There’s a lot of excitement these days about a type of cellular immunotherapy known as CAR-T, a method of modifying peoples’ immune cells to fight cancer. But you could also fill a book listing all the problems its makers will have to solve—how to test, manufacture, and even the define the nature of these cancer-killing cells—before the CAR-T story is a successful one.
These hurdles, not the hype, were the subject of a panel of experts from industry, academia, and the FDA at the Inaugural International Cancer Immunotherapy Conference in New York Thursday afternoon. The panelists included University of Pennsylvania professor Carl June, whose work has led to programs now in clinical testing at Novartis; Adaptimmune executive vice president Gwendolyn Binder-Scholl; and GlaxoSmithKline’s head of immuno-oncology Cedrik Britten, among others.
CAR-T stands for chimeric antigen receptor T cell, which describes an engineered version of the immune system’s attack dogs. CAR-T cells are a patient’s own T cells altered outside the body to be cancer killers, then put back in to go after tumor cells.
CAR-T therapies from Novartis, Juno Therapeutics (NASDAQ: JUNO), and Kite Pharma (NASDAQ: KITE) have produced impressive results so far for certain blood cancers, leading to long-lasting remissions in some patients.
But the field is early in its development. Researchers are trying to figure out how to make these therapies useful for more common cancers, such as lung, breast, and ovarian, and how to mitigate the overactive immune responses they can cause. Biotechs and pharma companies developing autologous therapies—which modify the cells of each individual patient—are wrestling with how to manufacture and distribute them at scale.
But a different, larger question looms, and it gets to the heart of why autologous T cell therapy is truly a new medical frontier. The cells that are delivered back into the patient are not what ends up doing the bulk of the therapeutic work.
Once the modified cells are put back into a patient, they seek out their cancer targets—probably proteins that dot the surface of a tumor cell—bind to them, and kill the tumor cells. Then the T cells divide. A subset of new cells is formed to lead the cancer attack, and they effectively become the therapy. But all that activity could vary widely patient to patient, depending on the individual’s immune system. That makes it very tough for the FDA to regulate.
“How do you define a label for a product that bears very little, if any resemblance to what you put into a patient?” Michael Kalos, Eli Lilly’s chief scientific officer of cancer immunobiology, noted on the panel.
There was a lot of concern on the panel about this potential problem, but not everyone in the industry feels this way. For instance, I asked Chuck Wilson, who wasn’t on the panel but runs a cell therapy startup, Unum Therapeutics, and was one of the key CAR-T dealmakers at Novartis, for his thoughts. He said that if a company’s manufacturing process leads to a product that yields “reproducible effects” on patients in trials, then it should be just as sound when used to make a marketed product at a large scale—regardless of how different that product is from the cells first extracted from a patient. Wilson said that it’s expected that the cells taken from each patient will be different, as will the new cells that divide in the patient’s body after treatment.
“But if clinical testing shows that this heterogeneity is tolerable with respect to efficacy (consistently active) and safety (consistently well tolerated), it’s a viable product,” he said.
The panelists Thursday noted how T cell therapies could throw a wrench into a typical, and crucial, clinical strategy. Early in the clinical testing of a drug, companies usually run what are known as dose escalation studies. Different doses of products are tested, low to high, to establish a trend of responses, see what safety issues pop up, and pick the optimal dose to move forward.
But because CAR-T cell populations expand once they’re put back into a patient’s body, doses are harder to define. What’s more, cranking up a dose for such a powerful therapy could be dangerous. “Are classical trial designs applicable, or do they have to be changed?” asked GSK’s Britten. “You can not have a simple dose escalation [study] with a drug that replicates.”
Adaptimmune’s Binder-Scholl called for more guidance from regulators to help figure out a more standardized scheme for dose escalation studies.
“I think the biology is going to make [that type of guidance] awfully challenging,” says Juno’s chief financial officer Steve Harr, who also wasn’t on the panel. “I would like to think over time we get into something a bit more predictable, and maybe we have some type of a standard, but we’re very early in this process.”
Even if the clinical and regulatory challenges are met, and products come to market, their owners are likely to face a big commercial problem that’s more familiar to the consumer technology world: Constant new iterations of products. Put another way, CAR-T cells could be the iPhones of the biotech world.
CAR-T products are going to have “rapid” life cycles because there are going to be consistent improvements to them, Carl June said Thursday. “Hopefully the field can move forward at about the pace that cell phones change.”
Some of those improvements are already in progress. As Xconomy reported in July, researchers at Memorial Sloan-Kettering Cancer Center in New York have been working to get into clinical testing with a CAR-T product that has two “extra” modifications: A chemical signal that recruits the patient’s unmodified T cells to help with the attack; and a self-destruct mechanism that clinicians can trigger in case the cells go haywire.
The trial, for relapsed ovarian cancer, could be the first test of a so-called armored CAR.
CAR-T therapies could also see improvements in combination with other drugs or with more efficient manufacturing processes. But one problem with making manufacturing changes is that regulators can demand a new set of clinical trials, arguing in essence that the new process has created a new product. June said he hopes for “flexibility” from regulators, but no one really knows how they’ll react.
“I don’t think we can just willy-nilly make changes and expect regulators, patients, and physicians to accept them because the biology makes sense,” Harr says. “We’re going to have to define the associated clinical benefit.”
These are just a few of the myriad questions CAR-T products will have to answer as these therapies make their way to the FDA. Novartis is aiming for an FDA approval next year. Juno has said that it aims to file papers either late next year or early 2017 seeking approval of its first product, and Kite is targeting a potential launch that year as well. All of these products are designed to treat blood cancers like acute lymphoblastic leukemia, diffuse large B-cell lymphoma, and follicular lymphoma. You can expect the next wave to come quickly afterwards.
“We do view this a bit like software, where you will have relatively rapid cycles of innovation,” Harr says.
Photo of healthy T cell courtesy of NIAID via Creative Commons license.