New indications may accelerate ‘explosion’ in CAR T-cell therapy, but more education still needed
Years of research into chimeric antigen receptor T-cell therapy for cancer treatment is finally paying dividends.
When Renier J. Brentjens, MD, PhD, director of cellular therapeutics at Memorial Sloan Kettering Cancer Center, first presented a study on the potential of CAR T-cell therapy during a poster session at ASH Annual Meeting and Exposition 18 years ago, no one approached his poster to ask him questions about the results.
“We toiled in relative obscurity in which only a small subset of investigators knew about this, [then] there was a national explosion,” Brentjens told HemOnc Today. “But it didn’t go from zero to 60 in just one event. Enthusiasm [in academia] has been building for quite some time.”
Two decades later, CAR T-cell therapy has generated tremendous excitement in the hematologic community and has the potential to fundamentally shift cancer treatment across the world.
That enthusiasm has been furthered by two FDA approvals of CD19-directed CAR T-cell therapies.
Tisagenlecleucel (Kymriah, Novartis) is used to treat certain children or young adults with large B-cell lymphoma or acute lymphoblastic leukemia. Axicabtagene ciloleucel (Yescarta; Kite Pharma, Gilead) is used to treat certain adults with relapsed or refractory large B-cell lymphoma.
Numerous studies are ongoing to determine the efficacy of CAR T-cell therapy in other types of hematologic malignancies, as well as whether use of the therapy can be expanded to solid tumors. Some centers also are developing their own point-of-care production systems to avoid the delays of off-site manufacturing and increase treatment accessibility.
However, as the use of CAR T-cell therapy expands, so does the need for clinician education about potential side effects and patient eligibility.
“This technology would get a big boost if we can get similar results in solid tumors with point-of-care-modified CAR T cells, but that has yet to be, and may never be, demonstrated,” Brentjens said. “We may get there eventually, but it may be a little longer before the next home run in this field is appreciated.”
HemOnc Today spoke with experts in cellular therapy about newly reported data on the long-term durability of CAR T-cell therapy and combinations with checkpoint inhibitors, the potential for point-of-care therapy to dramatically change treatment accessibility, and how “armored” CAR T cells might expand their use into the solid tumor setting.
Durability of outcomes
Studies presented in December at ASH Annual Meeting and Exposition further confirmed the efficacy of CAR T-cell therapy in certain patients with leukemia and lymphoma.
An updated analysis of the phase 2 ELIANA trial showed the persistence of benefit of a single infusion of tisagenlecleucel among children and young adults with relapsed or refractory B-cell ALL.
“I am pleasantly surprised about the durability,” Stephan A. Grupp, MD, PhD, lead trial investigator and director of the Cancer Immunotherapy Program at Children’s Hospital of Philadelphia, told HemOnc Today. “There are very few events happening after a year. For patients out 1 year or 2 ... the likelihood that they are going to stay in remission is very good.”
The global, 25-center ELIANA trial included 97 patients (median age, 11 years; range 3-24) with CD19-positive relapsed or refractory B-cell ALL. Seventy-five patients received a single tisagenlecleucel infusion (median dose, 3 x 106 CAR-positive viable T cells/kg; range, 0.2-5.4 x 106) following lymphodepleting chemotherapy.
Median time from infusion to data cutoff was 24 months (range, 4.5-35), which represented an additional 11 months of follow-up from previous reports.
Overall remission rate — which included complete remission plus complete remission with incomplete blood count recovery — within 3 months and maintained for 28 or more days served as the study’s primary endpoint.
The study met this endpoint, with an overall remission rate of 82% (95% CI, 72-90).
Median duration of remission was not reached, but responses were ongoing in 29 patients at 29 months.
“We already knew the short-term efficacy was really good and we were excited about that, but if you’re going to do a complex therapy like this ... the best outcome is a one and done,” Grupp said. “We now have really good follow-up data and are starting to see what it looks like in that 1-, 2- or 3-year time frame in terms of how patients are holding up.”
