Spatial analysis reveals distinct immune phenotypes and tertiary lymphoid structure-like aggregates in pediatric acute myeloid leukemia [Nanostring PanCancer IO 360]

Comparison of immune-infiltrated bone marrow samples (n=6) with immune-depleted bone marrow samples (n=17) of pediatric acute myeloid leukemia patients using the Nanostring PanCancer IO 360 panel. 6 immune-infiltrated and 17 immune-depleted FFPE bone biopsies (the non-leukemic controls’ median value of cytotoxic T cells was used as the cut-off value) from a cytogenetically representative cohort of pediatric AML patients were used for this experiment. After passing quality control, raw data were normalized using ROSALIND® according to Nanostring’s recommendations (https://rosalind.bio/; San Diego, CA, USA). Normalized data were uploaded to the online iDEP platform and differentially expressed genes between immune-infiltrated and immune-depleted samples were identified using DEseq2 with a false-discovery rate (FDR) cut-off of 0.05 and a minimum fold change of 2 (integrative Differential Expression and Pathway analysis; http://bioinformatics.sdstate.edu/idep96/; V.0.96; Ge, Son & Yao, 2018). Pathway analysis was performed using parametric gene set enrichment analysis with the GO Biological Processes gene set with an FDR of 0.05 (Ashburner et al., 2000; The Gene Ontology Consortium, 2020). The abundance of M2-like macrophages was estimated using a published gene signature of M2-like macrophages compared to M1-like macrophages, inside the TIDE environment (tide.dfci.harvard.edu/; Beyer et al., 2020; Jiang et al., 2018). Similarly, the T cell inflamed gene expression profile, the CD8-, IFN-γ-, and TIDE scores were all calculated using TIDE. >>>Submitter states: Patients did not provide consent for open access of raw data. Therefore, raw data are made available in a controlled manner via the EGA platform (https://ega-archive.org/). <<<