Fig 2: CAR T-cell pharmacokinetic and pharmacodynamics. A, CD4/CD8 ratio in the leukapheresis product used for CAR T-cell manufacturing in phase A and B subjects. Groups were compared using the Wilcoxon rank-sum test. B, Peripheral blood levels of CART-BCMA and huCART19 in the first 28 days after infusion determined by qPCR for vector sequences. C, Median levels of huCART19 (blue) and CART-BCMA (red) in phase A (solid line) and B (dashed line) subjects in the first 28 days after infusion. D, CART-BCMA area under the curve through day 28 (left) and CART-BCMA levels at months 2 and 3 after infusion (right) as determined by qPCR for vector sequences compared between subjects on the previously reported phase I study of CART-BCMA in relapsed/refractory MM (Prior Ph1) and the present study. Groups were compared using the Wilcoxon rank-sum test. E, CART-BCMA levels by qPCR for vector sequences in peripheral blood (PB) and bone marrow (BM) at postinfusion time points when paired samples were available. Lines connect paired measurements from the same subject. Paired PB and BM measurements were compared using the Wilcoxon sign-rank test. F, CART-BCMA and huCART19 levels by qPCR for vector sequences at 1, 2, and 3 months after infusion. Red lines highlight subjects with an increase in CAR T-cell levels between 2 and 3 months after infusion. G, Comparison of CD4/CD8 ratio in the leukapheresis product used for CAR T cell manufacturing in subjects who did (yes) and did not (no) exhibit an increase in CART-BCMA levels between months 2 and 3 after infusion. Groups were compared using the Wilcoxon rank-sum test. H, Soluble BCMA levels measured by ELISA at pre- and postinfusion time points. Shaded box indicates normal range for soluble BCMA in this assay based on mean ± 2*SEM from a sample of 19 normal donors. I, Peripheral blood frequency of CD19+ cells at pre- and posttreatment time points as determined by flow cytometry.

Fig 3: Cell-surface BCMA. A, MFI of multiple myeloma cells from bone marrow aspirate stained with phycoerythrin (PE)-conjugated anti-BCMA antibody in a multicolor flow cytometry panel to characterize multiple myeloma cells (see Methods), comparing patients at the pretreatment time point with relapsed/refractory multiple myeloma on our prior phase I study of CART-BCMA (left) and the current study (right); subjects without detectable multiple myeloma cells or nonevaluable bone marrow aspirates by flow cytometry are not depicted (N = 8 on prior phase I, N = 6 on the current study), and some subjects from the current study are depicted twice because bone marrow samples were analyzed both at enrollment (scr) and preinfusion (pre) time points in some subjects (N = 10). B, Example of lower pretreatment BCMA levels on the current study (top panels) compared with the prior study (bottom); multiple myeloma cells are identified by CD138 and CD38 expression (left) and confirmed by expression of other cell-surface markers cytoplasmic kappa or lambda light-chain restriction (see Methods). C, Subject-level comparisons of pretreatment BCMA MFI on multiple myeloma plasma cells from the prior phase I study and the current study from the negative control panel without the anti-BCMA antibody and the full panel; pink horizontal line indicates the median. D, Subject-level BCMA MFI on multiple myeloma cells divided by T cells as an internal negative control to adjust for batch effects; pink horizontal line indicates the median. E, BCMA intensity by IHC on pretreatment bone marrow core biopsies. F, Comparisons of pre- and posttreatment BCMA MFI on multiple myeloma cells in example patients with high pretreatment BCMA.

Fig 4: Late clinical response after maintenance initiation. Prestudy and on-study disease markers (pink) in subjects 16 (A) and 34 (B) who exhibited disease progression or stability in the first 1 to 2 months after infusion before responding upon initiation of maintenance therapy coincident with a rise in peripheral blood CART-BCMA levels (blue). C, Preprogression CART-BCMA levels in subjects 16 and 34 (pink) in comparison with other subjects with sufficient baseline detectable disease markers to track over time. D, Landmark analysis of TTP from 120 days after treatment in subjects with late response to maintenance therapy vs. other phase B subjects.

Fig 5: Immunohistochemical analysis of 92 Merkel cell carcinoma (MCC) samples reveals significant interindividual expression for numerous neuroendocrine/B‐cell‐related markers (BCMA, CD10, CD19, CD93, Chromogranin A, CK20, IgA, PAX5, TdT). (A) Overview of primary and metastatic MCC samples with positive marker expression shows abundant expression (> 50% samples) of Chromogranin A, followed by CK20, CD10, PAX5, BCMA and CD93, moderate expression (25–50% samples) of IgA and TdT, and modest expression (< 25% samples) of CD19. (B) Breakdown of individual marker expression (measured as optical density [OD]) by tumor type reveals significantly higher expression of CD93 and lower expression of Chromogranin A in primary versus metastatic samples (mean OD, Wilcoxon test, P < 0.05, n – number of tested samples); bars indicate standard error of the mean. Representative positive samples are shown for each marker from the primary (top) and metastatic (bottom) group (visualized at 15× magnification). (C) Co‐expression analysis of immunohistochemical markers from all MCC samples (left) reveals two distinct networks of moderate‐to‐strong correlation of B‐cell (BCMA, CD10 and CD93) and neuroendocrine with B‐cell (Chromogranin A, CK20 and PAX5) markers (Spearman's rho > 0.5). This trend was also observed in sub‐analyses of primary (top right) and metastatic (bottom right) samples.