Phase I results for anti-BCMA CAR T-cell therapy bb2121 

In May 2019, Noopur Raje, Massachusetts General Hospital Cancer Center, Boston, MA, Jesus Berdeja, Sarah Cannon Research Institute and Tennessee Oncology, Nashville, TN, Yi Lin, Mayo Clinic, Rochester, MN, and colleagues, published the results from a phase I dose escalation and expansion trial (CRB-401), investigating bb2121, a proposed novel CAR T-cell therapy, in patients with relapsed/refractory (RR) multiple myeloma (MM).¹

Background

The anti-B-cell maturation antigen (BCMA) CAR T-cell therapy, bb2121, has shown promising preclinical results across a range of BCMA expression levels. This led to the development of the CRB-401 (NCT02658929) multi-center clinical trial to evaluate the safety and efficacy as a potential treatment for RRMM.² bb2121 is formed by transducing autologous peripheral-blood mononuclear cells with a second-generation chimeric antigen receptor (CAR):

• Vector: lentiviral
• Single-chain variable fragment: anti-BCMA
• Co-stimulatory domain: CD137 (4-1BB)
• Signaling domain: CD3-zeta

Patient characteristics

• Patients with RRMM (N = 33)
  • Three (or more) prior lines of therapy including a proteasome inhibitor (PI) and an immunomodulatory drug (IMiD), or refractory to both classes:
    • Prior autologous stem cell transplant: 97% (n = 32)
    • Refractory to both PI and IMiD: 79% (n = 26)
  • Additional inclusion criteria, by trial stage were:
    • Dose-escalation phase: BCMA expression ≥ 50% on marrow plasma cells
    • Dose-expansion phase: prior exposure to daratumumab and refractory to most recent line of therapy
• Median age: 60 (37–75)
• Stage II or III disease: 67%
• High-risk cytogenetics (del[17p], t[4;14] or t[14;16]): 45%

Study design

• Prior to receiving the bb2121 CAR T infusion, patients were lymphodepleted on days -5, -4 and -3 with:
  • Fludarabine: 30 mg/m²/day
  • Cyclophosphamide 300 mg/m²/day
• In total, 36 patients were enrolled and underwent leukapheresis
  • Leukapheresis success rate: 100%
  • Three patients discontinued prior to bb2121 infusion due to disease progression
• Dose escalation cohort (n = 21):
  • Single infusion of bb2121 at doses: 50 x 10⁶, 150 x 10⁶, 450 x 10⁶ or 800 x 10⁶
  • Median prior lines of therapy: 7 (3–14)
• Dose expansion cohort (n = 12):
  • Range: 150 x 10⁶ to 450 x 10⁶
  • Median prior lines of therapy: 8 (3–23)
• Primary end point: safety
• Secondary end points: response rate and duration of response
• Data cut-off: 6.2 months

Safety

• All patients experienced an adverse event (AE), with 97% (n = 32) experiencing a ≥ grade 3 event
• Non-hematologic events grade ≥3 were uncommon
  • Grade ≥3 events were: fatigue (3%), pyrexia (3%), hypophosphatemia (9%), peripheral edema (3%), hyperglycemia (3%), hypotension (6%) and hyponatremia (6%)
  • Infections occurred in 42% (n = 14) of patients; 2 were grade 3
  • The most common other non-hematological events (>30%) of any grade were fatigue (42%) and headache (30%)
• Hematological AEs of grade 3 and above were most commonly neutropenia (85%), leukopenia (58%), anemia (45%) and thrombocytopenia (45%)
• Neurologic AEs occurred in 42% (n = 14) of patients, of which 39% (n = 13) were grade 1–2
  • One grade 4 event occurred 11 days post-infusion and was reversible
• Cytokine release syndrome (CRS): 76% (n = 25)
  • Graded by Lee et al. criteria³
  • Most (70%) were grade 1–2 events, but 6% (n = 2) were grade 3
  • Median time to onset: 2 days (1–25)
  • Median duration: 5 days (1–32)
  • Treatment for CRS (n = 11):
    • Tocilizumab (n = 7)
    • Glucocorticoids (n = 4)
    • Did not negatively affect CAR T expansion
  • Higher rates of CRS were observed with:
    • Doses > 150 x 10⁶
    • High peak CAR T expansion
    • High peak serum C-reactive protein
    • High peak of tumor necrosis factor α
    • High baseline ferritin, tumor-associated free light chains and serum BCMA

Efficacy

Table 1 below shows the best overall responses by the dose of bb2121 administered. The objective response rate (ORR), defined as partial response (PR) or better, for the whole cohort was 85% (95% CI, 68.1–94.9). Three of the patients achieved a complete response (CR, 9%), whilst 12 achieved a stringent CR (sCR, 36%).

Peak CAR T expansion was after 11 days in patients who received doses higher than 150 x 10⁶ CAR T cells. A very good partial response or better (≥VGPR) was only seen in patients receiving over 150 x 10⁶ CAR T cells, indicating frequency and duration of response are dose-dependent.

Factors associated with achieving a PR (or better):

• Higher PR: progressive disease, exposure to daratumumab in most recent therapy, lack of bridging therapy and extramedullary disease
• Lower PR: high-risk cytogenetics, no CRS, dose < 150 x 10⁶ and low in vivo expansion
• Not associated: BCMA levels and prior treatment exposure

For the whole cohort, median progression-free survival (PFS) was 11.8 months (95% CI, 6.2–17.8). The median duration of response was 10.9 months (7.2–could not be estimated) with a median time to first partial response, or better, of 1 month (0.5–3.0).

In 18 patients, minimal residual disease (MRD) status was available, with 16 patients who achieved a partial response (PR) or better, being MRD-negative (≤10^-4).

Based on BCMA expression, response rates were similar. In patients receiving 450 x 10⁶ CAR T cells, 100% of patients with <50% BCMA expression responded, while 91% of patients ≥50% BCMA expression responding.

At a median follow-up of 11.3 months (6.2–22.8), 14 patients had ongoing responses whilst 17 experienced disease progression (12 of whom had responded to therapy).

Table 1: Best overall response by dose of bb2121 (%, [n])

Conclusion

At doses above 150 x 10⁶ CAR T cells, CRs were observed at all levels. Clinical responses were observed in 85% of patients in this study, lasting on average for 10.9 months. The single-infusion administration of bb2121 may offer an advantage over current salvage therapies offered in the RRMM setting, which require more frequent visits and follow-up. The bb2121 CAR T cells were shown to persist in vivo and the safety profile was manageable and in line with anti-CD19 CAR T studies and data.