PDF
Citation
Print
Introduction
B-cell acute lymphoblastic leukemia (B-ALL) is one of the most common malignant disease in childhood, with a 5-year overall survival (OS) rate of up to 80%. However, a few patients still experience refractory/relapsed (R/R) disease [1,2], which is associated with a poor outcome and an OS lower than 40% [3]. Patients with persistent minimal residual disease (MRD) after induction chemotherapy or those with MRD changing from negative to positive (MRD re-emergence) during treatment are at a higher risk of relapse and have a poor prognosis [3,4].
The advent of chimeric antigen receptor (CAR) T cell therapy has been a significant breakthrough for these highrisk patients. Recent large clinical trials of CD19 CAR T-cell therapy have provided evidence of excellent efficacy for R/R B-ALL patients with complete remission (CR) rate as high as 80.9% to 93% [5-9]. However, relapse remains common over time, occurring in approximately 40% of patients [5,7,8]. Clinical characteristics, different disease statuses, MRD levels and the kinetics of CAR T-cell expansion and persistence that may affect or predict treatment outcomes need to be analyzed in order to identify patients at higher risk of recurrence and determine strategies for consolidation. Additionally, while previous clinical trials have mainly focused on treating R/R B-ALL, the potential of CD19 CAR T-cell therapy in reducing the risk of relapse or obviating the need for hematopoietic stem cell transplantation (HSCT) in patients with MRD re-emergence warrants further investigation. In this study, we presented the long-term follow-up of R/R and MRD re-emergent B-ALL patients treated with CD19 CAR T-cell therapy and analyzed the factors associated with treatment response and durable survival in our institution.
Materials and Methods
1. Trial design and participants
This phase 2, single-center trial aimed to investigate the efficacy of CD19 CAR T-cell therapy for patients with R/R B-ALL and those experiencing MRD re-emergence. From June 2017 to March 2021, a total of 115 eligible patients were enrolled in our clinical trial, registered at the Chinese Clinical Trial Registry as #ChiCTR-OPN-17013507 (Fig. 1). The study protocol was approved by the Peking University People’s Hospital Ethics Committee and conducted in accordance with the principles of the Declaration of Helsinki. Written informed consent was obtained from all patients or their guardians.
2. Clinical procedures
Eligible patients, aged under 25 years old, included those with CD19-positive R/R B-ALL or MRD re-emergence, defined as having at least two consecutive detectable MRD levels (sensitivity 1:10,000), despite achieving CR based on morphological assessment. Patients with extramedullary disease (EMD), such as central nervous system leukemia (CNSL), or those with a history of allogeneic HSCT (allo-HSCT) without active graft-versus-host disease, were also considered eligible.
Three to ten days before CAR-T-19 infusion, lymphodepletion chemotherapy was administered, primarily consisting of cyclophosphamide (250 mg/m2 for 3 days) and fludarabine (25 mg/m2 for 3 days). Some patients additionally received high-dose methotrexate, ifosfamide, or idarubicin based on the MRD condition in the bone marrow and extramedullary sites. CAR-T-19 cells were then administered via intravenous infusion, with no reported adverse reactions related to the infusion.
Patients achieving CR were offered the option of allo-HSCT, and the conditioning regimen for HSCT followed previously reported protocols [10].
3. Anti-CD19 CAR T-cell manufacture
The CD19 CAR T-cell therapy was conducted according to established methods [10,11]. Peripheral blood was collected from either patients or donors, and peripheral blood mononuclear cells were isolated and transduced with a lentiviral vector encoding the anti-CD19 CAR gene (including the CD19 recognition domain, transmembrane link domain, 4-1BB intracellular domain, and CD3ζ intracellular domain). Following lentiviral transduction, CAR-T-19 cells were cultured in vitro for approximately 5 to 11 days to obtain sufficient numbers of cells for infusion.
The final product, CAR-T-19, was prepared by diluting CAR-T cells with 2% human albumin-containing saline. The main component of the final product consisted of CAR-positive CAR-T-19 cells, with a small proportion of natural killer cells. The cell preparation process met the product quality and activity requirements outlined in the guidelines for CAR-T cell therapy products. The final product exhibited a cell activity of not less than 70% and remained effective within 12 hours at the temperature ranging from 10℃ to 25℃.
