CD19 CAR-T in relapsed t(8;21) AML: a single-center prospective phase II clinical trial

Abstract

Approximately 78.3% of patients with t(8;21) acute myeloid leukemia (AML) express CD19, making it a potential target for chimeric antigen receptor (CAR)-T cell therapy focused on CD19. This prospective phase II trial (NCT03896854) evaluated the safety and efficacy of CD19 CAR-T cell treatment in 10 relapsed CD19-positive t(8;21) AML patients. This study enrolled eight patients with hematologic and two with molecular relapsed AML. The median bone marrow blast percentage was 12.4% (0.1–50.2%), and the blasts exhibited a median CD19 positivity of 55.7% (22.6–97.1%). Genetic profiling revealed TP53 alterations (n = 1), KIT (n = 3) and FLT3-ITD (n = 1) mutations. After lymphodepletion with fludarabine and cyclophosphamide (FC), 5–20 × 10⁶ cells per kilogram of CAR-T cells were administered. All patients experienced grade 3 or higher hematologic toxicities following tumor-reduction chemotherapy and the FC regimen, which were managed for a median of two weeks after CAR-T treatment. Non-hematological toxicities were mild and reversible. Eight patients presented with mild (grade 1–2) cytokine release syndrome (CRS), and one experienced grade 3 CRS. The immune effector cell-associated neurotoxicity syndrome was not observed. All patients achieved complete remission (CR) after CAR-T, with 60% achieving a molecularly MRD-negative CR. RUNX1::RUNX1T1 fusion transcript levels demonstrated a median 2.5-log reduction (range: 0.7–4.5 log; P = 0.002). At a median follow-up of 64.6 months (range: 11.2–88.8 months), the median overall survival and leukemia-free survival were 11.6 and 3.8 months, respectively. The 12-month cumulative incidence of relapse was 53.3%. These findings indicated that CD19 CAR-T was a safe and effective option for relapsed CD19-positive t(8;21) AML.

To the editor

Despite the initial response to standard chemotherapy in most t(8;21) acute myeloid leukemia (AML) patients, approximately 35% experience relapse or persistent measurable residual disease (MRD) positivity, conferring a dismal 5-year overall survival rate of ≤ 20% in this setting [1,2,3]. This poor prognosis stems from the limited therapeutic options available for this molecularly defined subgroup.

Notably, CD19 expression is observed in 78.3% of t(8;21) AML cases [4], representing a potential therapeutic target for CD19-directed chimeric antigen receptor (CAR)-T cell treatment. We first demonstrated the efficacy of CD19 CAR-T cells in two relapsed t(8;21) AML patients, and this finding was subsequently revalidated by Israeli investigators [5, 6]. Based on these early proof-of-concept results, we conducted this prospective trial to evaluate CD19 CAR-T treatment in CD19 + relapsed t(8;21) AML.

This single-center, Phase II trial (NCT03896854, ClinicalTrials.gov) enrolled 10 relapsed t(8;21) AML patients between October 18, 2017, and December 11, 2023, at the First Affiliated Hospital of Soochow University, China. The study protocol received institutional review board approval in accordance with the Declaration of Helsinki guidelines. Baseline characteristics, including prior therapies, cytogenetic/molecular profiles, pre-treatment, blast percentages, CAR-T cell source, CAR-T cell dose, response rates and survival, are tabulated (Table 1). The methodological components, including subject selection criteria, study protocol, CAR construct design, CAR-T cell production procedures, statistical approaches, and endpoint evaluations, were thoroughly documented in the supplementary materials. The study cohort comprised ten patients (eight males and two females). The median age was 31 years (range: 13–52 years). The median follow-up duration was 64.6 months (range: 11.2–88.8 months).

Table 1 Patient baseline characteristics and treatment outcomes

All patients experienced grade 3 or higher hematologic toxicities following tumor-reduction chemotherapy and lymphodepletion regimen with fludarabine and cyclophosphamide (FC), including anemia in 8 patients (80%), thrombocytopenia in 8 patients (80%), and neutropenia in 9 patients (90%). The median time of red blood cell transfusions (hemoglobin < 70g/L) was 11 days (range: 8–26 days), and for platelet transfusion independence (platelet counts ≥ 20 × 10⁹/L) was 14 days (range: 0–30 days). The median time to neutrophil recovery (ANC ≥ 0.5 × 10⁹/L) was 12 days (range: 1–24 days). Ultimately, the median time for recovery from all hematologic toxicities was 14 days (range: 1–30 days).

