Editor's Note: This is part two of a two-part series on ways to help more patients realize the promise of immunotherapy drugs.

As far as clinical use goes, checkpoint inhibitors are furthest along, with four approved therapies and dozens of others wending their way through clinical trials.

But perhaps no cancer immunotherapy has captured the public's imagination like chimeric antigen receptor (CAR) T cells, despite the fact that there are not yet any FDA-approved CAR T cells.

Part of the reason is that one of the first rather Lazarus-like remissions in patients receiving CAR T cells was reported not in the New England Journal of Medicine or its peer journals, but in the New York Times. (See BioWorld Today, Aug. 12, 2011.)

Much like the public did not truly notice checkpoint blockers until Keytruda (pembrolizumab, Merck & Co. Inc.) worked for Jimmy Carter, it was the story of 6-year-old Emily Whitehead, who has been in remission since being both the first pediatric patient and the first acute lymphoblastic leukemia (ALL) patient to receive CAR T cells in 2011, who introduced the general public to the treatment approach.

Whitehead, who was near death when she received the CAR T cells, is clearly the best-case scenario. But since her recovery, CAR T cells have proved their mettle more general in the larger, messier, and more convincing world of clinical trials. Even the recent setback involving Juno Therapeutics Inc.'s Rocket trial testing JCAR015 is unlikely to dampen the fervor, with analysts predicting a delay in the timeline but no decreased chance of success. (See BioWorld Today, July 11, 2016.)

At the recent annual meeting of the American Society of Clinical Oncology (ASCO), the University of Pennsylvania's David Porter told his audience that there's a reason for the hoopla surrounding CAR T cells.

The cells, which combine the antigen recognition site of a B cell with the killing ability of a T cell, "has the specificity of antibody therapy, but is a living drug," he said, with long-term follow-up showing that CAR T cells can persist for more than five years after treatment.

CAR T cells could therefore "overcome many of the limitations of conventional chemotherapy, as well as other types of immunotherapy," he said.

Porter said that "there's a great deal we still don't know" about how to use CAR T cells, including the ideal dose, the optimal mix of killer, helper and memory T cells."

But he predicted that "by 2020, CAR T cells will be routine therapies for B-cell malignancies."

Like checkpoint blockade inhibition, there is a lot of interest in combining CAR T cells with other therapies, including checkpoint blockers themselves, since CAR T cells are subject to the same checkpoint inhibition that can stop unmodified T cells in their tracks. There are case reports of patients who initially did not respond to CAR T cells, but had a response after checkpoint inhibitors were added.

Perhaps even more important to their larger success, though, will be safety improvements that are being made in their development.

As CAR T cells expand after transplantation, they release pro-inflammatory cytokines. Those cytokines are important to their success, but there can definitely be too much of a good thing where cytokines are concerned. In the earliest T-cell trials, that cytokine storm regularly put patients into the intensive care unit in its own right. (See BioWorld Today, Dec. 12, 2013.)

Early concerns that the cytokine storm might be necessary for an effective response have proved unfounded. In fact, cytokine storm is "a negative predictor of efficacy, because then we have to turn off the CAR T cells," Steve Harr, chief financial officer of Juno Therapeutics Inc., told BioWorld Today.

Shaking up the immune system at least slightly does appear to be necessary for successful therapy with CAR T cells. "You're unlikely to have a profound response without a fever," Harr said. But "if you develop a fever in the first 72 hours, those cells are growing really significantly . . . and we have ways to slow that down now."

In general, the likelihood and severity of cytokine release syndrome appears to be related to the speed of expansion, which depends on the number of cells as well as the composition of cell types that is infused.

It also depends on the amount of antigen the cells have available to them, which means, somewhat counterintuitively, that sicker patients may need to be treated with fewer cells initially. In a trial of Juno's JCAR014 presented by Cameron Turtle, research associate at Fred Hutchison Cancer Research Center, at this year's ASCO meeting, Turtle described using "risk-adjusted dosing" that takes into account the patient's individual disease burden. Another strategy has been preconditioning the bone marrow with chemotherapy, which increases the durability of the CAR T cells.

The ratio of different T cell types in the treatment also affects toxicity. Optimal ratios are still being worked out, but already "the vast majority of patients in our defined cell trials are being treated as outpatients," Harr said.

The ability to treat in an outpatient setting itself increases the potential reach of CAR T cells. While ALL patients are mostly treated at major medical centers, "a lot of NHL and CLL patients are treated in the community," Harr said.

ARMORED CARS

Like checkpoint inhibitors, CAR T cells work better in some tumors than others for both known and as yet unknown reasons. Their first and greatest successes were in B-cell malignancies, which is due to the fact that B cells express the CD19 antigen. Primus inter pares of the B-cell malignancies is ALL, the type of cancer that befell Emily Whitehead. Porter said that for ALL, the success of CAR T cells has been the equivalent of knocking one out of the park in baseball. Non-Hodgkin's lymphoma (NHL) and chronic lymphocytic leukemia (CLL) are other B-cell malignancies that are "into the outfield," while acute myeloid leukemia (AML) is "on deck."

That antigen is not specific to cancerous B cells, and patients treated with CAR T cells end up unable to make antibodies because they have no B cells at all. Lacking B cells ends up being a surprisingly manageable problem. But in other cancer types, indiscriminate killing of tumor and normal cells won't do, and it will be necessary to identify cancer-specific antigens.

How many such antigens there are, Harr said, is "the biggest unknowable" in expanding the range of tumor types where CAR T cells can be used.

"We're highly confident that there are at least some, and we know of some," he said. But some is an imprecise amount.

Other problems are by their nature easier to tackle. In principle, CAR T cells can be targeted to solid tumors. A phase I trial in glioblastoma has shown some responses, and companies including Kite Pharma Inc. are testing antigens expressed on multiple different tumor types, such as NY-ESO1 and Tn-MUC1, in preclinical studies.

There is also the question of how to overcome an immunosuppressive microenvironment, a problem that is shared with other immunotherapies.

One possibility is to make what Harr termed "armored CARs," which would be armed with both the chimeric antigen receptor and a protein that the cells would secrete to combat the microenvironment's attempts at suppression.

Compared to targeting B-cell tumors, getting the approach to work in solid tumors "will be harder, will take time," Harr acknowledged. "It might be measured in quarters, and it might be measured in years."

But in the long run, he has no doubt that "this technology will work on solid tumors. . . . These are more engineering problems."