Tumor-informed ctDNA monitoring during surveillance for early detection of recurrence in patients with stage II/III esophageal cancer treated with chemoradiation.

Abstract

440 Background: The recurrence rate in patients with esophageal cancer (EC) is high, with up to 50% recurring within 2 years of curative treatment. Conventional surveillance imaging is often confounded by post-treatment inflammation and scarring, leading to delayed detection. Circulating tumor DNA (ctDNA) has emerged as a minimally invasive and highly sensitive biomarker for the early detection of recurrence. This study evaluated the role of longitudinal ctDNA monitoring in patients with EC during post-treatment surveillance. Methods: A single institution prospective observational study of ctDNA testing on longitudinal plasma samples collected pre-, on-, and post-treatment from patients with EC at the provider’s discretion (N=31). ctDNA analysis was performed using a clinically validated personalized, tumor-informed 16-plex PCR-NGS assay (Signatera, Natera, Inc.). ctDNA positivity rates and association of ctDNA status in the surveillance window (12 weeks post-end of definitive treatment or 2 weeks post-end of additional treatment until recurrence/death from any cause/end of follow-up) with event-free survival (EFS) was evaluated. Results: Patients (N=14) with >1 ctDNA result and >12 weeks of clinical follow-up available, post-definitive treatment were included. The median patient age was 71 years (range: 46-81 years). Adenocarcinoma was reported in 93% (13/14), and squamous cell carcinoma was reported in 7% (1/14) of patients. The majority of patients (11/14; 79%) had stage III disease. Median follow-up was 6.2 months (range: 1.9-27.6). In this cohort, 64% (9/14) of patients received definitive CRT, and 35% (5/14) received neoadjuvant CRT and surgery. ctDNA-positivity rate was 100% (11/11) at pre-treatment and 70% (7/10) in the surveillance window. Longitudinal analysis in the surveillance window [median 2 (range: 1-5) tests per patient] demonstrated that ctDNA-positive patients had significantly inferior recurrence-free survival (HR: 28.3, 95% CI: 3.1 - 3767.7, p=0.001). Recurrence/death event was observed in 85% (6/7) of patients with ctDNA-positivity during surveillance, whereas 100% of the serially ctDNA-negative patients (N=3) remained recurrence-free, one of whom died from a non-cancer cause. ctDNA-positivity at any time post-definitive treatment preceded clinical recurrence by a median of 4.1 months (0.7-7.3 months), with some patients undergoing earlier imaging as a result of a positive ctDNA test. Conclusions: Longitudinal, tumor-informed ctDNA monitoring is a valuable tool for surveillance, detecting recurrence ahead of radiographic imaging. Further prospective studies can provide valuable insights into the clinical utility of ctDNA monitoring to guide possible watchful waiting post chemoradiation.