Brexu-cel + dasatinib: Safety and feasibility of dasatinib pulses after brexucabtagene autoleucel to modulate CAR T cell activity in relapsed/refractory B cell acute lymphoblastic leukemia

Introduction : Brexucabtagene autoleucel (BA) is an effective chimeric antigen receptor T cell (CAR-T) therapy for B cell acute lymphoblastic leukemia (ALL); however, it causes high-grade immune toxicities. Preclinical models show that short dasatinib (D) pulses given after CAR-T transiently disrupt CAR signaling, promoting a reversible “rest” state that improves T cell function. We hypothesized that D given soon after BA in vivo could induce transient rest during rapid expansion, thereby mitigating toxicities while preserving efficacy. We therefore designed a trial to evaluate safety and feasibility of D pulses after BA in relapsed/refractory (r/r) ALL patients (pts). Methods : Adult r/r ALL pts meeting treatment criteria for BA were eligible for this open-label Phase 1b trial at Stanford University (BA+D group). After BA infusion, D 100mg daily was started between Day +4 to +10 and continued on a 3 days on/4 days off pulse schedule during month 1. Primary endpoints were feasibility (≥ 2 D pulses in month 1) and safety. Clinical outcomes were compared to a cohort of 11 Stanford pts who received BA for r/r ALL from 2022-2024 (BA-only group). Results : Eleven pts were enrolled; 4 withdrew prior to receiving BA; 7 were evaluable and received BA+D. At time of BA infusion, pt median age was 44 (range 35-56), median prior lines of therapy was 2 (1-3), 4 pts (57%) had morphologic disease, and 3 (43%) had measurable residual disease (MRD). D pulses began at median Day +5 after BA (range 4-13). Feasibility was met as 6/7 pts (86%) completed ≥ 2 D pulses following BA. D was well-tolerated without excess toxicity. For BA+D vs BA-only, respectively, rates of cytokine release syndrome [CRS, all grade (AG): 100% vs 73%; G3+: 0% vs 0%], were similar; rates of immune effector cell-associated (IEC) neurotoxicity syndrome (ICANS, AG: 29% vs 55%; G3+: 29% vs 36%), and IEC hemophagocytic syndrome (IECHS, AG: 29% vs 36%; G3+: 0% vs 18%) were numerically lower. Rates of complete remission (CR, 85% vs 73%), MRD-negative CR (57% vs 45%), 1-year relapse-free survival (57% vs 44%) and 1-year overall survival (85% vs 73%) were similar. Pts in BA+D who had immune toxicity trended towards lower total steroid exposure relative to similar pts in BA-only (median 2 vs 6 days). For BA+D, CAR-T expansion was assessed by flow assisted cell sorting. 5/7 pts (71%) had CAR-T cells that expanded despite D pulses, but with lower median peak (34 CAR-Ts/uL) than reported with BA-only. All 5 pts with CAR expansion had detectable circulating CAR-T at D+28. Further comparative correlative data will be presented at AACR. Conclusion : This direct translation of preclinical work shows that 3-day pulses of D after BA are feasible and safe. While we could not clearly determine if D pulses reduced CAR-T toxicity in this small cohort, BA does expand in the setting of D.