CD19/CD22-targeted CAR T therapy for relapsed/refractory B-cell malignancies

Chimeric antigen receptor (CAR) T-cell therapy is under investigation for the treatment of several hematological malignancies. Currently, the only United States (U.S.) Food & Drug Administration (FDA) and European Medicines Agency (EMA) approved indications for CAR T are diffuse large B-cell lymphoma (DLBCL) and acute lymphoblastic leukemia (ALL). Many of the CAR constructs being investigated in clinical trials for the treatment of B-cell malignancies target CD19.^1-4 However, malignant B cells have developed mechanisms to evade immunotherapies that target specific antigens such as CD19. These include mutation, alternative splicing, protein misfolding of CD19 or myeloid lineage switch of CD19-positive malignant cells. Antigen escape can lead to relapse following CD19-targeted therapies and these patients have poor outcomes.

One proposed method of reducing antigen escape, and subsequently lowering relapse rates, is to infuse two single-specific CAR T products against different antigens. Na Wang, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, CH, and colleagues, therefore conducted a pilot study (ChiCTR-OPN-16008526), in 89 patients with relapsed/refractory (R/R) B-cell malignancies, using two CAR T products administered sequentially.⁵

The Lymphoma Hub previously covered CAR T therapy as an educational theme. Read more here.

Study design⁵

• Open label, single center, single-arm pilot study
• Aim: evaluate the efficacy and safety of sequential infusion of two single-targeted CAR T products (anti-CD19 and anti-CD22)
• CAR constructs:
  • Third-generation CAR
  • Single chain variable fragment from a murine monoclonal antibody against CD19 or CD22
  • Two costimulatory domains: CD28 and 4-1BB
  • Activation domain: CD3ζ

Patient characteristics5

• Patients with R/R B-cell malignancies screened: 105
  • Sixteen patients did not receive the CAR T product due to production failure, serious infection, withdrawal of consent and disease progression (PD)
  • In total, 84.4% (n=89) of patients who were screened received the CAR T infusions
  • ALL; n=51
  • Non-Hodgkin lymphoma (NHL); n= 38
    • All patients had aggressive clinical disease
    • Twenty-three patients had relapsed three or more times, or had primary refractory disease
• Median age (total cohort): 36 years (range: 9–71)

Lymphodepletion, chemotherapy and CAR T manufacturing and infusion

• Lymphodepletion chemotherapy was given for three days (days -4 to -2):
  • Fludarabine: 25mg/m²
  • Cyclophosphamide: 300mg/m²
• The two CAR T-cell products were separately infused on successive days from day 0 (81% of patients received anti-CD22 CAR T cells first) and were generally given in two divided doses
  • Median manufacturing time:
    • Anti-CD19 CAR T product: 12 days (range: 8–20)
    • Anti-CD22 CAR T product: 13 days (range: 9–22)
  • Dosing of CAR T products for patients with B-cell NHL:
    • Anti-CD19 CAR T product: 5.1 ± 2.1 ×10⁶/kg
    • Anti-CD22 CAR T product: 5.3 ± 2.4 ×10⁶/kg
• Treatment occurred between March 2016 and January 2018
• Patients were followed-up until death, lost to follow-up, or withdrawal of consent

Efficacy⁵

Reported here only in patients with NHL. To read the results in ALL, please refer to the original paper.

In patients with NHL who were evaluable for efficacy (n= 36), with a minimum follow-up of three months, 18 patients had achieved complete responses (CR), with eight more achieving a partial response (PR), giving an overall response rate (ORR) of 72.2% (Table 1). The ORR was consistent across subgroups when analyzed by pathologic subtype or cytogenetic risk.

Table 1. Efficacy of CAR T infusions in patients with B-cell NHL, and sub-analysis in patients with diffuse large B-cell lymphoma (DLBCL)

* Median follow-up of 14.4 months (range 0.4–27.4) in patients with B-cell NHL, and 14.3 months (0.4–27.3) in patients with DLBCL

 

Patients with B-cell NHL

(n= 36)

95% CI

Patients with DLBCL

(n=23)

95% CI

ORR

72.2%

54.8–85.8

-

-

CR

50.0%

32.9–67.1

-

-

PR

22.2%

10.1–39.2

-

-

Median progression-free survival (PFS)*

9.9 months

3.3–not reached (NR)

5.8 months

3.0–NR

Twelve-month PFS

50.0%

33.4–64.5

47.8%

26.8–66.1

Median overall survival (OS)*

18.0 months

6.1–NR

18.0 months

5.2–NR

Twelve-month OS

55.3%

38.3–69.3

-

-

Responses:

