Re-convolving the compositional landscape of primary and recurrent glioblastoma reveals prognostic and targetable tissue states

Abstract

Glioblastoma is an aggressive diffusely infiltrating neoplasm that spreads beyond surgical resection margins, where it intermingles with non-neoplastic brain cells. This complex tissue harboring infiltrating glioma and non-neoplastic brain cells is the origin of tumor recurrence. Thus, understanding the cellular and molecular features of the glioma margin is therapeutically and prognostically important. Here, we used single-nucleus RNA sequencing (snRNAseq) of primary and recurrent glioma to define compositional tissue-states that correlate with radiographic and histopathologic features. We found that glioma cells can be clustered into proliferative, astrocyte-like/mesenchymal, and progenitor-like/proneural states in both the primary and post-treatment recurrence settings. We focused on non-neoplastic microenvironment cells including oligodendrocytes, myeloid cells, neurons, and astrocytes - the latter two are under-represented in single-cell RNAseq studies. Cell type-specific signatures of the astrocyte-like/mesenchymal glioma, and a subpopulation of non-neoplastic astrocytes correlated with poor prognosis, the latter correlated with glioma recurrence. Notably, astrocytes were enriched for metabolic and neurodegenerative gene signatures. Leveraging snRNAseq-derived compositional information, we define three tissue-states that correlate with radiographic localization of primary and recurrent glioma. Our findings define a compositional approach to the glioma microenvironment and reveal prognostically and anatomically relevant features paving the way to new mechanistic and therapeutic discoveries.

Glioblastoma states are defined by cohabitating cellular populations with progression-, imaging- and sex-distinct patterns

“…To interpret how tumor composition contributes to this trajectory, each image-localized sample underwent RNA sequencing deconvolution to estimate the abundance of the individual cell populations (Figure S2A; Table S2). Eighteen cell populations, as defined by snRNAseq (20), were predicted: six malignant glioma, five immune (e.g., myeloid cells, T-cells), and seven “other” (e.g., astrocytes, neurons, endothelial cells). The sums of malignant cells, immune cells, and other cells are highest in states C, B, and A, respectively (Figure 2D).…”

“…Recent methodological advancements, (i.e., single nucleus RNA sequencing) circumvent these limitations and allow for a more complete characterization of all populations including malignant, immune, and other cells. We leveraged a unique cohort of GBM snRNAseq as training data for a cellular deconvolution algorithm (20). This paradigm allowed us to quantify all GBM cell subpopulations in a large number of bulk RNA sequencing samples, which has never been done before.…”

“…Single nucleus RNAseq (snRNAseq) circumvents this limitation by more comprehensively sampling malignant, immune, and other cells. We leveraged a snRNAseq data set (20), derived from seven patients with GBM, which contained eighteen cell types: six malignant, five immune, and seven “other.” Malignant cell populations included mesenchymal/immunoreactive (gl_Mes1 and gl_Mes2), proliferative (gl_Pro1 and gl_Pro2), and neural/oligodendrocyte precursors (gl_PN1 and gl_PN2). Immune cells were characterized as T-cells, myeloid cells present at baseline (Myel1), proliferative tumor-associated macrophages (prTAM), monocyte-derived cells (moTAM), and microglia-derived cells (mgTAM).…”

“…This method relies on marker gene input and assumes that expression signals are linearly additive, such that marker gene expression is related to the proportion of that cell type in the sample. Each cell in our snRNAseq matrix was labeled as one of eighteen cell populations (20). This served as the input to CIBERSORTx, where a signature matrix was derived to define the unique features of each cell type.…”

Biologically-informed deep neural networks provide quantitative assessment of intratumoral heterogeneity in post-treatment glioblastoma

“…Two independent patient samples were used to develop (cohort A) and test (cohort B) the model. Cohort A was acquired as part of a retrospective, observational study designed to study patients who had undergone repeat surgical resection for recurrence of high-grade gliomas after treatment with chemotherapy and radiation therapy (49). Cohort B was acquired as part of a prospective, clinical trial for convection enhanced delivery of topotecan chemotherapy (50).…”