IMMU-20. SINGLE-CELL RNASEQ OF TUMOR INFILTRATING IMMUNE CELLS FROM NEOADJUVANT ANTI-PD1 TREATED GBM PATIENTS REVEALS GLOBAL TRANSCRIPTIONAL CHANGES AND IMMUNOSUPPRESSIVE ADAPTIVE RESPONSES BY MYELOID CELLS. (9th November 2020)

• Record Type: Journal Article
• Authors: Lee, Alexander; Mochizuki, Aaron; Chow, Frances; Reynoso, Jeremy; Orpilla, Joey; Sun, Lu; Chavez, Carolina; Galvez, Mildred; Kienzler, Jenny; Yong, William; Everson, Richard; Nathanson, David; Bensinger, Steven; Liau, Linda; Cloughesy, Timothy; Hugo, Willy; Prins, Robert
• Is Part Of: Neuro-oncology. Volume 22(2020)Supplement 2
• Page Start: ii108
• Page End: ii109
• Publication Date: 2020-11-09
• Subjects: Brain Neoplasms -- Periodicals; Brain -- Tumors -- Periodicals; Brain -- Cancer -- Periodicals; Nervous system -- Cancer -- Periodicals; 616.99481
• DOI: 10.1093/neuonc/noaa215.450
• Languages: English
• ISSNs: 1522-8517

Abstract

Abstract: INTRODUCTION: Neoadjuvant anti-PD1 therapy (neo-aPD1) was previously shown to significantly increase the survival of recurrent glioblastoma patients in a small randomized clinical trial. However, neo-aPD1 alone was not curative so defining the limitations of neo-aPD1 and discovering where other immunotherapies can be used alongside neo-aPD1 is needed. METHODS: To understand how immune cells in the tumor microenvironment change with neo-aPD1, we used single-cell RNAsequencing to analyze cells from 27 glioma patients (n = 105, 143 cells) of which 9 patients had received neo-aPD1 (n = 33, 325 cells). Using unsupervised clustering and pseudotime trajectory analysis, we characterized the transcriptional changes within immune cells and how these populations changed with therapy. RESULTS: We defined the immune landscape of the glioblastoma tumor microenvironment. Compared to no immunotherapy treatment, neo-aPD1 significantly increased the ratio of T cells to myelo-monocytic cells and led to significant increases in the effector and memory T cell populations but no significant changes in myeloid cell composition. Our differential gene expression analysis of the myeloid compartment showed significant increases in interferon-γ-responsive genes and down-regulation of genes associated with M2 macrophages and MDSCs, suggestive that neo-aPD1 influences the transcriptional profile of myeloid cells in the tumor microenvironment. Interestingly, our psuedotime trajectory analysis showed that neo-aPD1 was associated with cells expressing both lymphoid and myeloid-related genes, which we theorized to actually be lymphoid-myeloid cell doublets caused by increased interactions between myeloid and lymphoid cells. These doublets were highly enriched in MHC I and II, macrophage, T cell, and T cell activation and exhaustion genes indicating that neo-aPD1 may result in some adaptive immunosuppressive mechanism by increasing these interactions. This could explain why neo-aPD1 alone is not curative for glioblastoma patients. CONCLUSIONS: In total, neoadjuvant anti-PD1 therapy enhances effector T cell activity, but may concomitantly induce adaptive resistance mediated by myeloid cells in glioblastoma.