• Ongoing responses, at data cut-off (31^st March 2019):
  • CR: 15/18 maintained response
  • PR: 3/8 maintained response
  • Eleven patients subsequently underwent SCT (nine autologous and two allogeneic)
    • Six maintained initial CR without PD
• In patients who did not respond at three months:
  • Median progression-free survival (PFS): 2.4 months (95% CI, 1.0–2.7)
  • Median overall survival (OS): 5.1 months (95% CI, 1.9–6.1)

Analysis by disease status, response at three months, and cytogenetics:

• Patients receiving CAR T infusions at first relapse had superior survival to those with primary refractory disease or multiple relapses
• Patients who achieved an ORR at month three (R3M) had a significantly prolonged PFS (p<0.0001) and OS (p<0.0001) compared to others
• Prognostic impact of genetic abnormalities:
  • Patients with MYC rearrangements (including those with MYC and BCL2 or BCL6), median PFS and OS were NR
  • Of patients with del(17p) or TP53 mutation, median PFS was 8.8 months (95% CI, 1.0–NR) and median OS was 14.5 months (95% CI, 2.7–NR)

Disease progression:

• PD eventually occurred in 50.0% of patients (n=18)
• Median time to progression (TTP): 3.3 months (range: 1.0–14.8)
• At the time of progression, CD19 and/or CD22 CAR T cells were detected in two patients, with B-cell aplasia in three
• However, repeated biopsy and immunophenotyping did not reveal loss of CD19 or CD22

Safety⁵

For total cohort (patients with B-ALL and B-cell NHL)

• Cytokine release syndrome (CRS; graded as per Lee criteria⁶), any grade: 95.5% (n=85)
  • Grade 1–2 CRS: 77.6%
  • All incidences of grade ≥ 3 CRS and neurotoxicity were reversible aside from one
    • Grade 5 event in a patient with B-ALL
  • High-grade (grade ≥ 3) CRS was significantly associated with:
    • Higher levels of serum interleukin-6 (IL-6) in B-ALL and B-cell NHL
    • Higher percentages of bone marrow blasts in B-ALL
    • Increased levels of serum C-reactive protein in B-cell NHL
    • Higher international prognostic index scores in B-cell NHL
  • High-grade (grade ≥ 3) CRS was not significantly associated with:
    • Disease stage
    • CAR T dosage
  • CAR T-cell-related encephalopathy syndrome (CRES): 12 patients (13.5%; all reversible)
• Most common ≥ grade 3 adverse events (AEs) during the first month: cytopenias
  • The most common grade 3 or higher AEs occurring in more than 5% of patients are shown in Table 2
• Fatal AEs within the first month: 3 patients
  • One patient died of grade 5 CRS and pneumonia despite the administration of tocilizumab, glucocorticoid and plasma exchange
  • Two patients died of serious pneumonia following infection with cytomegalovirus or stenotrophomonas maltophilia
• Fatal AEs after the first month: 9 patients
  • Cause of death was an infection in five of these patients

Table 2. AEs grade ≥ 3 occurring in ≥ 5% of patients within the first month of infusion

N=89

Grade 3, %

Grade 4, %

Grade 5, %

Lymphocytopenia

0

100

0

Neutrocytopenia

2

98

0

Leukopenia

7

92

0

Anemia

72

10

0

CRS

17

3

1

Fever

36

0

0

Thrombocytopenia

11

52

0

Hypotension

9

0

0

Activated partial thromboplastin time prolonged

6

0

0

Hypokalemia

13

3

0

Diarrhea

6

0

0

Aspartate aminotransferase  increase

7

0

0

Lung infection

18

0

3

Conclusion

The sequential infusion of two separate CAR T products, each targeting a different B-cell antigen, has shown robust responses with a reduction in antigen escape in patients with R/R B-ALL and B-cell NHL.

In patients with B-cell NHL, the survival advantage was most apparent in patients who achieved an overall response at three months, indicating R3M may be a good predictor of survival in this population. This approach was also most beneficial at first relapse, indicating a potential role as a second-line treatment for patients with R/R B-cell NHL.

However, superior long-term survival was not observed following this dual-targeted approach, with comparable efficacy to single-specific CAR T-cells. The authors suggest efforts are focused on sustaining CAR T persistence in vivo to enable CAR T to become definitive, as opposed to bridging, therapy.