The probability of RFS was 66% (95% CI, 52-77) at 18 months and 62% (95% CI, 47-75) at 24 months.
Median OS was not reached. OS probability at 18 months was 70% (95% CI, 58-79).
Grade 3 or grade 4 cytokine release syndrome (CRS) occurred in 77% of the patients. Of those, 39% received tocilizumab (Actemra, Genentech) with or without other anticytokine therapies. All cases were reversible.
Tisagenlecleucel also induced durable responses among heavily pretreated adults with relapsed or refractory diffuse large B-cell lymphoma, according to 19-month follow-up results from the phase 2 JULIET trial presented at ASH.
The trial included 167 adults (median age, 56 years) with relapsed or refractory DLBCL who received at least two lines of prior therapy — including rituximab (Rituxan; Genentech, Biogen) and an anthracycline — and either failed or were ineligible for autologous stem cell transplant.
One hundred fifty-five patients, 77% of whom had stage III or stage IV disease, received one dose of tisagenlecleucel. Median dose was 3 x 108 CAR-positive viable T cells/kg (range, 0.1 x 108 to 6 x 108).
The efficacy analysis included 99 patients who had at least 3 months of follow-up.
Results showed an overall response rate of 54% (95% CI, 43-64); 40% of patients achieved complete response and 13% achieved partial response. Median duration of response was not reached, although investigators estimated a 64% (95% CI, 48-76) 12- and 18-month RFS rate. Investigators observed no relapses beyond 11 months.
Median OS was 11.1 months (95% CI, 6.6-not estimable) for all infused patients but was not reached (95% CI, 21-not estimable) for patients in complete response. Investigators calculated OS probabilities of 48% (95% CI, 38-57) at 12 months and 43% (95% CI, 33-53) at 18 months.
Within 8 weeks of infusion, grade 3 or grade 4 adverse events of special interest included cytopenia lasting more than 28 days (34%); CRS (23%); infections (19%); febrile neutropenia (15%); neurologic events, including one case of grade 2 cerebral edema (11%); and tumor lysis syndrome (2%).
Investigators reported no treatment-related mortality.
“When you have these aggressive lymphomas, most of the recurrences occur in the first 12 months. When you get to 19 months, you start to really believe that 5 years is possible,” Richard Thomas Maziarz, MD, study researcher and professor of medicine in the division of hematology and medical oncology at Knight Cancer Institute at Oregon Health & Science University, told HemOnc Today at the time of the presentation. “No one is at the point of saying ‘cure’ with CAR T, but when you start to see the remission-free survival paralleling survival in general, we have to start to think that it may become a potential outcome for these patients.”
Still, not all patients treated with CAR T cells respond to the therapy or have a durable response.
Also, despite initial complete response rates of 80% to 90% among patients with B-cell ALL, RFS declines to 60% within the first year. CD19-positive relapses usually occur due to CAR T-cell loss.
To increase the proportion of patients who respond to therapy and extend the life of CAR T cells, researchers began studying the addition of checkpoint inhibitors to CAR T. The PD-1 checkpoint pathway may be involved in CAR T-cell exhaustion, and addition of checkpoint inhibitors may overcome this effect and help increase CAR T function and persistence.
“For a fraction of patients who do not have durable remissions with CAR T cells, we can potentially improve upon the function and persistence of their CAR T cells by adding checkpoints,” Shannon Maude, MD, PhD, attending physician in the Cancer Center of Children’s Hospital of Philadelphia, told HemOnc Today at ASH.
Maude and colleagues evaluated data from 14 patients (age range, 4-17 years) with pretreated, relapsed B-ALL (n = 13) or B-cell lymphoblastic lymphoma (n = 1) who demonstrated repeated early CAR T-cell loss or partial/no response to CAR T-cell therapy.
