Combined biparatopic nanobody-based B cell maturation antigen chimeric antigen receptor T cell therapy and pomalidomide for relapsed or refractory multiple myeloma: In-vitro characterisation and phase 1 dose-escalation study

Abstract

Abstract Background Relapsed or refractory multiple myeloma (RRMM) remains incurable with conventional therapies. Existing B cell maturation antigen (BCMA)-directed chimeric antigen receptor T (CAR-T) cells can induce deep responses, yet relapse is common. Pomalidomide, an immunomodulatory drug, augments T cell function and may synergize with CAR-T cells. To improve efficacy and reduce relapse, we generated a biparatopic nanobody-based BCMA CAR-T construct that simultaneously engages two non-overlapping BCMA epitopes. In in-vitro studies this construct exhibited superior cytotoxicity compared with the commercially available ciltacabtagene autoleucel (Carvykti; Legend Biotech) and pomalidomide further promoted memory-like differentiation and prolonged effector persistence. Methods In-vitro work: A humanised, synthetic alpaca-derived nanobody phage library was panned against recombinant monomeric and dimeric BCMA. Two non-competing binders were linked to create the biparatopic nanobody, which was inserted into a second-generation CAR containing a 4-1BB costimulatory domain. Short-term cytotoxicity assays and serial tumour re-stimulation assays (effector:target = 2:1, 5 cycles with 3-day rests) were performed against BCMA-positive targets in the presence or absence of pomalidomide (1µM). Flow cytometry quantified stem cell memory (CD45RA⁺CCR7⁺CD95⁺) and central memory (CD45RA⁻CCR7⁺) subsets. Clinical study: This single-center, single-arm, 3+3 dose-escalation trial enrolled patients ≥18 years with RRMM or plasma cell leukemia, ECOG performance status 0-2, who had progressed after autologous stem cell transplantation or ≥2 prior lines of therapy. Patients received a single infusion of autologous biparatopic BCMA CAR-T cells at three dose levels: DL1 0.75×10⁶ CAR-T cells/kg, DL2 1×10⁶ cells/kg, DL3 ≥1.5×10⁶ cells/kg. Pomalidomide 1-2 mg/day was started on the day of infusion or upon CAR-T expansion decline and given on days 1-21 of each 28-day cycle until progression or intolerance. The primary endpoint was safety and maximum tolerated dose (MTD); secondary endpoints included overall response rate (ORR), minimal residual disease negativity (MRD; next-generation flow, 10⁻⁵), progression-free survival (PFS), overall survival (OS), and CAR-T expansion kinetics. The protocol was approved by the institutional ethics committee; all participants provided written informed consent. Results In-vitro: The biparatopic CAR-T demonstrated a significantly lower EC₅₀ against BCMA-positive targets than ciltacabtagene autoleucel. After five consecutive tumour re-stimulation cycles, biparatopic CAR-T maintained complete target-cell elimination, whereas ciltacabtagene autoleucel showed reduced killing after the fourth cycle. Co-culture with pomalidomide increased stem cell memory and central memory T-cell fractions and enhanced sustained killing by 1.8-fold within 48 hours. Clinical: Between February 2023 and July 2025, 18 patients were enrolled and 16 received CAR-T infusion (13 multiple myeloma, 3 plasma cell leukemia). Median age was 63 years (range 42-75), 62.5% were male, 81% were >60 years, 68.75% had extramedullary disease, and 68% were triple-refractory to CD38 antibody. Median prior lines of therapy were 4 (3-10). Any-grade cytokine release syndrome occurred in 93% (grade ≥3 in 6.7%), and grade 1 immune effector cell-associated neurotoxicity syndrome in 13%; no dose-limiting toxicities were observed. MTD was established at DL2 (1 × 10⁶ cells/kg). With a median follow-up of 241 days (range 11-879), 14 patients were evaluable for response: ORR 100%, complete response 57% (8/14), and MRD negativity 93% (13/14). One-year PFS was 63.6% and one-year OS 69.1%. Patients receiving DL2 achieved the highest complete response rate (71%) and peak CAR transgene levels (median 22688 copies/µg gDNA). Among 5 patients with isolated medullary disease, all achieved complete response (CAR peak 30055 copies/µg gDNA), whereas among 9 with extramedullary disease, complete response was 44% (CAR peak 21494 copies/µg gDNA), suggesting that disease distribution influences efficacy. Conclusions In-vitro data demonstrate superior potency and pomalidomide-mediated memory differentiation of biparatopic BCMA CAR-T cells. In heavily pretreated RRMM patients, the combination was well tolerated, with MTD/RP2D defined as 1×10⁶ CAR-T cells/kg plus pomalidomide maintenance. High rates of deep, MRD-negative responses support advancement to phase 2 evaluation.