4. Evaluation of outcomes and toxicity
Baseline bone marrow assessments were performed before lymphodepletion to evaluate pre-CAR disease status. The initial therapeutic response was evaluated 14 to 30 days after CAR-T-19 infusion using multiparameter flow cytometry (FCM) of a bone marrow aspirate, with routine surveillance thereafter. MRD negativity was defined as the absence of detectable leukemic blasts (less than 0.01%) among mononuclear cells by FCM. Patients with CNSL before CAR-T-19 therapy were also assessed by cerebrospinal fluid (CSF) examination via FCM following CAR-T-19 infusion. MRD-negative CR was defined as the absence of leukemic blasts in the bone marrow by FCM and no evidence of extramedullary leukemia.
Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) were graded based on the American Society for Transplantation and Cellular Therapy consensus criteria reported by Lee et al. [12]. CRS grade 3 or higher was considered severe. B lymphocyte aplasia (BCA) refers to the absence of CD19+ cells in blood samples by FCM.
5. Evaluation of CAR-T-19 cell counts in blood
Blood samples were collected before lymphodepletion, before CAR-T-19 infusion, and at intervals after infusion. The absolute counts of CAR-T-19 cells were determined using FCM, quantified using quantitative polymerase chain reaction (qPCR), and reported as the number of transgene copies per μg DNA. The area under the curve (AUC) of CAR T-cells (transgene copies per μg of DNA) at day 28 was calculated using a trapezoidal rule computational algorithm.
6. Statistical analyses
The outcome data used for the analysis were last updated on August 31, 2023. OS was calculated from the date of CD19 CAR T-cell infusion until the last follow-up or death. Leukemia-free survival (LFS) was calculated from the date of infusion to relapse or death, or last follow-up. Data were not censored when new therapy (including chemotherapy and allo-HSCT) was initiated in the absence of disease. KaplanMeier curves were generated for LFS and OS, and subgroup comparisons were performed using the log-rank test. Cox regression analysis was conducted to identify factors associated with LFS and OS. The Fisher’s exact test was used to compare the difference of categorical variables among the subgroups, while the Kruskal-Wallis or Wilcoxon MannWhitney tests were used to compare continuous variables. Logistic regression was employed to evaluate factors influencing CRS. Statistical analysis was performed using SPSS ver. 23.0 (IBM Corp., Armonk, NY). A p-value < 0.05 was considered statistically significant.
Results
1. Patients
Between June 2017 and March 2021, a total of 115 patients were consecutively enrolled in the study. The median age at infusion was 8 years (range, 1 to 25 years). Among the participants, 12 patients had EMD, 14 patients had undergone a prior transplantation before infusion, and six had previously received CD19 CAR T-cell therapy (S1 Table). Detailed characteristics of the patients are summarized in Table 1. Among the enrolled patients, 67 presented with R/R B-ALL, with five being primary refractory cases, while 48 had MRD re-emergence. We also analyzed the characteristics of patients with MRD re-emergence and R/R B-ALL (Table 1). There were no differences in baseline characteristics before CAR T-cell infusion between the two groups, except for a lower MRD level in the MRD re-emergence group.
Out of the 115 patients, 111 patients received LD-Flu/Cy treatment, while four were treated with alternative regimens, including high-dose methotrexate (n=1), cyclophosphamide alone (n=1), or cytarabine (n=2). The median dose of infused cells was 4.03 (range, 0.03 to 6.33)×106 CAR-T cells per kg. The median CAR-T cell transfection efficiency was 41.9% (range, 2.6% to 83.6%).
2. Treatment outcomes
All patients attained morphologic CR and 111 out of 115 (96.5%) achieved MRD-negative CR at 30 days after CD19 CAR-T cell infusion. Due to the limited number of patients in the MRD-positive CR category (n=4), factors influencing the response to treatment were not analyzed.
With a median follow-up time of 48.4 months (range, 3.2 to 75.2 months), the estimated 4-year OS and LFS rates were 70.7%±4.3% and 68.7%±4.5%, respectively. The median time to relapse was 9.9 months (range, 0.4 to 40.4 months) postCAR T-cell infusion among the 37 patients who relapsed. At relapse, the CD19 immunophenotype on leukemia blasts was CD19+ (n=16), CD19-/dim (n=10), and CD19 unknown (n=11). Notably, one patient experienced myeloid lineage switch and succumbed to transplant-related mortality (TRM). Among the 37 relapsed patients, 19 subsequently underwent HSCT, nine received chemotherapy, CD22-specific CAR-T cell therapy, or donor lymphocyte infusion (for those with prior allo-HSCT), and nine cases did not receive any additional treatment. Univariate predictors of outcome are presented in Table 2. Subsequently, multivariable modeling was performed using variables with p < 0.3 from univariable analyses. In the multivariate analysis, being female, receiving allo-HSCT after CAR-T, and achieving MRD-negative CR were associated with better LFS (Table 3).
