Next generation circulating tumour DNA detection and subclonality analysis for gastrointestinal cancer therapy personalisation

Abstract

Background: Gastrointestinal cancers are associated with poor outcomes as many patients present with advanced disease. Intratumour heterogeneity (ITH) is a major barrier to predicting the evolution of cancers and successful treatment. ITH is represented through somatic mutations and copy number aberrations (CNAs). These can be detected through genome profiling of tumour biopsies, however this can be subject to sampling bias. Minimally invasive circulating free (cf)DNA analysis can portray genomic landscapes but highly sensitive approaches are necessary to detect mutations with low variant frequencies. Methods: I have developed a fully customisable error-corrected targeted sequencing assay for low cfDNA input amounts to interrogate the genomic landscape of patients with metastatic colorectal cancer (mCRC). Results: Through optimising the target enrichment technology with molecular barcode and novel duplex DNA-molecule identification tool for error correction, I have established a sensitive cfDNA-seq technology with a mutation detection limit <0.15%, which is necessary for the reliable detection of evolving drug resistant subclones, I applied this assay to mCRC patients to detect driver mutations and CNAs and investigate the subclonal genomic landscape that may drive therapy resistance. I have also designed a translational clinical trial (ICONIC) to improve treatment approaches in localised gastro-oesophageal adenocarcinomas (GOAs). The trial utilises a novel perioperative chemo-immunotherapy combination with the aim of generating synergies by enhancing progression through the cancer-immunity cycle via the pro-immunogenic effect of docetaxel, oxiliplatin, leucovorin and fluorouracil (FLOT) chemotherapy and simultaneous avelumab-induced release of the candidate predictive biomarkers of tumour response through longitudinal blood (ctDNA) and tumour sampling. The translational study will also investigate dynamic changes ctDNA load and mutant allele frequency, changes in immune inflitrates in baseline and on-treatment biopsies and correlate neoepitope load and faecal microbiome with tumour response. Conclusions: Overall, the error-corrected cfDNA-seq assay with a customisable target region enables broad insights into cancer genomes and evolution.