In 2013 when Drs. Stephen Forman and Jacob Rowe referred to the “myth of the second remission of acute leukemia in the adult,” they were justified in doing so.1 For relapsed B-cell acute lymphoblastic leukemia (B-ALL), the salvage chemotherapy regimens available at the time rarely led to a durable second remission (CR2), and therefore few patients were able to proceed to allogeneic hematopoietic stem cell transplantation (allo-HSCT) for potential curative therapy. The five-year overall survival (OS) after first relapse was estimated to be three to 10 percent, based on data from several large trials.2–4 A more recent analysis of the historical data from that time (1,706 adult patients diagnosed between 1990-2013 from 11 study groups in Europe and the United States) has confirmed that most patients do not achieve a CR2 with standard chemotherapy salvage regimens:5 The overall CR rate following a first salvage chemotherapy regimen is estimated at 40 percent, and the rate of response decreases further with successive salvage chemotherapy regimens (21% following a second salvage chemotherapy attempt, and 11% following third or greater salvage). The one- and three-year OS rates also decrease considerably following successive attempts at treatment: first (26% and 11%), second (18% and 6%), and third or greater salvage (15% and 4%).
But those predictions may no longer apply. Immunotherapy, specifically the targeted antibody therapies blinatumomab and inotuzumab and chimeric antigen receptor T-cell (CAR-T) therapy, for relapsed and refractory (R/R) B-ALL has proven to be very effective for achieving CR2 with significantly improved response rates, including a significant proportion of the remissions with undetectable measurable residual disease (MRD).6–8 Moreover, these therapies are associated with decreased toxicity, thereby allowing patients to maintain sufficient performance status to undergo consolidation allo-HSCT, and thus enabling these agents to serve effectively as a “bridge” to allo-HSCT for potential cure in CR2. Now, these targeted antibody therapies are being tested in the frontline setting in pediatric and adult B-ALL trials as a strategy to eradicate MRD early in the treatment course, and thus overcome intrinsic chemotherapy resistance and prevent relapse (NCT02003222, NCT03150693, NCT03914625, NCT03959085) and as a “substitution strategy” to reduce and/or eliminate cytotoxic chemotherapy for older adults with B-ALL (NCT03739814).
Phase I/IIa studies of CAR-T therapy in children and young adult patients (≤25 years) with R/R B-ALL have demonstrated strong antileukemia activity with CR rates of 81 to 93 percent and durable remissions with 12-month relapse-free survival of approximately 60 percent and OS of 76 percent.9,10 In August 2017, the U.S. Food and Drug Administration (FDA) approved tisagenlecleucel, an anti-CD19 CAR-T product for pediatric and young adult patients up to age 25 years with refractory B-ALL or in second or later relapse.11 The efficacy of CAR-T therapy for adult patients is similar from single institution phase I data showing a CR rate of 62 to 83 percent and a median event-free survival (EFS) of 6.1 months as well as a median OS of 12.9 months.12–15 The first multicenter phase I trial for adult patients, ZUMA-3, showed that the CAR-T product KTE-X19 was safe and established the recommended phase II dose as 1 × 106 cells/kg.16
In the current article, Dr. Bijal Shah and colleagues report on the phase II results of the ZUMA-3 trial (NCT02614066).17 This trial led to the FDA approval of brexucabtagen autoleucel (Tecartus) for adult patients (≥18 years) with R/R B-ALL, based on the high rate of CR following a single infusion of CAR T-cells and a manageable safety profile. I chose this trial and the FDA approval of KTE-X19/brexucabtagen autoleucel as this year's best advancements for ALL as they bolster the probability of achieving CR2 and potentially of cure for adult patients with R/R ALL. Equally as important, the approval will expand the access of CAR-T therapy to many more adult patients and establish CAR-T therapy as a core element of B-ALL treatment.
The ZUMA-3 trial was a single-arm, multicenter, international, phase II trial. Eligible patients had R/R B-ALL with morphological disease in the marrow (>5% blasts) further defined as first relapse with remission of less than 12 months, R/R after at least two prior lines of systemic therapy, or R/R following allo-HSCT. Of note, prior blinatumomab therapy was acceptable as long as the leukemic blasts retained more than 90 percent expression of CD19. The CAR-T product KTE-X19 is the same product approved for the treatment of adults with R/R mantle cell lymphoma and notable for an additional step in the manufacturing process that removes malignant cells to reduce the potential for activation and exhaustion of the anti-CD19 CAR T-cells in the ex vivo manufacturing process. The conditioning chemotherapy was fludarabine at a dose of 25 mg/m2 on days -4, -3, and -2 and cyclophosphamide 900 mg/m2 on day -2. KTE-X19 was administered at a dose of 1 × 106 on day 0. Hospital admission following KTE-X19 infusion was required for a minimum of seven days.
This trial enrolled 71 adult patients (≥18 years) between October 1, 2018, and October 9, 2019, at 25 trial sites making it the largest population of adult patients reported to date. KTE-X19 was manufactured successfully for 65 (92%) patients and administered to 55 (77%). The median age of the 55 treated patients was 40 years, though eight (15%) were 65 years or older. The median time from leukapheresis to KTE-X19 manufacturing release was 13 days for U.S. patients and 14.5 days for European patients. The overall response rate was 71 percent, and the CR rate was 56 percent. The rate of undetectable MRD was 76 percent for all treated patients, and among patients with response, 97 percent had undetectable MRD. The median OS for all treated patients was 18.2 months and was not reached in responding patients.