3. Toxicity
CRS manifested in 86 patients (74.8%), with 34 patients (29.6%) experiencing grade 3-4 CRS. ICANS occurred in 22 patients (19.1%), with 12 patients (11.3%) experiencing grade 2-4 ICANS (Fig. 2A). Thirty-seven patients received tocilizumab, and 21 patients received steroids. Patients aged over 10 years, those with bone marrow blasts ≥ 5%, and those with MRD ≥ 10–3 before infusion exhibited a significantly higher incidence of severe CRS. Patients with a history of HSCT tended to have a higher incidence of grade 3 to 4 CRS, although the difference was not statistically significant when compared to those without a prior transplantation history (p=0.081). In univariate analysis, the MRD re-emergent group showed a lower CRS grade (p=0.010). However, disease status did not correlate with CRS severity in multivariate analysis (p=0.494), suggesting that CRS severity is more associated with MRD levels before infusion than baseline disease status (Fig. 2B). Additionally, higher CAR-T cell expansion and IL-6 levels were indicative of severe CRS after infusion (Fig. 2C and D). No significant difference was found among different ICANS groups, partly due to the small number of patients in the severe ICANS group.
Out of the 115 cases, 112 cases (97.4%) showed BCA. B lymphocytes in peripheral blood disappeared within 1 day to 20 days after CAR-T-19 infusion. Except for three patients whose B cells began to rise at 2 weeks after CAR-T-19 infusion, the other patients experienced persistent BCA until 4 weeks after CAR-T-19 infusion. Two untransplanted patients had BCA that lasted longer than 1 year. The median value of BCA persistence time was 30 days (range, 11 to 933 days) after infusion.
Four patients developed hemophagocytic lymphohistiocytosis after CAR-T infusion, and all of them recovered through symptomatic treatment, such as corticosteroids and plasma exchange.
4. Persistence of CAR-T cells in the peripheral blood of patients
Robust proliferation of CAR-T cells was observed in the majority of patients. The median CAR-T qPCR peak level was 37,300 copies, occurring between days 4 and 22 after infusion (median, 11 days). In patients without consolidative HSCT, the median persistence time of CAR-T cells was 61 days. Patients with R/R B-ALL exhibited a tendency toward higher peak proliferation levels and CAR T-cell AUC from day 0 to day 28 (AUC28), but these differences were not statistically significant compared to those with MRD re-emergence (S2A and S2B Fig.). The peak proliferation levels of MRD-negative CR patients (median, 42,389 copies/μg DNA) were relatively higher than those of MRD-positive CR patients (median, 3,712.8 copies/μg DNA) (p < 0.001) (S2C Fig.). Furthermore, MRD-negative CR was associated with a higher CAR-T cell AUC28 (p < 0.001) (S2D Fig.).
5. Role of consolidative therapy (allo-HSCT vs. non-HSCT)
After achieving CR following CAR T-cell treatment, patients were given the option of receiving consolidative transplantation or not. Characteristics of patients who bridged into allo-HSCT or did not bridge were provided in Table 1. As shown in Table 1, patients with a history of transplantation and those experiencing severe CRS after infusion were less inclined to undergo consolidative transplantation after CAR T-cell therapy. Excluding one patient lost to follow-up, 75 out of 114 (65.8%) patients proceeded with consolidative allo-HSCT (66 from haploidentical donors, 5 from human leukocyte antigen-identical sibling donors, and 4 from matched unrelated donors). The median time to allo-HSCT was 68 days from CAR T-cell infusion. The 4-year OS and LFS rates were 72.4%±5.3% and 73.9%±5.5% for patients who underwent allo-HSCT, whereas they were 59.1%±8.4% and 50.1%±8.2% for patients who did not proceed with consolidative transplantation (p=0.014 for LFS and p=0.197 for OS). The median time to relapse was similar in the two groups, occurring approximately 10 months after infusion. TRM was 10.67% for patients bridging into allo-HSCT. For patients not bridging into transplantation, the power of BCA to predict outcome became more pronounced with time, and our study showed that whether loss of BCA in the first 3 months or not led to an LFS rate of 35.8%±10.3% vs. 73.1%±11.7% (p=0.014), while patients whether lost BCA at 6 months or not had an LFS rate of 36.7%±9.3% vs. 87.5%±11.7% (p=0.004) (S3 Fig.). However, the measure is also suboptimal as there are patients with CD19-negative relapse.