The non-hematological toxicities were transient and manageable, including pulmonary infection (grade 2; n = 1), transient hepatotoxicity (grade 1; n = 1) and hypotension (grade 3; n = 1). Cytokine release syndrome (CRS) occurred in 9 patients (90%), with grades 1 in 4, 2 in 4, and 3 in 1. The grade 3 event was managed with steroid intervention (methylprednisolone 1 mg/kg/day intravenous for 3 days) and supportive care. No immune effector cell-associated neurotoxicity syndrome (ICANS) (ASTCT criteria) [7] was observed.

Ten out of 10 patients achieved complete remission (CR) (100% CR rate) post-CAR-T treatment, with 60% attaining molecular MRD-negative CR (CR_MRD-) (Fig. 1a, b, Table S2). RUNX1::RUNX1T1 fusion transcript levels showed a median 3.27-log reduction (range: 0.7–4.5 log; P = 0.002) post-CAR-T (Fig. 1c).

Fig. 1

Treatment Response and Survival of Patients. a Complete remission (CR) status of patients post-CAR-T infusion. Patients 1, 3, 4, 5 and 8 achieved molecular remission. Patients 2, 3, 6, 7 and 8 relapsed within four months, and Patient 1 experienced relapse at 12.2 months after the CD19 CAR-T treatment. b Schedule of the CD19 CAR-T treatment regimen. c Variation in RUNX1::RUNX1T1 fusion gene expression in the bone marrow (BM) measured by quantitative real-time PCR (RT-PCR). d CAR-T cell copy numbers were detected using vector copy numbers (VCN) analysis. e The IL-6 cytokine results during CAR-T cell treatment. f & h 12-month overall survival (OS) and leukemia-free survival (LFS) rates were 45.0% and 46.7%, respectively. h The cumulative incidence of all patients is displayed

The median time to peak levels of CAR-T cells in blood was 8 days after infusion (range: 4–24 days) (Fig. 1d). Longitudinal cytokine profiles (e.g., IL-6) are shown in Fig. 1e and Figure S3. The median peak of IL-6 levels occurred on day 5 (range: 1–10). Figure S1 demonstrated CD19 dynamics and MRD clearance in a representative case (patient 4).

After CAR-T treatment, three patients (patients 3, 4, and 10) underwent allogeneic hematopoietic stem cell transplantation (allo-HSCT), 2 received decitabine-based regimens, and 5 declined further consolidation treatment. Among the three allo-HSCT patients, the patient 10 died of grade IV acute intestinal graft-versus-host disease (GVHD) at 1.5 months post-transplantation, the patient 3 with TP53 alterations succumbed to relapse at 2 months post-transplantation, and the patient 4 receiving decitabine maintenance remained alive in remission. Six patients who underwent allo-HSCT (patients 1, 2, 3, 6, 7, and 8) experienced relapse at a median time of 3.4 months (ranging: 2.1–12.2 months) post-CAR-T therapy. This included five cases of CD19-negative relapse and one case of CD19-positive relapse (patient 8). Several salvage treatments were administered to patients who relapsed after CAR-T, including venetoclax with hypomethylating agent (patients 1, 3), decitabine plus HAAG (Homoharringtonine, Cytarabine, Aclarubicin, combined with Granulocyte colony-stimulating factor) (patient 3), salvage transplantation (patient 6), decitabine combination with CD38 CAR-T (patient 3), and palliative care (patients 2, 7, 8). Finally, seven patients died, and three patients (patients 4, 5, 9) survived after CAR-T treatment. (Figure S2).

The median OS and leukemia-free survival (LFS) were 11.6 months (95% CI, 18.4% to 75.3%) and 3.8 months (95% CI, 18.4% to 75.3%). The 12-month cumulative incidence of relapse (CIR) was 53.3% (95% CI, 20.9% to 77.7%) (Fig. 1f-h). By the last follow-up, three patients were alive (Table S2).

Our study was the first to report the effectiveness of CD19 CAR-T therapy in t(8;21) AML, expanding the research with a larger group, longer follow-up, and a focus on post-CAR-T consolidation. Although all patients achieved CR, the median LFS was only 3.8 months, highlighting the necessity for larger sample sizes and the need for multi-center, collaborative trials to refine treatment strategies. While sequential allo-HSCT or maintenance treatment may help consolidate CAR-T-induced remission, future trials should focus on larger cohorts to evaluate these strategies.