Patients received CD19-directed CAR T-cell therapy (CTL019, n = 4; CTL119, n = 10) in combination with pembrolizumab (Keytruda, Merck; n = 13) or nivolumab (Opdivo, Bristol-Myers Squibb; n = 1). Researchers administered the checkpoint inhibitor at least 14 days after CAR T-cell infusion.
Three patients treated with CD19 CAR T cells in combination with a checkpoint inhibitor re-established B-cell aplasia — an indicator of CAR T-cell function — for 5 to 15 months. Two of those patients had persistent B-cell aplasia while taking pembrolizumab.
Researchers observed two complete and two partial responses among four patients with bulky extramedullary disease unresponsive to or who relapsed after CAR T-cell therapy.
One patient experienced CAR T-cell proliferation days after beginning checkpoint inhibitor treatment.
Partial responses occurred with the addition of pembrolizumab among four patients who did not go into remission after CAR T-cell infusion. One patient progressed with CD19-negative/dim disease.
“This is still a small study, so we need to look at this more broadly and identify biologic features that will tell us the mechanism of why patients are responding and why they’re not, and if there are predictors in those who would most benefit,” Maude said.
CRS occurred among three patients within 2 days of starting checkpoint inhibitor therapy. Other adverse effects, including acute pancreatitis, hypothyroidism, arthralgias, urticaria and cytopenias, were tolerable or reversible after discontinuation.
Two patients who stopped checkpoint inhibitor therapy due to adverse events experienced relapse or disease progression within weeks.
“Checkpoint inhibitors in combination with CAR T cells make perfect sense,” Brentjens said. “The fact that the two of them work together is not surprising and it’s encouraging. This is just one example; there are multiple antibodies that are being investigated that may have utility to complement or rescue the function of the CAR T cell.”
Huge fanfare, little knowledge
These trials and approvals have led to a rapid rise of CAR T-cell therapy; however, that speed has left little time for medical professionals to remain up to speed themselves.
A study at ASH showed clinicians in the hematologic community have little knowledge of the therapy itself, including on CAR components, dosing and FDA-approved indications.
Based on a CME-certified clinical practice survey consisting of multiple-choice questions about the therapy, Willis and colleagues found that 61% of hematologists/oncologists could not correctly identify the components of a CAR construct, 45% did not recognize that currently approved CAR T-cell therapies are dosed as a single infusion and 62% could not identify the FDA-approved indication for axicabtagene ciloleucel.
“The results aren’t surprising because the vast majority of hematologists don’t directly care for patients who are receiving CAR T cells,” Edward A. Copelan, MD, chair of the department of hematologic oncology at Levine Cancer Institute at Atrium Health and a HemOnc Today Editorial Board Member, told HemOnc Today. “As the indications for this therapy grow, use will become more widespread, and hematologists will become more knowledgeable in its application.”
Brentjens said that considering the therapy is still relatively new and that most hematologists don’t directly deal with patients undergoing this treatment, he is pleased to see the numbers as high as they were.
“Most CAR T-cell studies are done at academic centers, and many of those clinicians probably didn’t answer the questionnaire,” Brentjens said. “Right now, the indications are in leukemia and lymphoma, and perhaps multiple myeloma soon. Those are, relatively speaking, rare diseases compared with breast cancer and lung cancer. I’m pleased to see that there are 40% who do know the components. That’s pretty impressive for a therapy that is not yet mainstream.”
Study results also showed that 44% of hematologists/oncologists could not identify signs of CRS associated with CAR T-cell therapy, 43% lacked knowledge that elevated serum C-reactive protein is associated with the highest level of CRS in patients with lymphoma receiving axicabtagene ciloleucel, 35% could not identify signs of neurotoxicity associated with CAR T-cell therapy, and 54% could not identify the appropriate role of corticosteroid therapy in managing CRS and neurotoxicity.
