Fig. 1: The identification and role of the immunoglobin-secreting malignant cells in the invasive frontiers of SCLC.

a Schematic workflow and grouping information for scRNA-seq using the 10× Chromium platform and spatial transcriptomics analysis. b Uniform Manifold Approximation and Projection (UMAP) plot of Louvain clustering (Scanpy) of jointly analyzed single-cell transcriptomes from 109,462 cells from SCLC specimens. c Distribution of canonical molecular markers for indicated subpopulations visualized in the dot plot. d Expression of selected Ig genes (IGKC, IGLC2, and IGHG3) and proliferation markers (TOP2A) among various cells of SCLC visualized in feature plot. e UMAP embedding of jointly analyzed scRNA-seq from malignant cells annotated by cell type. f Velocities derived from the dynamical model for various cells in SCLC specimens are visualized as streamlines in a UMAP-based embedding. g Representative image of multiplexed immunofluorescence staining (mIF) for ASCL1 (red), CD20 (orange), IGHA1 (brown), KI67 (yellow), and DAPI (blue) in SCLC patients (×200 magnification). h The expression pattern of IgG1, IgG2, IgG3, IgG4, free light chains κ, λ, and heavy chains γ in SCLC cell lines (DMS53, SW1271, NCI-H1339, NCI-H146, and NCI-H526). i Hemoxylin and eosin (H&E) staining and spatial feature plot (IGKC, IGLC2, and IGHG3) of the sections with invasive frontiers from SCLC patients. j Spatial trajectory inference analysis of cluster 1 spots using stlearn algorithm showing the intratumor heterogeneity of SCLC tumor cells with spatial information. k Multiplexed IF staining (ASCL1, CD68, IGHA1, LRP1, SLAMF7, and DAPI) showing co-aggregation of LRP1-high macrophages (LRP1⁺CD68⁺ cells) and SLAMF7-high tumor cells (SLAMF7⁺IGHA1⁺ cells) at invasive fronts of SCLC patients. l Spatial cell–cell interaction analysis of invasive frontiers of an SCLC patient using stlearn. Representative receptor and ligand pairs are visualized and zoomed. Progression-free survival (m) and overall survival (n) curves of 40 patients with SCLC from our validation cohort with bulk-RNA sequencing information, which were grouped by the expression of ASCL1 and IGHA1.