6. Efficacy of CAR-T therapy for patients with MRD reemergence vs. those with R/R B-ALL
In the MRD re-emergent group, 66.7% (32/48) of patients received consolidative transplantation after CAR T-cell therapy, which was comparable to the percentage (65.2%, 43/66) in the R/R B-ALL group. The median time to relapse was 44.6 months for patients in the MRD re-emergent group and 39.8 months for patients in the R/R B-ALL group. In the MRD re-emergence group, CD19 immunophenotype on leukemia blasts was CD19+ (n=5) and CD19-/dim (n=4), while in the R/R B-ALL group, the corresponding patients were 11 and 6, respectively. There were no significant differences in longterm efficacy between the different disease statuses before infusion (4-year OS: MRD re-emergence vs. R/R B-ALL, 70.6%±6.6% vs. 66.5%±6.1%, p=0.755; 4-year LFS: MRD reemergence vs. R/R B-ALL, 67.3%±7.0% vs. 63.8%±6.2%, p=0.704) (Table 2). For MRD re-emergent patients, there was no significant difference in OS and LFS whether they underwent HSCT or not after CAR T-cell therapy (4-year OS: non-HSCT vs. HSCT, 81.3%±9.8% vs. 65.3%±8.5%, p=0.252, Fig. 3A; 4-year LFS: non-HSCT vs. HSCT, 73.9%±11.3% vs. 63.9%±8.8%, p=0.317, Fig. 3B). However, patients with R/R B-ALL had a better 4-year OS and LFS if HSCT was performed after CAR T-cell infusion (4-year OS: non-HSCT vs. HSCT, 44.7%±11.0% vs. 78.0%±6.6%, p=0.020, Fig. 3C; 4-year LFS: non-HSCT vs. HSCT, 32.6%±10.4% vs. 81.4%±6.4%, p < 0.001, Fig. 3D). S4 Table presented the LFS and OS in different patient and treatment characteristics between MRD re-emergence and R/R B-ALL group.
Discussion
We investigated the safety and efficacy of CD19 CAR T-cell therapy in a large cohort of patients with MRD re-emergence and R/R B-ALL at our single institution. With a median follow-up of 48.4 months, the estimated 4-year OS and LFS rates for the 115 patients treated with CD19 CAR T-cell therapy in our study were 70.7%±4.3% and 68.7%±4.5%, respectively. These results were comparable to those reported in other studies [13,14].
All patients achieved morphologic CR, and the rate of MRD-negative CR rate after infusion was 96.5%, significantly associated with the in invo amplification of CAR-T-19. On the other hand, the expansion of CAR-T-19 in vivo was the primary cause of CRS. Consequently, monitoring the expansion of CAR-T cells in vivo could aid in predicting treatment response and the severity of CRS. Moreover, a high tumor burden before infusion also indicated severe CRS. Fortunately, in our study, none of the patients succumbed to severe CRS-related complications.
Previous studies have indicated that a high tumor burden, the presence of TP53 mutation, and the absence of consolidative HSCT post-infusion are associated with inferior outcomes in the aftermath of CAR-T cell therapy [7,9,15]. In our multivariate analysis, males, MRD-positive CR after infusion, and CAR-T without subsequent consolidative allo-HSCT were identified as independent risk factors for lower LFS. The disease status and tumor burden before infusion did not significantly influence LFS or OS, partly because most of the patients, particularly those with R/R B-ALL, had already undergone treatment interventions such as chemotherapy prior to lymphodepletion.
