Fig. 1: CD19 CAR-T cell expansion and B cell aplasia in patients with paediatric B-ALL with initial high or low tumour burden, and rationale for a DC-targeting CD19 CAR stimulating vaccine.
a, Scatterplot of peripherally circulating CAR-T cells in patients with paediatric B-ALL with initially high (≥40%, n = 40) or low tumour burden (<40%, n = 90) measured by qPCR over time during the first month after infusion. Solid lines show trends generated by locally estimated scatterplot smoothing. The median time to peak CAR-T expansion was 10 days after infusion in both cohorts. b,c, Maximum concentration (b) and area under the curve (AUC) (c) of CAR-transgene levels measured by qPCR, broken down by tumour burden. ***P < 0.001, by two-sided Wilcoxon rank-sum tests (n = 40 for high tumour burden, n = 90 for low tumour burden for b and c). d, Kaplan–Meier analysis of B cell aplasia for all patients (black) or patients with high (orange) or low (blue) tumour burden. B cell aplasia is defined as the time to the emergence of ≥1% CD19-positive B cell in bone marrow aspirate or ≥3% B cell by peripheral blood flow cytometry or CD19-positive relapse. Data were censored for patients who had CD19-negative relapse, reinfusion for hematogones in the marrow, alternative therapy including other CAR-T therapy or hematopoietic stem cell transplant, or non-relapse mortality. The probability of continued B cell aplasia at 2 years was similar in both cohorts (P = 0.76 by log-rank test). At 6 months, it was 72% (95% CI 55–95) in high and 67% (95% CI 57–78) in low tumour burden patients. e, Box plot of co-stimulatory molecule module scores expressed by B cells (likely healthy (n = 24) and likely malignant (n = 21)) and DC (n = 19). The box plots show median, 25th percentile and 75th percentile. The whiskers extend from each hinge to the most extreme value within 1.5 interquartile range (IQR). Data beyond this range are plotted individually. Each point represents the average of all cells of that phenotype within a patient, by a two-sided Wilcoxon rank-sum test. NS, not significant.
