How Durable Are CAR T-Cell Therapies?

Renier J. Brentjens, MD

The use of chimeric antigen receptor (CAR) T-cell therapies for the treatment of hematologic malignancies is still in its early stages, but when the FDA approved tisagenlecleucel (Kymriah) and axicabtagene ciloleucel (axi-cel; Yescarta) in 2017, this gave hope to oncologists and patients with some types of leukemia and lymphoma who have exhausted all other options. The findings and approvals for this approach were so welcome that the American Society of Clinical Oncology (ASCO) named CAR T-cell therapy as the Advance of the Year as part of the Clinical Cancer Advances 2018: ASCO’s Annual Report on Progress Against Cancer.¹

These new drugs are categorized as adoptive cellular therapies. In the 2 agents approved thus far, the patients’ T cells are reengineered to target cancer cells expressing CD19 in certain B-cell leukemias and lymphomas. Study results show certain patients respond well, particularly compared with historical averages, but because experience with this novel therapy is still limited, there are questions about response durability.

Recent Findings for CAR Therapies

The clinical trials have focused on initial clinical outcomes and long-term studies are not yet available. There also may be some questions about the reported responses being for the patients treated, rather than a percentage of the intentto-treat population. So, it is still not known how often these treatments provide complete remission, how long patients can expect to remain in remission, and how the durability of the treatment matches the high cost of treatment.Tisagenlecleucel was the first FDA-approved CAR T-cell therapy; the agency approved it in August 2017 to treat children and young adults (up to 25 years) with B-cell precursor acute lymphoblastic leukemia (ALL) that is refractory or in second or later relapse.² The therapy was approved based on outcomes among 63 patients who received a single dose of tisagenlecleucel during the phase II single-arm international ELIANA trial. Among the infused patients, 83% achieved a remission, including 63% with a complete remission (CR).

In updated results from the study published in February 2018, investigators reported an overall remission rate of 81%, including 60% with a CR, among 75 patients who had received a tisagenlecleucel and had at least 3 months’ follow-up.³ The rates of event-free survival (EFS) and overall survival (OS) were 73% and 90%, respectively, at 6 months. At 12 months, the EFS rate was 50% and the OS was 76%. In addition, tisagenlecleucel was evident in the blood for up to 20 months following infusion and the median duration of remission was not reached. These results compared favorably with historical findings for approved therapies for relapsed B-cell ALL, where response rates have ranged from 20% to 39%, investigators noted.

Similarly, long-term data from a phase I trial involving 53 adults with relapsed B-cell ALL who received an infusion of autologous T-cells expressing 19-28z CARs showed high response rates.⁴ These results showed 83% of patients achieved a CR; after a median follow-up of 29 months, the median EFS was 6.1 months and the median OS was 12.9 months.

The initial response rate for CAR therapy is very high, according to Renier J. Brentjens, MD, PhD, but the relapse rate is significant, he said. “Depending on which study you’re looking at, we’re getting more longterm follow-up now, and it looks like maybe half of the patients can relapse and some of those patients relapse with a disease that no longer expresses the CD19 target,” he said in an interview with OncologyLive^®. Brentjens, a leading researcher into CAR therapies, is director of Cellular Therapeutics at Memorial Sloan Kettering Cancer Center in New York, New York.

Terry J. Fry, MD, principal investigator on several CAR T-cell trials at the Pediatric Oncology Branch at the National Cancer Institute (NCI), said in an article published by the NCI, that “even in those who experience a complete response, up to a third will see their disease return within a year.”⁵ In most cases of B-ALL, CD22 is expressed in addition to CD19, so researchers at the NCI are testing CAR T cells that target this protein. The results of a small phase I study of 21 children and young adults (17 previously treated with CD19- directed immunotherapy) were promising. CRs were obtained in 73% of the patients and median remission duration was 6 months.⁶

CAR T-cell therapy was approved by the FDA for treatment in children before it was for adults, something that Fry said is almost unheard of in cancer. In an interview with OncologyLive^®, Ivan M. Borrello, MD, an associate professor of oncology at Johns Hopkins Sidney Kimmel Comprehensive Cancer Center in Baltimore, Maryland, said that tisagenlecleucel may be a better fit for ALL, especially in children. “I think the reason for that is that children’s immune systems are much more resilient,” he said. “They have a functional thymus that pretty much disappears once you get to puberty. What that does, I think, is it supplies a healthier, younger T cell, which in the end may make a difference.”

The approval for axi-cel, which came 2 months after the tisagenlecleucel decision, is for the treatment of adults with relapsed or refractory large B-cell lymphoma after 2 or more lines of systemic therapy. The cancer types covered by the indication include diffuse large B-cell lymphoma (DLBCL) not otherwise specified, primary mediastinal large B-cell lymphoma (PMBCL), high-grade B-cell lymphoma, and DLBCL arising from follicular lymphoma (FL).

