbacks of their autologous counterparts still create a definite clinical need for them.
“The problem that happens right now with autologous CAR-T cells is when patients … don’t have an an effective way of collecting T cells needed for manufacturing,” Deol said.
As such, allogeneic therapy will likely be an option for patients to whom autologous CAR-Ts can’t be delivered, Gilham said, adding that “they can very easily coexist.”
The two to three weeks between apheresis and infusion is another reason allogeneic CAR-Ts can be attractive. Allogeneic cells, Deol said, are ready to use like an ordinary medicine and can be taken off the shelf and infused. This can be a real advantage for patients whose disease is progressing and not controlled and thus don’t have the luxury of time, he said.
In its second quarter 2018 earnings in July, Novartis disclosed that it had encountered unspecified “variability” in the manufacture of Kymriah, particularly for DLBCL. In the company’s earnings call, Novartis Oncology CEO Liz Barrett said the company had been able to deliver the product to “the majority of patients.” Sales of Kymriah were $76 million in fiscal year 2018, according to the company’s financial reports.
But despite enthusiasm about the potential of off-the-shelf CAR-Ts, they do not appear ready for prime time. “If I had to guess, we may see that allogeneic CAR-T cells can work,” Locke said. “It may be some time before they’re able to work to the degree of FDA approval.”
He said development will mostly focus on CD19-targeting CAR-Ts, meaning they’ll be up against the better-established Kymriah and Yescarta. Thus, while the allogeneic cells are likely to eventually win FDA approval, the question that remains is whether they will be able to show similar efficacy or better toxicity profiles than the already approved autologous CAR-Ts.