Cytokine release syndrome (CRS) was observed in 89 percent of patients with grade 3 to 4 CRS observed in 24 percent. Neurologic toxicity was observed in 60 percent of patients with grade 3 to 4 observed in 25 percent. There was one grade 5 neurologic event (brain herniation on day 8) related to KTE-X19. Tocilizumab was administered to most patients (80 percent) along with glucocorticoids (75 percent) for the treatment of CRS and neurologic events. Infections of grade 3 or higher occurred in 25 percent of patients, and one grade 5 event (septic shock on day 18) related to KTE-X19. Importantly, no new safety signals have emerged among patients treated in the phase I portion of this trial.
Less than 10 years ago, the likelihood of a successful CR2 in B-ALL was low with the available salvage chemotherapy. The approval of CD19-directed CAR-T therapy for adult patients with R/R B-ALL represents a clinical milestone for immunotherapy in 2021. The high response rates in this heavily pretreated patient population are striking, and the long-term remissions for this cohort are promising. This approval also presents new questions, particularly about the optimal timing of the CAR-T therapy in R/R B-ALL. With its availability, how should we sequence the different immunotherapies of blinatumomab, inotuzumab ozogamicin, or CAR-T therapy? Because of the differences in the eligibility criteria and patient cohorts (regarding the number of prior salvage attempts and percentage of patients with prior allo-HSCT) enrolled on the clinical trials that led to the FDA approvals for these immunotherapies, there is not enough evidence that the efficacy of any one agent is superior to another for an individual patient. Currently, most clinicians consider efficacy along with toxicity, disease, and patient characteristics to determine the best choice of immunotherapy for a patient with R/R B-ALL.
A second important question in the field is whether CAR-T therapy should be incorporated into frontline therapy for high-risk B-ALL? For example, for patients with detectable MRD following induction therapy, should CAR-T therapy be used before developing overt relapse? Though most CD19-specific CAR-T trials have been evaluated in patients with overt relapse, the Memorial Sloan Kettering Cancer Center has reported on a subgroup of patients with low disease burden (<5% marrow blasts) before CAR-T infusion.13 Patients with low disease burden had longer EFS and OS (median EFS, 10.6 months; median OS, 20.1 months) compared to patients with high disease burden (median EFS, 5.3 months; median OS, 12.4 months). Also, the incidence of severe toxicity was significantly reduced in patients with low disease burden (CRS 5% and neurotoxicity 14%) compared to patients with high disease burden (CRS 41% and neurotoxicity 59%). The data presented in the ZUMA-3 trial revealed that CAR-T expansion was inversely associated with bone marrow blasts at screening; patients with the highest quartile of blasts in the marrow (75%-100%) were less likely to have CAR-T expansion, and thus less likely to have a response. CAR-T expansion was highest in patients who achieved a CR or CR with incomplete count recovery. For patients who achieved an MRD-negative remission, the median CAR-T levels were more than 60 times higher in patients with MRD-negative versus MRD-positive status after infusion. This data suggests that CAR-T therapy may be most beneficial before overt relapse. The Children's Oncology Group is testing this idea currently with tisagenlecleucel in a phase II study (Cassiopeia trial, NCT03876769) for patients ages 1 to 25 years who have MRD detectable (>0.01%) at the end of consolidation therapy. Notably, tisagenlecleucel will be used as definitive therapy in this trial with no planned consolidative allo-transplant.
A third important question is what is the role of consolidation allogeneic transplant? Can CAR-Ts replace transplantation in first CR? Ten patients (18 percent) underwent allo-HSCT after KTE-X19 infusion. The median duration of remission both with and without censoring patients at the time of allo-HSCT was 12.8 months. Whether cure may be achieved for patients with relapsed B-ALL disease without transplantation is unknown in adults, but certainly CAR-T therapy has demonstrated durable remissions without subsequent allo-HSCT in pediatric patients.18 In adults, we know that CAR-T therapy is an effective bridge to transplant, yet it also provides hope for those who are transplantation ineligible. In this trial, patients aged 65 years and older were treated with manageable toxicity; this result is significant as older patients with B-ALL have a specific need for less toxic, targeted therapy, and one that will provide opportunity for long-term survival without subsequent HSCT. It must be remembered that 20 to 30 percent of ALL cases occur in adults aged 55 years and older.19 Historically, older adults with ALL have dismal outcomes with long-term cure rates of less than 20 percent.20
In the next year, we anticipate that many more adult patients with R/R B-ALL will be treated with brexucabtagene autoleucel, thus offering many more patients with relapsed disease an opportunity for potential curative therapy, possibly without allo-HSCT. Drs. Forman and Rowe in their 2013 commentary spoke of the promise of new immune therapies to achieve CR2, “…bringing the probability of cure for patients with relapsed acute leukemia from myth to reality.” With the approval of CAR-T therapy for adults with R/R ALL, we have moved beyond the myth, and the hope for better therapy they spoke of has been realized.
Dr. O’Dwyer was a co-author on Shah et al.16