Patients with the TCF3::PBX1 fusion gene typically manifest a more aggressive disease course [15,16]. In our study, individuals with TCF3::PBX1 positivity tended to exhibit lower OS and LFS rates, although the difference did not reach statistical significance, possibly due to the small sample size. Nevertheless, given the limited number of cases, consolidation allo-HSCT or careful observation should be considered for this subgroup. Patients with BCR::ABL or KMT2Ar fusion genes, initially classified as medium to high risk at initial diagnosis of ALL, demonstrated a prognosis similar to that of patients without fusion genes in our study. This suggests that CAR-T cell therapy might overcome the poorer outcomes associated with genomic instability. Most patients who had previously undergone allo-HSCT did not proceed to a second transplant for various reasons; however, we did not observe statistically significant differences in outcomes between patients with or without a history of HSCT. Therefore, CAR T-cell therapy may be an effective and safe option for patients experiencing relapse after transplantation, especially in the absence of more effective treatment options. Few studies have described CAR-T therapy for B-ALL patients with CNSL or other extramedullary relapse, primarily due to concerns regarding poor response and treatment-related complications such as ICANS. Our results suggest that CD19 CAR-T can induce similar high response rates patients with EMD (4-year OS: with EMD vs. without EMD, 81.8%±11.6% vs. 69.6%±4.6%, p=0.348; 4-year LFS: with EMD vs. without EMD, 72.7%±13.4% vs. 67.1%±4.8%, p=0.613). CD19 CAR T-cell therapy may offer a potential treatment option for previously excluded patients with CNSL or other extramedullary relapse, with manageable neurotoxicity.
Consolidative HSCT is considered to be an effective treatment for ensuring long-term survival post–CAR-T therapy, primarily due to challenges in sustaining CAR-T cells in vivo and the potential for the escape or downregulation of target tumor antigen expression [17]. In our study, bridging to HSCT significantly impacted LFS and OS in R/R B-ALL patients (4-year OS: non-HSCT vs. HSCT, 44.7%±11.0% vs. 78.0%±6.6%, p=0.020; 4-year LFS: non-HSCT vs. HSCT, 32.6%±10.4% vs. 81.4%±6.4%, p < 0.001). These findings were consistent with those reported in other studies [7,9,18]. Therefore, timely implementation of bridging therapy is crucial for R/R B-ALL patients after CAR T-cell treatment. However, in the overall group, whether to undergo transplantation or not exhibited significant differences only in terms of LFS, not OS. This discrepancy may be attributed to the inclusion of MRD re-emergent patients and a high incidence of treatment-related mortality.
Initial clinical studies of CAR T-cell therapy primarily focused on R/R B-ALL. However, the presence of positive or re-emergent MRD during treatment is closely associated with recurrence post-treatment [5,19,20]. The Children’s Oncology Group initiated a phase 2 open-label study (AALL1721) targeting patients younger than 25 years at initial diagnosis of CD19-positive B-ALL with de novo National Cancer Institute (NCI) high-risk features and MRD ≥ 0.01% after consolidation chemotherapy. The study administered tisagenlecleucel following maintenance chemotherapy, with the aim of demonstrating the durable persistence of CAR-T cells in patients and potentially obviating the need for HSCT [21].
In our study, 48 B-ALL patients with MRD re-emergence were treated with CD19 CAR T-cell therapy. The LFS and the OS rates were 67.3±7.0% and 70.6±6.6%, respectively. Whether or not bridging to HSCT did not significantly affect the survival in this subgroup. As previously reported, the 5-year OS and cumulative incidence of relapse for patients with re-emergent MRD were 49.8±4.3% and 60.2±4.3%, respectively. Therefore, patients experiencing MRD re-emergence may also benefit from CAR T-cell therapy and may even be exempted from HSCT. These results suggest that we may extend the indications of CAR T-cell treatment to MRD re-emergent patients. However, further studies are warranted to substantiate this conclusion.
There were some limitations in our study. Firstly, while our results indicate that HSCT following CAR-T therapy can improve the overall outcome of R/R B-ALL patients achieving MRD-negative CR, it remains uncertain whether CAR-T treatment can make R/R B-ALL patients benefit from HSCT. Further investigations are needed to compare the survival outcomes between patients who received CAR-T cell therapy before HSCT and those who underwent HSCT directly. Secondly, for MRD re-emergent patients, owing to the limited duration of CAR-T cells in vivo, alternative treatments other than HSCT are still necessary to maintain LFS. Therefore, a multicenter study on MRD re-emergent patients receiving CAR T-cell therapy is imperative to identify which patients may not require HSCT after CAR T-cell therapy in the future.
In summary, our study showcased the effectiveness of CD19 CAR T-cell therapy in enabling children and young adults with R/R B-ALL to achieve MRD-negative CR. Bridging to HSCT after CAR T-cell infusion proved to be a beneficial factor in extending OS. For MRD re-emergent patients, CAR T-cell therapy effectively eradicated MRD and prevented rapid disease progression. This resulted in some patients achieving long-term survival without the need for HSCT.
XML Download