Single-Cell Analysis Reveals Intratumoral Heterogeneity in Glioblastoma

Intratumoral heterogeneity occurs when cells within a single tumor differ according to mutation, phenotype, or epigenetic state, and can lead to complications in clinical diagnosis, treatment, and disease recurrence. Despite advances in targeted therapy, cancers with increased heterogeneity, such as glioblastoma, continue to display high mortality rates, reinforcing the need for individual tumor cell characterization in order to achieve better patient prognosis. To systematically analyze intratumoral heterogeneity, Patel, Tirosh, and colleagues used single-cell RNA sequencing to generate full-length transcriptomes for 430 individual tumor cells isolated from five freshly resected human glioblastomas. Estimation of copy-number variation from expression data in individual tumor cells revealed that tumors largely displayed homogenous chromosomal aberrations. However, in contrast to this large-scale analysis, extreme diversity in transcriptional profiles was observed across individual cells. Multiple receptor tyrosine kinases exhibited mosaic expression, and mutually exclusive expression of oncogenic EGFR variants was detected in individual cells. In addition, hierarchical clustering defined four meta-signatures enriched for genes involved in the cell cycle, hypoxia, and immune response that showed intratumoral variation. In contrast to in vitro models, only a fraction of individual cells expressed active cell-cycle genes, and intratumoral gradients were observed for the hypoxic and stem-cell signatures, reinforcing the notion of in vivo complexity. Of note, a continuous spectrum of stemness and differentiation states was identified across tumor cells and was correlated with the expression of candidate regulatory transcription factors. Furthermore, although the current glioblastoma subtype classification scheme detected dominant subtypes via bulk tumor analysis, scoring individual cells based on this scheme revealed a mixed-subtype population in each tumor. Using this in turn to infer heterogeneity in bulk tumors showed that increased heterogeneity was associated with decreased survival. Together, these findings highlight intratumoral transcriptional variability in glioblastoma and emphasize the need to account for tumor heterogeneity in future therapeutic design.

Patel AP, Tirosh I, Trombetta JJ, Shalek AK, Gillespie SM, Wakimoto H, et al. Single-cell RNA-seq highlights intratumoral heterogeneity in primary glioblastoma. Science 2014;344:1396–401.

Note: Research Watch is written by Cancer Discovery Science Writers. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at http://CDnews.aacrjournals.org.