In updated results from the pivotal phase II ZUMA-1 trial, investigators reported results for 101 adult patients with refractory DLBCL, PMBCL, or transformed FL who received axi-cel.⁷ The objective response rate (ORR) was 82% and the CR rate was 54%. The median follow-up for this study was 15.4 months, with 40% of patients having a lasting CR. The OS rate was 52% at 18 months. The median duration of response was 8.1 months.

“With existing therapies in this highly refractive patient population, the response rates are less than 30%, and the complete response rates are less than 10%,” said Sattva S. Neelapu, MD, a professor in the Department of Lymphoma/Myeloma at The University of Texas MD Anderson Cancer Center in Houston, in an OncLive interview. “And the median survival for these patients is about 6 months. With axi-cel, we are seeing more than half the patients are still alive 1 year later.”

Neelapu, who was a leading investigator on the ZUMA-1 trial, said investigators tried to identify causes of progression or relapse after axi-cel by examining tumor biopsies of 21 patients. “We found that about one-third of them had loss of the CD19 protein at the time of progression. In another two-thirds of the tumors, we also found an upregulation of PD-L1, which is an inhibitory ligand that can inhibit the function of the CAR T cells. So, what this suggests is that there are at least 2 potential mechanisms of resistance, and this could potentially guide us on future therapies.”

In fact, Neelapu noted, the ZUMA-6 trial is testing the combination of axi-cel and atezolizumab (Tecentriq), which blocks PD-L1, in adults with DLBCL (NCT02926833). In early findings reported at the 2017 American Society of Hematology Annual Meeting, investigators said the combination elicited responses in 8 of 9 patients, including 5 participants with CRs.⁸

According to Brentjens, the response rates to CAR therapies for patients with lymphoma may be lower than for those with other hematologic malignancies, but the responses themselves may be more durable. “Again, this is based on a limited number of studies with a limited number of patients,” he pointed out. “But that seems to be the picture that is coming into focus.”

It has been noted that the CAR trials report responses of the percentage of patients treated, rather than the intent-to-treat population, but Brentjens says that this is a reasonable approach at this point. “You want to have a sense of what a baseline is as far as response rates go,” he said. “I think that even if you did it on an intent-to-treat basis, the numbers would still be quite spectacular, certainly in ALL.”

There have not been any head-to-head studies comparing CAR T-cell therapy with other treatments and this may not be possible, since the number of patients that may be required to conduct a phase III trial may be excessive, as might the cost of these studies. “I think that even early on, we always thought that a phase II trial is where the FDA would be consulted for approval and probably what will happen is that the continued reporting on patient outcomes to the FDA will establish whether or not the FDA will validate that the FDA approved these as second-line treatment and that they may move to first-line [therapy],” Brenjens said.

Toxicity, Disease Burden, and Response

Neelapu anticipates that a clinical trial testing axi-cel versus autologous stem cell transplantation (ASCT) will be conducted in the second-line setting for patients with relapsed or progressive B-cell non-Hodgkin lymphoma, where ASCT is the current standard.Toxicity associated with CAR T-cell therapy is a significant issue. Cytokine release syndrome (CRS) is the most common adverse event (AE) and can be fatal. The rapid and massive release of cytokines into the bloodstream can lead to dangerously high fevers and low blood pressure. Tocilizumab (Actemra), an interleukin-6 receptor antagonist, was approved by the FDA in August 2017 for management of CRS, based on data indicating 69% of patients who received 1 or 2 doses for treatment of severe or life-threatening CRS had resolution of CRS within 2 weeks of receiving the medication.⁹ Tocilizumab is now standard therapy for this patient group.

Borrello pointed out that the toxicity is limiting the ability to move CAR T-cell therapy to upfront settings. “We can’t give this to someone who is newly diagnosed,” he said. Other AEs include neurotoxicities and B-cell aplasia. “CD19-targeted CAR can result in B-cell depletion and hypogammaglobulinemia,” Borrello said. Brentjens agreed, but pointed out that it is relatively anecdotal. “Intravenous immunoglobulin therapy can often help these patients,” he said. “We routinely give intravenous immunoglobulin therapy to our chronic lymphocytic leukemia patients, for example.

It’s a feasible thing to do if there are longterm B-cell aplasias and low antibody counts or levels.” There do not seem to be any long-term issues associated with the therapy, he added. There has also been concern about treatment-induced cerebral edema resulting from CAR T-cell therapy.⁵ Although not all studies have reported this as a significant problem, Juno Therapeutics halted further development of JCAR015, which had a 19-28z construct, after there was a greater than expected incidence of severe neurotoxicity and 5 deaths from cerebral edema during the phase II ROCKET trial in adults with relapsed/refractory ALL.¹⁰ The company has reported promising findings for several other CAR candidates that use a different construct.

Despite the possibility of strong toxic reactions to CAR T-cell therapies, researchers are not sure whether toxicity plays a role in treatment durability. “There is evidence to suggest that some toxicity has been associated with efficacy, but there’s also evidence to suggest that efficacy is independent of toxicity,” Borrello said. “The whole thing is very controversial at this point, but clearly there are some CAR Ts that are showing more toxicity and others that are showing less. Nobody truly understands why at this point.”

