Paper Details

ABSTRACT

Introduction: ABBV-2001 (also known as ISB-2001) is a first- and only-in-class trispecific T-cell engager (TCE) targeting CD38 and BCMA on myeloma cells and CD3 on T cells. Dual tumor-antigen engagement increases avidity-based binding, stabilizes immune synapses, and enhances cytotoxicity across variable antigen expression levels, while distal positioning of the CD38-binding domain may reduce off-tumor adverse events (AEs). TRIgnite-1 (NCT05862012) is an ongoing first-in-human global phase 1 dose-escalation (Part 1) and expansion (Part 2) study of ABBV-2001 in RRMM. Previously reported Part 1 data demonstrated favorable safety for ABBV-2001 and robust activity in difficult-to-treat patient (pt) subgroups (Lichtman et al., 2025). Here we report longer follow-up data from Part 1 and initial results from expansion Cohorts A and B of Part 2. 

Methods: Eligible pts with RRMM were triple class exposed (immunomodulatory drugs, proteasome inhibitors, and anti-CD38 therapies) and had ≥3 prior lines of therapy (pLoT). Prior BCMA exposure (including TCEs, CAR-T and ADC) was allowed. Part 1 included 9 dose levels (DL) ranging from 5–2700 µg/kg weekly (Q1W) with 2 step-up doses (SUD) on Cycle (C) 1 Day (D) 1 and C1D4 followed by target dose from C1D8 onwards. Part 2 included Cohorts A (600 µg/kg) and B (1200 µg/kg) with the same SUD schedule. Target doses were administered Q1W until C3D1, Q2W until C5D1, then Q4W or Q2W based on response. All doses were administered subcutaneously. 

Results: As of data cutoff, total n=70 pts were enrolled and included in the analysis. In Part 1, 42 pts received ABBV-2001; median (med) age 65 years (yrs); med (range) pLoT 6 (3–11); 57% anti-CD38 refractory, 45% prior BCMA exposure, 40% extramedullary disease (EMD) at baseline, and 42% with high-risk cytogenetics. For Part 2 (A and B; N=28), med age 72 yrs; med (range) pLoT 5 (2–15); 50% anti-CD38 refractory, 39% prior BCMA exposure, 22% EMD at baseline, and 48% with high-risk cytogenetics.  Med (range) follow-up was 12 (1–30) mo for Part 1 and 10 (1–15) mo for Part 2.

In Part 1, no dose-limiting toxicities occurred. Treatment-emergent adverse events (TEAEs) leading to treatment discontinuation occurred in 7%. Without prophylactic tocilizumab use, cytokine release syndrome (CRS) was 71% (G1: 57%; G2: 14%; no ≥G3) with med time to onset and resolution of 24 hours. Two pts experienced immune effector cell-associated neurotoxicity syndrome (ICANS; G1: 2%; G2: 2%; no ≥G3). G3/4 infections were 33%. In Part 2, ≥G3 TEAE incidence was 71%; no TEAEs led to treatment discontinuation. Without prophylactic tocilizumab, CRS was 68% (G1: 64%; G2: 4%; no ≥G3), with 1 pt with G2 ICANS. G3 infections were 17% (no G4 or fatal events). No G5 treatment-related AEs were reported for Parts 1 and 2.

In heavily pretreated patient population, of efficacy evaluable pts in Part 1 DL3–9 (N=40) and Part 2 (A and B; N=27), the overall response rates (ORRs) were 75% and 85% (≥complete response [CR]: 35% and 59%), respectively. Med (range) time to first response was 1.2 (1–6) and 1.2 (0–4) mo, and med (range) time to ≥CR was 5.4 (1–12) and 5.5 (3–12) mo, respectively. Among all efficacy evaluable BCMA-naïve pts, ORR of 83% was observed (Part 2: 95%), while ORR of 77% was observed in pts with prior BCMA CAR-T or TCE exposure (Part 2: 80%). Among all efficacy evaluable anti-CD38 mAb sensitive pts, ORR of 88% was observed (Part 2: 92%).  Among all evaluable pts with CR/stringent CR, minimal residual disease negativity (MRDneg) at 10^–5 was achieved in 83% (Part 2: 88%). ABBV-2001 demonstrated a med half-life of 17 days and consistent T-cell activation, further supporting a less frequent dosing schedule.

Conclusions: With longer follow-up and more pts treated, ABBV-2001 continued to show favorable safety (≤10% G2 CRS with no ≥G3 CRS without prophylactic tocilizumab) and deep activity (83% MRDneg) in heavily pretreated pts with RRMM. With less frequent dosing, robust efficacy was observed with an ORR of 85% (≥CR 59%). Alternative SUD schedule and monthly treatment intervals are being evaluated to determine the optimal RP2D. Updated data will be presented at the meeting.