GBM heterogeneity is primarily driven by the stochastic outcome of GBM stem cell hierarchical growth.

  • Major finding: GBM heterogeneity is primarily driven by the stochastic outcome of GBM stem cell hierarchical growth.

  • Approach: GBM stem cell clonal evolution was assessed by lentiviral barcoding of primary GBM.

  • Impact: GBM initiation may be associated with the reactivation of a normal developmental stem cell program.

Glioblastoma multiforme (GBM) is the most common adult brain malignancy and exhibits significant intratumoral heterogeneity. GBM stem cells (GSC), the radioresistant and chemoresistant CD133+ stem-like cell subpopulation of GBMs, has been shown to drive GBM initiation, maintenance, and post-treatment relapse, but the role of GSCs in GBM intratumoral heterogeneity has not been fully elucidated. To characterize the clonal dynamics of GSCs at different stages of malignant progression, Lan and colleagues performed serial orthotopic implantations of mice with primary and recurrent human GBM cells transduced with a lentiviral barcode library and treated the mice with either temozolomide or vehicle in the second and third passages of xenograft. Post-passage sequencing of untreated tumor cells present at and distal from the injection site showed that tumor growth is dependent on the tumorigenic capacity of many single-cell clones derived from GSCs and that although the growth behavior of individual clones is stochastic, each clone follows a predictable pattern. Evaluation of the distribution of relative clone size suggested that GBMs undergo nonexponential growth and harbor a small population of slow-cycling GSCs that eventually give rise to more rapidly cycling progenitor cells, which then undergo symmetric division to produce either progenitor cells or nondividing cells that rapidly undergo apoptosis. Consistent with these findings, stochastic simulations confirmed that GBM growth is dependent upon a defined GSC hierarchy. Comparison of the temozolomide-treated xenografts with untreated xenografts showed that although most of the cells were temozolomide-sensitive and exhibited nonexponential growth, temozolomide-resistant clones and a small subset of pretreatment clones did not exhibit nonexponential growth. In vitro drug screening identified epigenetic drugs that specifically targeted temozolomide-sensitive clones, such as an EZH2 inhibitor, or temozolomide-resistant clones, such as a Menin–MLL interaction inhibitor. These findings demonstrate how GBM stem cell growth dynamics determine GBM intratumoral heterogeneity and identify potential therapeutic epigenetic drugs that target specific GBM stem cell clonal populations.

Lan X, Jörg DJ, Cavalli FMG, Richards LM, Nguyen LV, Vanner RJ, et al. Fate mapping of human glioblastoma reveals an invariant stem cell hierarchy. Nature 2017;549:227–32.

Note:Research Watch is written by Cancer Discovery editorial staff. Readers are encouraged to consult the original articles for full details. For more Research Watch, visit Cancer Discovery online at