PTU-128 Novel insights into metastatic spread of oesophageal adenocarcinoma from whole genome sequencing data. (22nd June 2015)

• Record Type: Journal Article
• Authors: Noorani, A; Wedge, D; Lao-Sirieix, P; Weaver, J; Grehan, N; Debiram, I; Gilligan, D; Hindmarsh, A; Goddard, M; Hardwick, R; Fitzgerald, R
• Is Part Of: Gut. Volume 64(2015)Supplement 1
• Page Start: A119
• Page End: A119
• Publication Date: 2015-06-22
• Subjects: Gastroenterology -- Periodicals; 616.33
• DOI: 10.1136/gutjnl-2015-309861.243
• Languages: English
• ISSNs: 0017-5749

Abstract

Abstract : Introduction: The evolution of metastatic oesophageal adenocarcinoma (OAC) is poorly understood. Whole genome sequencing (WGS) across multiple metastatic sites provides the unique opportunity to study clonal evolution of OAC. Method: 2 cases of junctional OAC were extensively sampled at autopsy within 4 h of death. At presentation case 1 was palliative, case 2 had potentially curative disease but progressed during treatment. 20 fresh frozen tumour samples from primary tumour and multiple metastases (adrenal, liver, pancreas, station 102, left gastric and para-aortic lymph nodes) underwent WGS. Clonal analysis involved estimating copy number status and tumour purity, followed by point mutation clustering based on allele frequency (proportion of tumour cells harbouring a mutation); clusters were referred to as clones and were annotated with driver events of clonal progression (single nucleotide variants (SNVs) and copy number events). Results: 11 clones spanned all samples in case 1 compared to 5 in case 2. Median SNVs per clone were 5885 in case 1 (range 1091–29883) and 2149 (range 424–27031) in case 2. The founding clone in both cases contained the bulk of the SNVs and key potential drivers of carcinogenesis (case 1:KRAS and GATA4 amplification, case 2:TP53 mutation). Metastases in case 1 had a branching pattern of evolution; 6 clones were unique to individual metastases and only 1 was shared across sites. Clones were on average 9363 (+/-5827) SNVs distant from the founder clone demonstrating continuing evolution in metastases. Para-aortic nodal metastasis preceded local lymph node metastasis. In contrast case 2 demonstrated linear evolution with just 2 metastatic clones that were on average 1286 (+/-1219) SNVs distinct from the primary tumour; 1 was only in the adrenal, the second was shared across multiple sites suggestive of an embolic pattern of tumour seeding. The liver metastasis was identical to the primary founder clone demonstrating that metastasis to the liver was the initial event, even prior to local lymph node metastasis. Conclusion: Metastases can result from contrasting patterns of clonal evolution. Both cases demonstrate the striking phenomenon of early distant metastases, prior to the development of local lymph node metastases thus challenging our traditional algorithms of tumour staging and prognostication. Additional cases are being analysed to support our findings. Disclosure of interest: None Declared.