Fig 2: Discovery of high affinity and specific VHHs against human BCMA. (a) A total of 2119 samples from MM patients were immunostained with antibodies targeting BCMA, CD19, and CD22. Each data point on the graph corresponds to an individual patient’s sample, and the analysis was conducted using the UCSC Xena platform. (b) Phage-display library screening strategy. The process includes DNA library generation, biopanning, and identification of specific binders through iterative selection and analysis. (c) A diagram of the panning procedure, six candidate VHHs were selected for further study. (d) Polyclonal analysis of dilutions of the phage pool from each round of biopanning by ELISA. (e) Monoclonal analysis of rescued phage from randomly selected clones from rounds 3 and 4 of biopanning by ELISA. (f) The purity of six candidate VHHs was evaluated using HPLC-SEC. (g-i) Flow cytometry analysis of the binding activity of six candidate VHHs to target cells with heterogeneous BCMA expression. (j) ELISA analysis of the binding activity of six candidate VHHs to BCMAECD-His antigen. (k-l) Flow cytometry analysis of the binding activity of six candidate VHHs to BCMA-negative cell lines CHO and HEK293T. Data were acquired using a BD Biosciences FACS flow cytometer and analyzed with FlowJo software

Fig 3: Characterization of VHH-based BCMA CAR-T cells. (a) Schema of the multi-round repeated antigen stimulation experiment. Tumor cells were seeded in 12-well plates, and CAR-T cells were added at an E:T ratio of 1:1. Every 24 hours, cells were collected, washed with PBS, resuspended in fresh X-VIVO medium, and counted using AO/PI reagent. Cells were then analyzed by flow cytometry, and culture supernatants were collected for cytokine measurement by ELISA. (b-d) Six VHH CAR-T and mock T cells were separately co-cultured with tumor cells over successive rounds. Residual tumor cells (b), CAR-T cell numbers (c), and CAR MFI (d) were quantified after each round. (e) Concentrations of IFN-γ and IL-2 in supernatants after co-culture with tumor cells were measured by ELISA. (f) Flow cytometry analysis and quantitative summary of activation markers CD137 (4-1BB), CD25, and CD69 on CAR-T cells on Day 7 and Day 12 of culture. (g) Flow cytometry analysis and quantitative summary of exhaustion markers PD-1, TIM-3, and LAG-3 on CAR-T cells on Day 7 and Day 12 of culture. (h) Representative dot plots show the phenotype of CD4+, CD8+, and CAR+ T cells. Cells were stained with CCR7 and CD45RA antibodies and assigned to subsets as stem cell memory (TSCM, CCR7+/CD45RA+), central memory (TCM, CCR7+/CD45RA-), effector memory (TEM, CCR7-/CD45RA-), and effector (TEFF, CCR7-/CD45RA+) cells. Gating boundaries were defined using CCR7-FMO and CD45RA-FMO controls. (i) Quantification of T cell phenotypes. Frequencies of stem cell memory, central memory, effector memory, and effector subsets were determined for CD4+, CD8+, and CAR+ T cells based on the representative flow cytometry plots in (h)

Fig 4: Enhanced in vitro activity of biparatopic Nab5822 CAR-T cells. (a) Schematic of recombinant lentiviral vectors encoding monospecific BCMA CAR, bispecific Nab2258 CAR, and Nab5822 CAR. (b) A schematic illustrating the potential mechanisms of action of monospecific BCMA CAR, bispecific Nab2258 CAR, and Nab5822 CAR-T cells against tumor cells. (c-d) Surface BCMA CAR expression on T cells 72 hours after lentiviral transduction was measured by flow cytometry, with representative plots shown in (c) and quantification across four donors in (d). (e) Total expansion Fold of CAR-T cells cultured in vitro with anti-CD3/CD28 stimulation at the indicated time points. (f) Percentages of CD4 and CD8 subtypes in CAR-T and mock T cells. (g-j) Cytotoxicity of CAR-T and mock T cells against MOLP8 (BCMALow) (g), MM.1S (BCMAMid) (h), NCIH929 (BCMAHigh) (i), and K562 (BCMANeg) (j) target cells at E:T ratios of 1:1, 1:3, and 1:9, shown in top panels without sBCMA and bottom panels with sBCMA. Data are presented as mean ± SD of triplicate wells. (k) Serum soluble BCMA levels in 10 MM patients were measured by ELISA, with a mean concentration of 12,000 pg/mL. (l-o) Representative heat maps illustrating cytokine production of CAR-T and mock T cells against MOLP8 (BCMALow) (l), MM.1S (BCMAMid) (m), NCIH929 (BCMAHigh) (n), and K562 (BCMANeg) (o) target cells at E:T ratios of 1:1, measured by CBA. Quantitative data are shown for the indicated cytokines in the absence or presence of sBCMA. (p) Degranulation of CD8+ T cells was assessed by surface CD107a expression using flow cytometry after 4 h of co-culture at an E:T ratio of 1:1. (q-r) Intracellular expression of perforin (q) and granzyme B (r) In CD8+ T cells was assessed by flow cytometry after 4 h of co-culture at an E:T ratio of 1:1. Statistical significance was determined by one-way ANOVA, followed by Dunn’s multiple comparisons test, or by an unpaired t-test; ns, not significant; *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

Fig 5: In vivo antitumor activity of Nab5822 CAR-T cells is preserved despite sBCMA. (a) Schematic of the evaluation of in vivo anti-BCMA+ tumor activity of Nab5822. NSG mice received 3.0 × 10^6 MOLP8-luc cells on day −10 and 2.0 × 10^6 cryopreserved Nab5822 CAR-T or mock T cells on day 0. Soluble BCMA (120 ng/mL) was administered on day 1, and tumor growth was monitored weekly. (b) IFN-γ concentrations were measured by ELISA on day 7 after infusion of Nab5822 CAR-T or mock T cells, in the absence or presence of sBCMA. (c) Total flux represents the tumor burden in MOLP8-luc bearing mice at the indicated time points (n = 3 biological replicates). (d) Bioluminescence was monitored during treatment. Mice marked with # were euthanized upon reaching the humane endpoint because tumor burden exceeded 1.0 × 10^10 photons/s. (e) in vivo toxicity was assessed by monitoring mouse body weight during treatment. Data are presented as mean ± SEM. (f) Kaplan–Meier survival curves of MOLP8-Luc bearing mice treated as described in (c, d). (g-i) Kinetics of CD3+ (g), CD4+ (h) and CD8+ (i) T cells in peripheral blood of mice after infusion of Nab5822 CAR-T or mock T cells. Each line represents an individual mouse. Statistical significance was determined by one-way ANOVA followed by Dunn’s multiple comparisons test, two-way ANOVA followed by Tukey’s multiple comparisons test, or log-rank test