Increased understanding of the adverse-event profile of CAR T-cell therapy should improve these statistics, Copelan said.
“There is increasing understanding of the mechanisms underlying CRS and neurotoxicity,” he said. “There’s great emphasis in trying to identify patients in whom these complications will develop.”
Cost efficacy questioned
The promise associated with CAR T-cell therapy comes with a hefty price tag.
For their advanced lymphoma indications, both tisagenlecleucel and axicabtagene ciloleucel are priced at $373,000. Tisagenlecleucel carries a price of $475,000 for its pediatric leukemia indication. Neither of these price points include the cost of hospitalization and treatment for toxicities that may occur.
“As expensive as it is, it can’t just give a statistically significant enhanced DFS,” Brentjens said. “It has to be highly effective.”
In February, CMS — which currently has no national Medicare policy for CAR T-cell therapy, leaving the decision to local Medicare contractors — announced it would cover the therapy under “Coverage with Evidence Development.”
This would require Medicare to cover the therapy when it is offered in a CMS-approved registry or clinical study that monitors patients for at least 2 years following treatment. These data would then help CMS identify patients most likely to benefit from treatment who could be covered in the future without a data requirement.
Cost-effectiveness needs to be considered in terms of other treatment options for these patients, Brentjens said.
“I do respect the fact that the lump sum cost of these cells as it stands right now is not palatable for insurance companies,” he said. “But with an expensive one-and-done approach instead of many rounds of chemotherapy or other treatment, the overall cost could end up being lower for CAR T.”
At ASH, Baumgardner and colleagues presented an analysis showing CAR T-cell therapies break with past trends of low-value innovation marked by increases in pharmaceutical oncology treatments that have outpaced survival improvements.
Researchers identified all analyses of pharmaceutical treatments for cancer published since 2007 in the Tufts Medical Center Cost-Effectiveness Analysis Registry, and they derived cost-utility analysis data for CAR T-cell therapies from a study conducted last year by the Institute for Clinical and Economic Review.
A regression analysis showed that, for anticancer interventions overall, incremental quality-adjusted life years (QALYs) gained have declined by 0.079 (95% CI, –0.128 to –0.03) per year. Cost-effectiveness for treatments also worsened, with an increase of $36,147 (95% CI, –29,575 to 101,868) per year.
However, results showed that CAR T-cell therapy provides 5.17 (95% CI, 4.09-6.25) more incremental QALYs than pharmaceutical cancer innovations outside of hematology and 4.67 (95% CI, 3.44-5.9) more incremental QALYs than other treatments within hematology.
Treatments within hematology other than CAR T-cell therapy provided 0.5 (95% CI, –0.09 to 1.09) more incremental QALYs than new treatments outside of hematology.
However, at ASH, experts suggested biases may limit the findings of this study. Also, these costs may not include the “hidden” costs of CAR T cells that are incurred from toxicity and patient management.
“There are articles questioning the cost-effectiveness and the overall effectiveness in getting sustained remissions,” Copelan said. “CAR T cells are clearly effective, and they have been shown to be cost-effective as well. More knowledge will emphasize their effectiveness.”
As with any product, prices likely will decrease with time, Brentjens said.
“Industry takes responsibility by improving the cost of production, streamlining the process of making them, and potentially getting off-the-shelf cells,” he said. “All that will very likely decrease the cost. But, if the cells are very good, even at a higher price for the therapy, ultimately health care savings may be seen because patients would not need additional treatments.”
Point-of-care CAR T
Limitations to the expanded use of CAR T-cell therapy include its centralized production and off-site manufacturing, both of which delay the time to treatment and drive up costs.
Zhu and colleagues at Medical College of Wisconsin, however, recently demonstrated the feasibility of point-of-care CAR T-cell production for clinical use in a phase 1 trial presented at ASH.
