Abstract

Background: Chimeric antigen receptor-T (CAR-T) cell therapy has transformed B-cell acute lymphoblastic leukemia (B-ALL) treatment, yielding outstanding clinical results. However, less than half of patients achieve long-term remission, with treatment-associated adverse events a clinical challenge. Aims: This study aims to investigate the transcriptomic characterizations of CAR-T products and bone marrow mononuclear cells, and identify predictive biomarkers for efficacy and toxicity. Methods: Seventeen B-ALL patients were enrolled in this study. Single-cell RNA sequencing was performed on BMMCs (baseline and post-infusion) and IPs in 4 patients, and 120,780 residual cells were collected for bioinformatic analysis to identify biomarkers for the early clinical forecast. Another 13 patients-cohort was used for bulk RNA-sequencing validation. Results: An overview of our approach for sc-RNA-seq is shown in Figure A. Twenty-one distinctly transcriptional cell clusters were identified after extensive, unbiased clustering of all BMMCs and IPs (Figure B). Cell clusters were classified into five broad cell types -- IPs, HSPCs, NK/T cells, myeloid cells, and B-ALL cells (Figure C). For severe CRS prediction, 4 patients were divided into two groups based on tumor burden. Baseline cytokine-mediated signaling increased in high-tumor burden patients, mainly in B-ALL blasts, HSPCs (Pt 10 vs Pt 8); and low tumor burden patients, primarily in NK/T and Myeloid cells (Pt 11 vs Pt 9) (Figure D). IFITM1, a gene in IFN-related signaling pathways, was significantly upregulated in the above-mentioned cell types and validated in the bulk-RNA cohort (Figure E). Unlike complete response patients, NR (Pt 8) patients had diminished SRGN expression in every BMMC-1 cell type (Figure F), with probable arrests of IPs at the G1 phase (Figure G). In the NR patient tumor subsets, treatment-resistant tumor clusters (C1) in the IFN-I signaling pathway were enriched, whereas treatment-sensitive tumor cells (C3/5/10) underwent apoptosis (Figure H/I/J). Other ways, including the extracellular structure organization that predicts extramedullary relapse, were amplified (Figure K). The bulk RNA-sequencing cohort produced identical results. Image:Summary/Conclusion: IP and BMMC transcriptional feature heterogeneity possibly contributes to diverse clinical outcomes post-anti-CD19 CAR-T cell therapy and could be predictive.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call