Findings may help doctors ‘customize’ glioblastoma treatment

Patterns in how genes express themselves may help doctors personalize treatment for glioblastoma, the most deadly of brain cancers, according to a new MD Anderson study.

The team, led by Erik Sulman, M.D., Ph.D., associate professor of Radiation Oncology, identified gene expression patterns that differed from surrounding immune cells, demonstrating the effects of chemotherapy and radiation treatments. They further defined tumor-intrinsic transcriptional subtypes designated as proneural, neural, mesenchymal and classical.

The findings were published online in Cancer Cell. 

“The intrinsic capacity of glioblastoma tumor cells to infiltrate the brain impedes surgical eradication and predictably results in high rates of early recurrence,” says Sulman. “To better understand determinants of glioblastoma tumor evolution and treatment resistance, we performed high dimensional profiling and molecular profiling of nearly 600 glioblastoma tumors via The Cancer Genome Atlas.”

Sulman and his team were able to identify common mutations in genes such as TP53, EGFR, IDH1 and PTEN as well as the frequent and concurrent presence of abnormalities in the p53, RB and receptor kinase pathways. Additional analysis revealed four clusters – proneural, mesenchymal, neural and classical – all highly associated with genomic abnormalities.

“Glioblastoma tumor cells, along with the tumor microenvironment, together create a complex milieu that ultimately promotes tumor cell transcriptomic adaptability and disease progression,” says Sulman. “Our study explored the properties of the microenvironment in different glioblastoma expression subtypes before and after therapeutic intervention. In doing so, we improved the robustness of gene expression subtype classification through revised gene signatures and analytical methodology."

The team’s results indicate that the tumor microenvironment affects expression-based classification of glioblastoma, both at the primary disease stage and at disease recurrence, implicating a role for macrophages and microglia in treatment response.

Longitudinal transcriptome analysis showed that expression subtype is retained in 55 percent of cases,” says Sulman. “Gene signature-based tumor microenvironment inference revealed a decrease in invading monocytes and subtype-dependent increase in macrophages/microglia cells upon disease recurrence.”

Glioblastoma expression subtypes have been related to genomic abnormalities treatment response and differences in tumor microenvironment. Sulman’s team defined tumor-intrinsic gene expression subtypes, which establish a role for the tumor immune environment in shaping the tumor cell transcriptome.

The finding could have implications for new approaches to more personalized therapy as currently all glioblastoma patients are treated similarly. The study could lay the ground work to enhance understanding of how to more effectively stratifying patients, a crucial step for precision medicine and more targeted, effective therapies.

The tumor microenvironment, consisting of cells surrounding a tumor is normally made up of immune cells, supporting cells and other normal cells. Tumors donated to tissue banks consist of a mixture of microenvironment cells and cancer cells. The team isolated the intrinsic gene expression pattern of 364 glioblastoma tumors, demonstrating the impact of the standard cancer treatment such as radiation and  temozolomide gene expression patterns on the remaining tumor and non-tumor cells, after negating the impact of therapy on the tumor-associated non-cancer cells.

The team was able to describe a greatly enhanced picture of this tumor ecosystem of hundreds of tumors, and what types of cells are in the microenvironment, their contributions, and how treatment affects this microenvironment as well as the tumor cells themselves.

Through this approach, the researchers also found that the molecular markers previously defining the neural subtype of glioblastoma was associated with the presence of normal neural tissue in the tumor margin, and were not characteristic of the actual tumor subtype.

Analysis also showed that the presence of macrophages correlated with poorer outcomes for glioblastoma who received radiation therapy. Also, tumors with a significant large amount of certain mutations have an increased number of positive CD8-positive T cells, suggesting they may respond to checkpoint inhibitor immunotherapy.

“Our study provides a comprehensive transcriptional and cellular landscape of IDH (isocitrate dehydrogenase) wild type glioblastoma during treatment modulated tumor evolution,” says Sulman.