Copied to clipboard!

Tumour DNA methylation subtypes and expression levels of CFISH (CXCL8, FCN1, IFI30, SLC11A1, HLA-DRA) predict response and overall survival to immunochemotherapy.

Background: Predicting patient response to immunochemotherapy is vital but challenging and poorly studied in the (neo)adjuvant setting. We showed that tumour monocyte content (TMC) predicts overall survival of advanced oesophageal cancer (OC) in response to immunochemotherapy. We report clinical outcomes of 35 operable OC patients receiving ICI-based therapy, and molecular profiling of the unique phase-II LUD2015-005 trial treating 73 OC patients (38 advanced/inoperable and 35 operable) in the (neo)adjuvant setting. Methods: LUD2015-005 trial patients were treated with four-week immune checkpoint inhibitor (ICI)-only before combination with standard-of-care. Paired and longitudinally collected biopsies of tumour and multiple adjacent normal sites were bi-dissected to extract DNA and RNA. Whole-genome, DNA-methylome, single-cell and bulk RNA-sequencing data generated from the entire LUD2015-005 cohort were used for cell-deconvolution and data integration analysis. Findings: Neoadjuvant ICI-based combination therapy of operable OC in the LUD2015-005 trial revealed improved overall outcomes. Tumour monocytes contribute to tumour immune microenvironment (TiME) remodelling and associate with tumour reactive CD8+ T cells upon ICI+ chemotherapy/chemoradiotherapy. We identified five DNA-methylation subtypes with distinct TiME and a TMC-related five-gene signature CFISH (CXCL8, FCN1, IFI30, SLC11A1, HLA-DRA) as highly predictive biomarkers of overall survival in OC and beyond. Conclusions: These findings highlight the importance of the myeloid compartment in ICI response and the potential of CFISH and tumour DNA methylation in patient stratification for immunochemotherapy. Funding: This work is mainly funded by Ludwig Institute for Cancer Research, CRUK Grants (A21998), (C30423/A2541), and NIHR Oxford Biomedical Research Centre. AstraZeneca UK Limited provided durvalumab and tremelimumab for LUD2015-005.

Type: Other
Archive: European Genome-phenome Archive (EGA)

4 Datasets

Click on a Dataset ID in the table below to learn more, and to find out who to contact about access to these data

Dataset ID	Description	Technology	Samples
EGAD00001016161	RNA-Seq data generated from normal tissue from patients from the operable cohorts of the LUD2015-005 Oesophageal cancer study (NCT02735239, EudraCT 2015-005298-19).	NextSeq 2000	10
EGAD00001016345	RNA-Seq data generated from tumour samples from patients from the operable cohorts of the LUD2015-005 Oesophageal cancer study (NCT02735239, EudraCT 2015-005298-19).	Illumina NovaSeq 6000	85
EGAD00001016793	Whole genome sequencing (WGS) generated from the operable cohorts of the LUD2015-005 study (NCT02735239, EudraCT 2015-005298-19). WGS data was generated with the aim to assess previously reported (e.g. tumour mutational burden) and novel predictive genomic markers for immunochemotherapy treatment in this setting. Using these data, mutation and copy number profiles were generated for the LUD2015-005 study, which were assessed for correlation with patient outcomes from this trial.	Illumina NovaSeq 6000 Illumina NovaSeq X	118
EGAD00001016815	single cell RNA sequencing from primary tumour biopsies of 16 patients from the operable cohorts of the LUD2015-005 oesophagleal cancer study (NCT02735239, EudraCT 2015-005298-19).	NextSeq 2000	1