Relapse after chemotherapy and surgery occurs in >50% of resected EAC patients (pts). ctDNA following resection is prognostic in multiple cancers, however false positive results due to clonal haematopoiesis of indeterminate potential (CHIP) may limit the accuracy of tumour-naive ctDNA panels. We investigated the prognostic value post-surgical ctDNA in resected EAC and optimised a pipeline to reduce the effect of CHIP.
Pts were identified from the prospective national UK OCCAMS consortium dataset. A 77 gene tumour naïve ctDNA panel was deployed. Plasma samples were sequenced to a mean depth of 7,062x (range: 2,196 – 28,524). Pipeline optimisation included: removal of variants which were synonymous, >0.05% frequency across germline databases; those which fall in untranslated or upstream gene regions and variants at >5% VAF if always confirmed as germline in COSMIC. Post-op ctDNA +ve is defined as at least 2 variants detected in the same sample. Overall survival (OS) and disease-free survival (DFS) was estimated using the Kaplan-Meier method and compared using the log-rank test.
97 pts were identified; 83/97 (86%) male, median age 68 years (SD 9.5 years), 100% cT3/T4, 75% cN+. 96/97 (99.0%) had neoadjuvant chemotherapy. 78/97 (80.4%) had CHIP analyses; 18/78 (23.1%) had CHIP variants removed. Using stringent quality criteria 16/79 (20.3%) were ctDNA +ve post-resection; recurrence was observed in 12/16 (75.0%) of these. Median OS for ctDNA +ve pts 14.9 months (m) vs 29.5m for ctDNA -ve (HR 2.32, (95% CI 1.14- 4.73, p = 0.03). When CHIP was excluded 10/63 (16.9%) pts were ctDNA +ve and 9/10 of these (90.0%) recurred. With correction for CHIP median OS ctDNA +ve pts was 10.1m vs 29.5m ctDNA -ve. (HR 5.55, (95% CI 2.42- 12.71, p = 0.0003). One pt with short follow-up due to recent treatment (30 weeks) who was ctDNA +ve has not yet relapsed. Similar outcomes were observed for DFS.
We demonstrate in a large, national, prospectively collected dataset that the ctDNA in plasma following surgery for EAC is prognostic for relapse. Pipeline optimisation can reduce or eliminate false positives from CHIP. In future, post-operative ctDNA could be used to risk stratify patients into high and low risk groups for intensification or de-escalation of adjuvant chemotherapy.
E. Ococks: Travel/Accommodation/Expenses: Roche. R. Fitzgerald: Advisory/Consultancy, Shareholder/Stockholder/Stock options: Cyted Ltd.; Licensing/Royalties: Cytosponge; Research grant/Funding (self): Roche; Research grant/Funding (self): AstraZeneca. E. Smyth: Honoraria (self): Zymeworks; Honoraria (self): Astellas; Honoraria (self): AstraZeneca; Honoraria (self): Bristol-Myers Squibb; Honoraria (self): Merck; Honoraria (self): Celgene; Honoraria (self): Five Prime; Honoraria (self): Gritstone Oncology; Honoraria (self): Servier. All other authors have declared no conflicts of interest.