Fig. 3: Higher spatial intra-tumour heterogeneity, proliferation and stemness in LFS medulloblastoma as compared to sporadic medulloblastoma. | Nature Communications

Fig. 3: Higher spatial intra-tumour heterogeneity, proliferation and stemness in LFS medulloblastoma as compared to sporadic medulloblastoma.

From: Spatio-temporal transcriptomics of chromothriptic SHH-medulloblastoma identifies multiple genetic clones that resist treatment and drive relapse

Fig. 3

A Spatial distribution of cancer drivers. Heterogeneous expression of oncogenes and essential microenvironment factors in LFS (n = 500 spots) and sporadic (n = 2723 spots) medulloblastoma. Left, representative examples; right, distribution of the fraction of spatial variance for the corresponding genes for LFS (n = 8) and sporadic (n = 5) tumours. Two-sided Mann-Whitney U-test was applied, non-significant after Benjamini-Hochberg multiple testing correction. B. Fractions of malignant cell types for each Visium spot in one representative LFS (n = 881 spots) and one sporadic (n = 2723 spots) medulloblastomas. Malignant cell clusters of SHH-high and cycling cells co-localize but neuronal development clusters are located in separate regions. C Cell-cell communication analysis, revealing Midkine signalling (MDK) as essential driver for cell-cell communication in LFS medulloblastoma. Left, overview of midkine signalling in a representative sample (n = 3983 spots). Colour indicates the extent of signal emitted by Visium spot in the upper plot and the received signal in the lower plot (from blue to red). Arrows indicate the overall direction of signalling. Right, average COMMOT cell-cell communication score for the leading 10 pathways. D UMAPs (n = 15,265 cells) based on matched 10x RNA-seq data (4 LFS medulloblastomas) with cell type annotation and expression of the MDK ligand. E COMMOT cell-cell communication scores, averaged across locations, for all medulloblastomas (LFS, n = 8; sporadic, n = 5), displaying the 10 leading pathways. n.s., two-sided Mann-Whitney U-test with Benjamini-Hochberg multiple testing correction. F Kaplan-Meier survival curves for SHH alpha subtype (n = 51, enriched for LFS medulloblastomas) grouped by MDK expression. High expression of MDK is associated with poor outcome in the SHH alpha medulloblastoma subtype (log-rank test, Kaplan-Meier plots generated using the R2 database, see Methods). G Fraction of the Malignant Cycling cells across spatial locations in LFS medulloblastomas (n = 8) as compared to sporadic medulloblastomas (n = 5). Horizontal lines denote the first, second, and third quartiles. A two-sided Mann-Whitney U-test was used. H Validation at protein level by immunofluorescence using Ki67 as a proliferation marker. Representative image of one sample is shown. Scale bar, 20 μm. I Quantification of Ki67-positive cells for all medulloblastomas (LFS, n = 8; sporadic, n = 5). Median is indicated. Statistical significance was calculated using a two-sided Mann-Whitney U-test.

Back to article page