Skip to main content
Neuro-Oncology logoLink to Neuro-Oncology
. 2017 Nov 6;19(Suppl 6):vi116. doi: 10.1093/neuonc/nox168.478

IMMU-20. CHARACTERIZING THE IMMUNE LANDSCAPE OF PD-1/PD-L1 IN BRAIN TUMOR IMMUNE CELL POPULATION USING CyTOF MASS CyTOMETRY

Alexander Lee 1,2, Tom Davidson 3, Mildred Galvez 4, David Nathanson 2, Anthony Wang 3, Linda Liau 1, Robert Prins 1,5
PMCID: PMC5693003

Abstract

INTRODUCTION

Immunotherapy may be a viable and efficacious treatment strategy for glioblastoma and other brain tumors. However, the negative co-stimulatory molecule PD-1 and its ligand PD-L1, which contributes to T cell inhibition in the tumor context, may limit effective anti-tumor immune responses. We have shown elevated levels of PD-1/PD-L1 expressing immune cells in the blood and tumor environments of brain tumor patients relative to healthy individuals. Additionally, our murine studies have shown PD-L1 expression on the tumor surface and on tumor infiltrating myeloid populations.

METHODS

To study in an unbiased manner which immune populations express PD-1/PD-L1 in the human brain tumor context and to discover the immune landscape in brain tumors, we analyzed tumor infiltrating immune cells and PBMCs from 25 brain tumor patients and PBMCs from 4 healthy donors using CyTOF mass cytometry and performed an unsupervised analysis using viSNE and SPADE to form groups of phenotypically similar cells.

RESULTS

We observed different profiles between brain tumor-infiltrating immune cell populations, patient PBMC, and healthy donor PBMC. PD-L1 expression was found in the myeloid cell population (87% ± 5.8% in tumor, 89% ± 3% in patient blood, 43% in healthy blood) and PD-1 was highly expressed in the tumor infiltrating immune population (CD4: 86% ± 11% in tumor, 46% ± 9.6% in patient blood, 11% in healthy blood; CD8: 
87% ± 10.6% in tumor, 52% ± 7.7% in patient blood, 16% in healthy donor). Additionally, we report differing PD-1/PD-L1 co-expression patterns with other immune markers between the patient’s tumor immune population and PBMC.

CONCLUSIONS

In conclusion, we characterized the immune landscape of immune populations existing in brain tumor patients and highlight which cell populations express PD-1/PD-L1. Uncovering these expression patterns in the tumor immune population may potentially be used to tailor or develop more precise and effective immunotherapies for patients.


Articles from Neuro-Oncology are provided here courtesy of Society for Neuro-Oncology and Oxford University Press

RESOURCES