Prognostic value of ctDNA monitoring in patients with resectable pancreatic ductal adenocarcinoma during surveillance.

Abstract

778 Background: Pancreatic ductal adenocarcinoma (PDAC) has a high recurrence rate, with up to 80% of patients experiencing disease recurrence within two years of surgical resection. Standard postoperative monitoring with imaging and serum tumor markers has significant limitations. Here, we evaluated the prognostic value of longitudinal circulating tumor (ct)DNA assessment in an updated cohort of patients with resected PDAC, representing a larger sample size and longer clinical follow-up during post-treatment surveillance. Methods: We retrospectively analyzed 292 banked plasma samples from 43 patients with localized PDAC who underwent surgical resection +/- neoadjuvant and/or adjuvant chemotherapy. Longitudinally collected blood samples during surveillance (from post-definitive treatment to the end of follow-up/recurrence) were used for ctDNA analysis using the personalized, tumor-informed 16-plex PCR-NGS assay (Signatera RUO, Natera, Inc.). The association of ctDNA status with recurrence-free survival (RFS) was evaluated using the Kaplan-Meier method, and comparisons were accomplished using log-rank tests. A multivariable Cox proportional hazards model was used to identify the most significant prognostic factor associated with RFS. Results: Among the 43 patients included in this analysis, 37% (16) had stage I, 30% (13) had stage II, and 33% (14) had stage III disease. The median patient age was 69 years (range: 51-86), and the median follow-up was 25 months (range: 1-52). The median plasma volume was 4.5 mL (range: 0.6-6.8), with ~half of the patient samples with <2 mL of plasma. 32.6% (14/43) received neoadjuvant chemotherapy (NAC) and 62.8% (27/43) received adjuvant chemotherapy (ACT). During follow-up, 72% (31/43) of patients relapsed, 68% (21/31) of whom tested ctDNA-positive at one or more time points post-surgery. In the surveillance window, 32% (10/31) of patients were positive for ctDNA, all of whom relapsed. ctDNA-positivity in the surveillance window was associated with significantly inferior RFS (HR: 16.1, 95% CI: 3.79-68.17, P=0.0002). Upon adjusting for other clinicopathological factors (stage and surgical margin), the multivariate analysis confirmed ctDNA-positivity during surveillance to be the most significant prognostic factor associated with RFS (HR: 8.4, 95%CI: 2.84–24.8, P<0.001). Conclusions: Despite the low sample quality in this cohort, ctDNA showed prognostic value, demonstrating the clinical utility of longitudinal ctDNA monitoring during surveillance.