Supplemental figure illustrating T, B and NK cell gating strategy

Supplemental table detailing statistics from the GSEA analysis comparing ritux and RV

Supplemental figure describing the proportion of peripheral monocyte cell subsets based on clinical response

Supplemental table detailing blood counts on cycle 1 day1 by lymphoma grade

Anti-programmed cell death protein1 (anti-PD-1) based immunotherapies improve survival in oesophageal adenocarcinoma (OAC) patients, but not all patients benefit. Inhibitors of the histone methyltransferase EZH2 (EZH2i) have been shown upregulate major histocompatibility complex (MHC) class I and II expression which overcame anti-PD-1 resistance in other cancers. This in vitro study assessed the efficacy of EZH2i and their impact on MHC expression in OAC. Three OAC cell lines (FLO1, MFD1, OE33) and a EZH2i-sensitive line (HS-SY-II) were cultured in 2D and treated with a EZH2i panel. Cell viability (via EC50 calculation) was assessed via MTS assay. Differential gene expression analysis followed by gene set enrichment analysis of RNA-sequencing data from the cell lines treated with EZH2i or DMSO control was performed in R. Flow cytometry of MHC Class I (HLA-ABC) and II (HLA-DR) was performed in parallel replicate experiments. EZH2i demonstrated similar efficacy in all 4 cell lines (EC50 > 5 µM). In the RNA-sequencing data, EZH2i was associated with gene set enrichment of interferon alpha and gamma signalling pathways (p<0.001). In FLO1 cells, HLA-ABC stain intensity was significantly increased (p=0.006) following EZH2i but not the proportion of cells expressing HLA-ABC (99.3% vs 99.0%). EZH2i increased in the proportion of cells expressing HLA-DR (18.3% vs 1.0%, p=0.001). Our data demonstrate the potential of EZH2i as a therapy to upregulate tumour cell MHC 1 and 2 expression which in turn may enhance anti-PD-1 immunotherapy response in OAC. Future work requires investigating EZH2i in combination with anti-PD-1 and chemotherapies.

Supplemental figure portraying the cell communication analysis in both pre- and on-treatment biopsies in PR versus NR.

Supplemental table detailing the top 50 DEGs and leading edge genes from scRNAseq and bulk RNAseq and the resulting overlapping genes

Supplemental detailing the top 20 ligand–receptor interactions

Supplemental figure describing the proportion of peripheral lymphoid cell subsets based on clinical response

Supplemental table detailing statistics from the GSEA analysis comparing progressive disease and partial response

Supplemental figure showing the correlation analysis between peripheral blood and immune subsets

Supplemental figure highlighting the approach in identifying γδ T cells

Supplemental table detailing expression of CD27 on individual immune subsets by clinical response

Supplemental figure describing the treatment schema within the RiVa trial

Supplementary table detailing characteristics of (overall) clinical responses to treatment

Supplemental figure showing changes in peripheral lymphoid populations throughout the trial

Supplemental figure showing the comparison of cell fractions between high and low grade disease

Supplemental figure illustrating regulatory T cell gating strategy

Supplemental table detailing participant representation in the RiVa trial

Supplemental figure showing the consort diagram