Evaluating neuroblastoma and paraganglioma cancer heterogeneity and cell of origin through single cell transcriptomics

<p dir="ltr">Neuroblastoma is a cancer that arises in the adrenal medulla and sympathetic chain. Paraganglioma is a cancer that arises in the adrenal and extra-adrenal chromaffin cells. Neuroblastoma is a pediatric cancer with an average age at diagnosis of 18 months of age. Paraganglioma is most often present in adults, with an average age at diagnosis of 55 years of age. Neuroblastoma is highly metastatic, with almost 80% of high-risk cases having metastasize at diagnosis. Only 10 to 30% of paraganglioma metastasize.</p><p dir="ltr">In <b>paper I</b> we aimed to explore inter- and intra- tumor heterogeneity in human neuroblastoma. Particularly we investigated a cohort of high- and low-risk neuroblastoma samples to understand cancer cellular identities and their relationship with risk and age stratification. In <b>paper II</b> we aimed to characterize a postnatal progenitor-like population of cells in the healthy adrenal gland and its relationship with paraganglioma tumors. In <b>paper III</b> we aimed to explore inter- and intra- tumor heterogeneity in a cohort of neuroblastomas and paragangliomas. We looked to identify transcriptional similarities between these tumors and the healthy adrenal gland and between neuroblastoma and paraganglioma. In <b>paper IV</b> we aimed to examine transcriptional states, tumor microenvironment and epigenetic landscapes in neuroblastoma.</p><p dir="ltr">To this effect we used single cell/nuclei trancriptomics, bulk transcriptomics, spatial transcriptomics, and multiOmics. In addition, we performed stainings and lineage tracing.</p><p dir="ltr">In <b>paper I</b> we found that NB samples of different age-groups and risk stratification contained different contribution of cells. High-risk NB samples had a significantly higher proportion of progenitor-like undifferentiated cells, whereas low-risk samples consisted mostly of noradrenergic cells. We found that gene signatures for noradrenergic clusters are significantly upregulated in low-risk patients. In <b>paper II</b>, we identified 2 VHL-related paraganglioma tumors with Schawn-cell-like cells carrying an inferred 3p deletion. The same deletion was found in neuroendocrine cells from the same sample. However, the neuroendocrine cells carried additional copy number alterations. This suggests that in rare cases Schawn-cell-like stroma cells could potentially act as cell of origin of PPGL. In <b>paper III</b> we found a subpopulation of cells in malignant PPGL resembling cycling neuroblasts. This suggests neuroblasts may be influencing clinical progression in paraganglioma. We explore further this finding by running a deconvolution in a bulk transcriptomics dataset enriched with metastatic samples. We observed that metastatic samples had higher proportion of cycling neuroblasts when compared to non-metastatic samples. In <b>paper IV</b> we projected neuroblastoma cells onto a developmental trajectory reference. We identified 7 states that contributed to 2 classes, mesenchymal/neural crest, and sympathoadrenal committed. The 2 classes are linked by an intermediate bridge/connecting state. The bridge state was associated with high risk poorly differentiated neuroblastoma. We used MultiOmics sequencing to obtain information of gene expression and chromatin accessibility. We found that transcription factor E2F7 influences cell transitions and promotes NB malignancy through reinforcing the bridge state. We used spatial transcriptomics to deconvolve the developmental states of the reference into the tumor spatial spot. We quantified the distance between each tumor state-spot. We observed that bridge tumor spots were closer to cancer associated fibroblasts and tumor associated macrophages.</p><p dir="ltr">In conclusion, we explore cell states underlying a broad cohort of paraganglioma and neuroblastoma tumors and its relationship with the healthy adrenal gland. We identify progenitor-like population of cells that embryonically and postnatal could be influencing tumorigenesis in paraganglioma and neuroblastoma. Understanding the molecular mechanisms underlying PPGL and NB pathogenesis and progression has significant implications. The identification and validation of biomarkers is critical for diagnostic and targeted therapies.</p><h3>List of scientific papers</h3><p dir="ltr">I. Single-nuclei transcriptomes from human adrenal gland reveal distinct cellular identities of low and high-risk neuroblastoma tumors O. C. Bedoya-Reina*, W. Li*, M. <b>Arceo, M.</b> Plescher, P. Bulova, H. Pui, M. Kaucka, P. Kharchenko, T. Martinson, J. Holmberg, I. Adameyko, Q. Deng, C. Larsson, C. C. Juhlin, P. Kogner, S. Schlisio Nature communications (2021), 12(1), 5309. <a href="https://doi.org/10.1038/s41467-021-24870-7" rel="noreferrer" target="_blank">https://doi.org/10.1038/s41467-021-24870-7</a></p><p dir="ltr">II. Sustentacular glia-like cells as postnatal chromaffin progenitors and tumor cells of origin in a subset of VHL-related paragangliomas. P. Bulova*, P. Cui*, M. Arceo, J. Zhu, W. Li, M. Plescher, V. Poltorachenko, K. Stripling, C. Santangeli, M. E. Kastriti, C. Larsson, C. C. Juhlin, M. Mints and S. Schlisio [Manuscript preprint] <a href="https://doi.org/10.21203/rs.3.rs-6907400/v1" rel="noreferrer" target="_blank">https://doi.org/10.21203/rs.3.rs-6907400/v1</a></p><p dir="ltr">III. Cells resembling embryonic cycling neuroblasts as potential driver of malignant paraganglioma. M. Arceo*, J. Zhu*, W. Li, K. Stripling, V. Poltorachenko, K. S. Hose, P. Kogner, A. Stenman, C. Larsson, C. C. Juhlin, M. Mints and S. Schlisio [Manuscript]</p><p dir="ltr">IV. Single-cell MultiOmics and spatial transcriptomics demonstrate neuroblastoma developmental plasticity. Y. Zhu, D. Lou, P. Chen, G. Li, D. Usoskin, J. Pan, F. Li, S. Huang, C. Hess, R. Tang, X. Hu, J. Yu, M. Arceo, R. R. de Krijger, A. Tischler, S. Schlisio, P. Ernfors, Y. Hu, J. Wang Developmental Cell (2025) <a href="https://doi.org/10.1016/j.devcel.2025.04.013" rel="noreferrer" target="_blank">https://doi.org/10.1016/j.devcel.2025.04.013</a></p><p dir="ltr">(*) equal contributions</p>