Researchers collected peripheral blood mononuclear cells by apheresis from six patients — three with mantle cell lymphoma, two with DLBCL and one with chronic lymphocytic leukemia — and loaded them into a CliniMACS Prodigy (Miltenyi Biotec) device.
Researchers then used positive immunomagnetic selection to enrich CD4 and CD8 T cells. The culture process involved keeping the cells in a TecMACS medium while being supplemented with 3% human AB serum and 200 U/mL IL-2, then adding TransACT reagent to stimulate the T cells. The next day, researchers added lentiviral vector expressing anti-CD19 and anti-CD20 with CD3-zeta and 4-1BB stimulatory domains.
Culture washes and feeding were conducted on days 5, 6, 8, 10 and 12, and the final harvesting occurred on day 14.
Researchers gathered an average yield of 5.9e+8 (range, 4.3-7.9e+8) CAR T cells. Enriched T cells were 94.3% CD3-positive and average CD20.19 CAR expression was 20.8%.
Three patients received a cryopreserved product, and the other three received a fresh product.
Results showed that patient CAR T cells killed CD19- and CD20-positive target cells in vitro and produced interferon-gamma.
The time between taking the cells from the patients and reinserting them was 14 days. However, researchers believe they can reduce that to 8 to 10 days.
“What we’ve learned is that the majority of expansion of the CAR T cells occurs within the first 8 days,” Nirav N. Shah, MD, study researcher and assistant professor at Medical College of Wisconsin, told HemOnc Today. “We are exploring what the optimal number of days is for manufacturing for CAR T cells. The data we have support the idea that we don’t need 14 days for all patients.”
Based on these data demonstrating feasibility, point-of-care CAR T cells could someday greatly expand use of the therapy and drive down costs, experts say.
However, there are still significant obstacles — in terms of efficacy and safety — to overcome before the delivery of this therapy in a point-of-care setting becomes mainstream.
“Studies haven’t yet demonstrated the effectiveness of these point-of-care CARs compared with centrally processed CARs,” Copelan said. “As CAR production becomes more complex, it might complicate the ability to use the current systems that are available locally to reproduce this. Is gene editing going to be transferrable to this point-of-care model? I don’t know the answer to that.”
A decentralized CAR manufacturing system may potentially bring down costs, but it is unlikely to be used across cancer centers and hospitals for years to come, Shah said.
“The cost of therapy is complex and has a variety of factors associated with it,” he said. “The currently approved FDA product is a centralized model in which you ship the T cells to a third-party facility for manufacturing. There are costs obviously associated with that.
“We can save money because everything is done in-house,” he added. “We are manufacturing the CAR T cells in our center and don’t have to ship them out.”
FDA rules regarding the point-of-care setting are still lacking.
“This will move forward based on what the FDA says will be acceptable and what won’t be acceptable,” Brentjens said. “The technology is a simplified way to do this, and it is possible that this could be a better alternative than the larger facilities. But, I don’t necessarily see community hospitals investing in this technology in the foreseeable future.”
On solid ground
Much research is focused on expanding CAR T-cell therapy to other hematologic malignancies and solid tumors.
Pharmaceutical companies have demonstrated a growing interest in CAR T-cell therapy, which will help fund future clinical trials. In March 2018, Celgene acquired Juno Therapeutics for $9 billion, in large part because of its research into CAR T-cell therapy.
“The fact that there’s so much interest on behalf of these companies indicates that there are a lot of people betting on CARs,” Copelan said. “They are not betting on applications in B-cell malignancies alone, they are betting that their use will also spread to solid tumors, acute myeloid leukemia and other diseases.”
Studies have shown efficacy in treating multiple myeloma, the next predicted indication for FDA approval.
Expanding the therapy to solid tumors, however, has proved an extremely difficult task.
