Long-term data with CD19 CAR T-cell therapy in children and young adults with B-ALL

CD-19-directed chimeric antigen receptor (CAR) T-cell therapy has demonstrated highly encouraging results in children and young adults (CAYAs) with relapsed or refractory (R/R) B-cell acute lymphoblastic leukemia (B-ALL). However, relapse rates are still high and long-term follow-up data are limited. In the R/R setting, allogeneic hematopoietic stem cell transplantation (allo-HSCT) along with chemotherapy, is associated with improved disease-free survival in the long-term; however, the value of consolidative allo-HSCT following CD19 CAR T-cell therapy in CAYAs with R/R B-ALL is unknown.¹

Nirali N. Shah and colleagues conducted a phase I study of CD19.28ζ CAR T-cells in this patient population to investigate whether alternative chemotherapy regimens could reduce disease burden more effectively. Intensified lymphodepletion strategies were also used. Results with a long-term follow-up of 4.8 years from 50 patients were published in the Journal of Clinical Oncology.¹

This was a single center, phase I study (NCT01593696) of CD19.28ζ CAR T-cells (an anti-CD19 single-chain variable fragment plus TCR zeta and CD28 signaling domain).² In the dose-escalation phase, the primary objective was to define the maximum tolerated dose, discover the toxicity profile, and the feasibility of CD19 CAR T-cells. Twenty-one patients were treated in this phase and the results were previously published in The Lancet.² In brief, CD19.28ζ CAR T-cells demonstrated a 90% feasibility, the maximum tolerated dose was 1 × 10⁶ cells/kg, and all toxicities were reversible.²

Methods

• In this current analysis, 30 additional patients were treated in an expansion phase and results from a total of 50 patients were reported. A retrospective analysis was performed to investigate the impact of allo-HSCT on survival.
• Study design was as follows:

• Dose levels (DL): 1 × 10⁶ cells/kg once on Day 0 (DL1) and 3 × 10⁶ cells/kg once on Day 0 (DL2)

• • To reduce disease burden prior to CAR-T infusion, lymphodepletion regimens were evaluated in patients, as seen in Figure 1 below.

Figure 1. Lymphodepletion regimens used in the study* 

Cy; cyclophosphamide; Flu, fludarabine; HD, high dose; LD, low dose; MESNA, 2-mercaptoethane sulfonate; OD, once daily.
*Adapted from Shah et al.^1
†High burden disease was defined by ≥25% bone marrow blasts, circulating peripheral blasts or lymphomatous disease.

Patients

A total of 53 patients were enrolled in the study. All but two patients (who had diffuse large B-cell lymphoma, and who were excluded from these analyses) had B-ALL. One patient with B-ALL could not receive CAR T-cell infusion due to invasive fungal disease. Baseline characteristics of patients with B-ALL who were included in the analyses (n = 50) are presented in Table 1. Forty-five patients were treated at DL1 and five patients were treated at DL2.

Table 1. Baseline characteristics*

Results

Safety

Any grade of cytokine release syndrome (CRS) was reported in 35 patients (70%), and nine patients (18%) experienced Grade 3−4 CRS (Table 2). The median time to CRS onset was 5 days (range, 1−12 days). All events were resolved.

Table 2. Safety outcomes*

There was an association between disease burden and CRS severity. Among responders who received low-dose fludarabine/cyclophosphamide (Flu/Cy), Grade 3−4 CRS events were significantly higher in patients with an ≥M2 marrow (≥5% marrow blasts) compared with those who had an M1 marrow (<5% marrow blasts) (57.1% vs 6.67%, p = 0.005). Eight of nine patients who had Grade 3−4 CRS had ≥M2 marrow blasts. Grade 3−4 CRS and ≥M2 marrow blasts were reported in 70% and 80% of patients with neurotoxicity, respectively.

Efficacy

Of 50 patients, 31 (62.0%) achieved complete response (CR). Of those with CR, 28 patients (90.3) had minimal residual disease (MRD) negativity, corresponding to an overall MRD negativity of 56.0%.

CR rates were higher in patients who:

• were primary refractory (p = 0.0035)
• had fewer prior lines of therapy (p = 0.033)
• had an M1 marrow (MRD-positive, <5% marrow blasts; p = 0.0007)
• received any Flu/Cy regimen for lymphodepletion (p = 0.041)
  • CR rates with low-dose and high-dose Flu/Cy were 71.4% and 57.1%, respectively

• had low-burden disease (M1 marrow) (94.1% vs 45.5% with high-burden disease defined as ≥M2 marrow; p = 0.0007)

All patients with central nervous system (CNS) involvement who had CRS could be treated effectively, and patients achieved marrow response.

Higher CAR T-cell expansion and Grade 3−4 CRS were associated with CR; response rates did not differ with expression of the T-cell exhaustion markers on infused CAR T-cells. CD4+ and CD8+ CAR T-cell populations were balanced between central and effector immunophenotypes. Cytokine peak levels were differentially increased among those with low- and high-grade CRS.

One patient with prior exposure to a CD19 CAR T-cell therapy and four patients with prior exposure to blinatumomab failed to respond to the study treatment.

Long-term outcomes

Table 3 summarizes survival with a median follow-up of 4.8 years (range, 3.5−7.2 years). In the high-burden disease cohort, eight of 15 patients who achieved CR proceeded to allo-HSCT.

Table 3. Survival outcomes*

The impact of consolidative HSCT

Of 28 patients who achieved MRD-negative CR, 21 (75%) underwent allo-HSCT for consolidation; median time to transplantation from CAR infusion was 54 days (range, 42−97 days).

• Median overall survival (OS) from allo-HSCT Day 0 was 70.2 months (95% CI; 10.4 to not-estimable)
• Median event-free survival (EFS) was not reached
• 5-year EFS after allo-HSCT was 61.9% (95% CI, 38.1−78.8)

Eight patients died between 0.8 and 71 months after allo-HSCT due to transplant-associated complications, graft-versus-host disease (GvHD), infection, secondary malignancy 3 years following transplant, and relapse. Cumulative incidence of relapse posttransplant was 4.8% and 9.5% at 12 and 24 months, respectively.

Patients who did not underwent transplantation (n = 7) relapsed at a median of 152 days (range, 94−394) after CAR infusion, emphasizing the role of consolidative allo-HSCT.

Conclusion

This study represents the longest follow-up after CD19 CAR T-cell therapy for B-ALL. A consolidative allo-HSCT after CAR infusion was associated with:

• a relapse rate of <10% at 24 months
• 5-year EFS of 61.9%, and
• a median OS of 70.2 months

In managing B-ALL with CNS involvement, the findings indicate that CNS2 and CNS3 disease can be treated safely and effectively, and support further evaluation of CAR T-cells in this setting. Lymphodepletion with a Flu/Cy-based regimen was associated with improved responses over alternative regimens. Results also indicate that prior CD19-targeted therapies may negatively impact response to CD19 CAR T-cell therapy. Overall, CD19.28ζ CAR T-cell therapy followed by transplant may have a potential for long-term durable disease control in CAYAs with R/R B-ALL.