Safety and PK/PD of ALLO-647, an anti-CD52 antibody, with fludarabine (Flu)/cyclophosphamide (Cy) for lymphodepletion in the setting of allogeneic CAR-T cell therapy

2527 Background: Allogeneic chimeric antigen receptor (CAR) T cell therapy holds promise in addressing logistical/manufacturing challenges of autologous CAR T cell therapy. ALLO-501 (anti-CD19; uses Cellectis technologies) and ALLO-715 (anti-BCMA) are allogeneic CAR T cell products whose a) disrupted TCRα constant gene may reduce GvHD risk, and b) edited CD52 gene may permit use of ALLO-647 (a humanized anti-CD52 mAb) to selectively deplete host T cells. Methods: The ongoing ALPHA (ALLO-501) and UNIVERSAL (ALLO-715) trials include patients (pts) with relapsed/refractory large B-cell or follicular lymphoma and multiple myeloma, respectively. The lymphodepletion regimen included ALLO-647 (39 mg [low dose, LD; n = 33], 60 mg [n = 12], or 90 mg [high dose, HD; n = 27]) with Flu 30 mg/m2/d x 3d +/- Cy 300 mg/m2/d x 3d (Flu+Cy, n = 66; Cy, n = 6) (ALLO-647/Flu +/- Cy) before CAR T infusion. A mixed-effects population pharmacokinetic (PK) model was fit to ALLO-647 concentration vs. time data. Pharmacodynamic (PD) effects on host T cells, IL15, and CAR T cell expansion were also studied. Results: As of the data cut, 72 pts were treated. Common Grade ≥3 AEs were neutropenia (71%), thrombocytopenia (42%), anemia (39%), and lymphopenia (28%). Neutrophil, hemoglobin, and platelet counts did not differ by ALLO-647 dose, suggesting these dim-CD52+ cells were unaffected. Gr ≥3 infections were seen in 21% pts; 33 had infusion-related reactions (IRR; Gr 1: 13%; Gr 2: 32%; Gr 3: 1.4%). IRR incidence and severity were higher with HD ALLO-647. The optimal PK model included a saturable (concentration-dependent) elimination pathway; clearance varied as a function of baseline lymphocyte count (LC). Serum ALLO-647 levels increased with dose; median modelled Cmax was 4,224 and 14,139 ng/mL for LD and HD, respectively. With ALLO-647/Flu +/- Cy, all but 2 pts reached a LC nadir < 0.05x109 cells/L, typically by D-3. Duration of lymphodepletion was typically longer with HD ALLO-647. Median duration of T-cell suppression ( < 10 cells/µl) was ̃ 8.5 and 13.6 days from CAR T infusion, respectively, for LD and HD ALLO-647. With HD, T cell counts were < 10 through D+28 in 17 pts and > 10 in 2. In 68 evaluable pts, compared to LD, HD ALLO-647 was associated with higher D0 serum IL15 levels, which have been linked to improved clinical response (eg, Kochenderfer JCO 2017). Higher CAR T expansion was also observed post D+14 with HD ALLO-647 compared to LD, creating an opportunity for clinical response. Conclusions: ALLO-647/Flu +/- Cy had a tolerable safety profile and produced a deep and durable window of lymphocyte depletion. ALLO-647 exhibited target-mediated drug disposition; clearance increased with higher baseline LC. HD was associated with higher IL15 levels and better CAR T expansion, suggesting dose responses. Enrollment in both studies is ongoing; updated safety and PK/PD data will be presented. Clinical trial information: NCT04093596.