“One of the biggest differences between using CAR T-cell therapy for hematological malignancies vs. solid tumor malignancies is the need to overcome the hostile solid tumor microenvironment,” Roisin E. O’Cearbhaill, MD, research director of gynecologic medical oncology at Memorial Sloan Kettering Cancer Center, told HemOnc Today. “Various inhibitory factors in the tumor microenvironment have to be overcome, including T regulatory cells, myeloid-derived suppressor cells and inhibitory cytokines.”
To overcome these challenges, researchers have developed “armored” CAR T cells that are designed to secrete antibodies that block the PD-1/PD-L1 pathways or to secrete proinflammatory cytokines, such as IL-12.
“Armoring the CAR T cells with the ability to secrete IL-12 into the tumor milieu can work in an autocrine mode on the CAR T cells, as well as in a paracrine fashion to recruit and activate the endogenous tumor-infiltrating lymphocytes,” O’Cearbhaill said. “In addition to the direct cytotoxic effect of the CAR T cells, we also hope to reactivate the tumor-infiltrating lymphocytes from an anergic state in order to kill the cancer cells through this other mechanism.”
Brentjens’ laboratory developed two armored CARs, including one that recognized CD19 and the other MUC16, which is found in some ovarian and pancreatic cancers.
“About 70% to 80% of serous ovarian cancers express this antigen,” O’Cearbhaill said. “MUC16(ecto) is expressed only at low levels on normal tissue including the uterus, fallopian tubes, ovaries and corneal surface of the eye and is, therefore, a highly attractive target for CAR-based adoptive T-cell therapy.”
Mouse models showed the armored CARs persisted longer and improved outcomes compared with first- and second-generation CARs.
Whether the challenges of expanding CAR T to solid tumors will be overcome in the next few years remains to be seen, Grupp said.
“There are a lot of ways to try and make this work, and the good news is that there is a lot of interest and investment in this therapeutic area,” he said. “Someone is going to pay for that study, we’re going to learn what we need to learn, and hopefully it will lead to real positive progress in solid tumors.”
The “amazing results” seen in leukemia give Grupp hope that it can expand to other cancers.
“There is always a risk and a hype phase with new technologies — and we are in the hype phase for CAR Ts — but there is a lot of reality behind this,” he said. “All this money flying around means that people can invest in clinical trials, and it means somebody, somewhere will try to make this thing work. All it takes is one or two more significant impacts on cancer diseases and we will see this explosion get even bigger.”– by John DeRosier
Click here to read the , “Will CAR T cells become first-line therapy for hematologic malignancies?”
Rafiq S, et al. Nat Biotechnol. 2018;doi:10.1038/nbt.4195.
The following were presented at ASH Annual Meeting and Exposition; Dec. 1-4, 2018; San Diego:
Baumgardner J, et al. Abstract 322.
Grupp S, et al. Abstract 895. Li A, et al. Abstract 556.
Schuster SJ, et al. Abstract 1684.
Willis L, et al. Abstract 2269.
Zhu F, et al. Abstract 4553.
For more information:
Renier J. Brentjens, MD, PhD, can be reached at Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065; email: email@example.com.
Stephan A. Grupp, MD, PhD, can be reached at Children’s Hospital of Philadelphia, 3401 Civic Center Blvd., Philadelphia, PA 19104; email: firstname.lastname@example.org.
Edward A. Copelan, MD, can be reached at Levine Cancer Institute at Atrium Health, 1021 Morehead Medical Drive, Charlotte, NC 28204; email: email@example.com.
Roisin E. O’Cearbhaill, MD, can be reached at Memorial Sloan Kettering Cancer Center, 1275 York Ave., New York, NY 10065; email: firstname.lastname@example.org.
Nirav N. Shah, MD, can be reached at Cancer Center - Froedtert Hospital, 9200 West Wisconsin Ave., Milwaukee, WI 53226; email: email@example.com.
Disclosures: Brentjens reports being the co-scientific founder of Juno Therapeutics. Grupp reports a consultant role with Novartis. Copelan, O’Cearbhaill and Shah report no relevant financial disclosures.