That being said, researchers have seen that patients with more disease burden tend to have more toxic responses to treatment. “What we have seen is that if you have more disease on board when you get the T cells, you’re more likely to have some of the toxicities, like the cytokine release syndrome,” Brentjens explained. “Less disease or minimal residual disease patients tend to have very minor toxicities and tend to not require significant ICU care. So, we know that we can kind of predict going in who may or may not have worse toxicity. But whether there is, ‘the more toxicity you have the more likely it stays in remission,’ we haven’t seen a connection there at this time. I wouldn’t be surprised that really the only thing we can connect the 2 with is the amount of the disease at the time of treatment to correlate to toxicity.”

Interestingly, the CAR T-cell therapies are less effective among patients with higher disease burden, too. In one study, patients with minimal residual disease (less than 5% blast cells in bone marrow) fared significantly better than those with morphologic disease (5% or more blast cells in bone marrow).¹¹ “We’ve got a very expensive therapy that, when it works, it works very well, but when it doesn’t work, it doesn’t work well at all,” Borrello said. “It’s not like it goes down from a CR to a PR [partial response] by 5%. It goes down dramatically.”

Hoping to Lower Cost

At this point, other than the disease burden at the start of therapy, there is no way to know ahead of time who will respond to CAR T-cell therapy and who will not. The issue now is whether researchers can find effective biomarkers to develop a system or a set of assays that would predict the likelihood of response.This new therapy comes at a steep price. The debut price of a single dose of tisagenlecleucel is $475,000 and $373,000 for axi-cel; one analyst estimated that the total cost of CAR T-cell therapy could reach as high as $1.5 million per patient.¹² Given this high price tag, some may question the cost in relation to the uncertain response and durability. “However, the fact that we’re getting some really longterm remissions in some of these patients who are heavily pretreated, who have an absolutely dismal prognosis suggests that the therapy begins, at least, to justify the cost,” Brentjens said. “I think that the best [approach] is to try to compare it to the price tag of what a [bone marrow] transplant costs, and then argue that perhaps the success, certainly in ALL, seems to be greater with this procedure than with standard bone marrow transplants.”

Dana Cooper, a spokesperson for Novartis, agrees. In an interview with OncLive^® in August 2017, Cooper said, “The only potentially curative option for these pediatric and young adult patients is allogeneic stem cell transplant, which costs $500,000 to $800,000 for the first year. Additionally, the stem cell transplant economic burden in subsequent years is high. With the potential to extend survival and improve quality of life, [tisagenlecleucel] can deliver significant societal value.”

CAR Therapy as a Game Changer

The therapy’s high price comes from its origins. It is like giving patients a living drug, explained Brentjens. Blood must be drawn from the patient and the T cells separated out.⁵ These T cells are then genetically engineered using a disarmed virus so that they produce the CARs. Once this occurs, the T cells have to be expanded into the hundreds of millions before infusing them back to the patient.When a new treatment for a disease like cancer shows such promise, it is hard not to get too excited over the results. But we must still be cautious, say the researchers. “CAR T-cell therapy is a game changer,” Borrello said. “It’s a major advance in the treatment of at least these liquid tumors, but it’s not a universal cure. It comes at a significant toxicity, and there’s certainly a significant price tag associated with it that needs to be considered.

“Furthermore, I think the other thing is that what is still unclear is what is the benefit of CAR T-cell therapy to patients who don’t achieve a complete remission? I think a lot of this will only come once we have more mature data, and we’re beginning to get this mature data now.”

Brentjens also noted that CAR T-cell therapy is a very novel approach that is not yet in its final iteration. He likes to use a car comparison for the CAR therapy: “We have a Model T Ford and what we really want is a Mustang. Which is to say that the success that we’re having now is very promising, but we continue to try, we continue to acknowledge that we’re not yet at 100% long-term durable remissions. This means there’s further engineering that we could do to these immune cells as we understand the process better, to make them even better. The CAR T cells that you can get now, I would predict, are not the same cells that you’re going to get 10 years from now or maybe even as soon as 5 years from now, as we continue to innovate and make these cells more potent.” The other thing to consider is, will this therapy be a new way to treat other cancers?

“Is this a proof-of-principle for treating more cancers or cancers in general?” Brentjens asked. Researchers such as Steven A. Rosenberg, MD, PhD, a Giants of Cancer Care^® award winner who is chief of the Surgery Branch in the NCI’s Center for Cancer Research, aren’t sure. “Efforts to identify unique antigens on the surface of solid tumors have largely been unsuccessful,” he said in an article published on the NCI website.⁵ In hematologic cancers, the therapy can reach the tumor antigens but most tumor antigens in solid cancers are inside the cell, making them unreachable.

“Can we move this technology to solid tumor malignancies? That’s not an easy question to answer and one that probably won’t be answered in the near future,” Brentjens said.