Epigenetic biomarkers of T-cells in human glioma

Abstract

Immune factors are thought to influence glioma risk and outcomes, but immune profiling studies to further our understanding of the immune response are limited by current immunodiagnostic methods. We developed a new assay to capture glioma immune biology based on quantitative methylation specific PCR (qMSP) of two T-cell genes (CD3Z: T-cells, and FOXP3: Tregs). Flow cytometry of T-cells correlated well with the CD3Z demethylation assay (r = 0.93; p < 2.2 × 10⁻¹⁶), demonstrating the validity of the assay. Furthermore, there was a high correlation between qMSP and immunohistochemistry (IHC) in quantifying tumor infiltrating T-cells (r = 0.85; p = 3.4 × 10⁻¹¹). Applying our qMSP methods to archival whole blood from 65 glioblastoma multiforme (GBM) cases and 94 non-diseased controls, GBM cases had highly statistically significantly lower T-cells (p = 1.7 × 10⁻⁹) as well as Tregs (p = 5.2 × 10⁻¹¹) and a modestly lower ratio of Tregs/T-cells (p = 0.024). Applying the methods to 120 excised glioma tumors, we observed that tumor infiltrating CD3+ T-cells were positively correlated with glioma tumor grade (p = 5.7 × 10⁻⁷), and that Tregs were enriched in tumors compared with peripheral blood indicating active chemoattraction of suppressive Tregs into the tumor compartment. Poorer patient survival was correlated with higher levels of tumor infiltrating T-cells (p = 0.01) and Tregs (p = 0.04). DNA methylation based immunodiagnostics represent a new generation of powerful laboratory tools offering many advantages over conventional methods that will facilitate large clinical epidemiologic studies and capitalize on stored archival blood and tissue banks.

Keywords: :

• DNA methylation
• glioma
• Tregs
• T-cells
• biomarkers

Disclosure of Potential Conflicts of Interest

The authors have no potential conflicts of interest to report.

Acknowledgments

The authors wish to acknowledge the contributions of Yuanyuan Xiao for comparing CD3Z expression levels among expression classes of GBMs within the TCGA database and Cynthia Caudrey, Tarik Tihan for providing tumor specimens and pathology review, Terri Rice, Lucie McCoy for data collection and analysis, and Katie Wegenman for sample processing.

Financial Disclosures

This work was supported by grants from the National Institutes of Health, R01CA52689, CA126939, CA100679, University of California San Francisco Brain Tumor Specialized Program Of Research Excellence P50CA097257, K08 NS063456, and the National Institute of Environmental Health Sciences ES06717.

Supplemental Materials

Supplemental materials may be found here: www.landesbioscience.com/journals/epigenetics/article/22675