LAUSR

Background: Biliary tract cancer (BTC) refers to malignancies arising from the bile ducts, gallbladder, and ampulla of Vater. ctDNA-based MRD testing has shown promise in identifying surgical patients at high risk of recurrence, although there is a paucity of evidence on its utility in BTC. In this study, we prospectively evaluated the performance of Burning Rock Patient-specific pROgnostic and Potential tHErapeutic marker Tracking (brPROPHET), a patient-specific, tumor-informed MRD assay. Methods: Surgical specimens and two serial blood biopsies (before and approximately 4 weeks after surgery) were collected. The brPROPHET MRD assay (Burning Rock Biotech, China) uses whole exome sequencing for detecting patient-specific somatic alterations from tumor tissues and for blood a personalized panel designed to target up to 50 variants. For head-to-head comparison, the samples were also subjected to a tumor-informed, fixed-panel assay using 520- and 168-gene panels respectively for tumor and blood samples. Results: A total of 17 patients were analyzed. Most were men (11, 64.7%), had extrahepatic cholangiocarcinomas (13, 76.5%), or stage II disease (15, 88.2%). Median age was 64 years (range 57-68). Disease-free survival (DFS) was available for 14 patients at a median follow-up of 11.5 months (range 6.5-16.0). Baseline and postoperative ctDNA positivity rates were 64.7% (11/17) and 11.8% (2/17) with brPROPHET and 17.7% (3/17) and 5.8% (1/17) with the fixed-panel assay using 520- and 168-gene panels. Moreover, of the 3 patients who have relapsed, 1 was postoperatively ctDNA-positive by both assays and 1 only by brPROPHET. These findings suggested superior technical sensitivity of patient-specific over fixed-panel approach in detecting tumor-associated aberrations, and this advantage could translate into more sensitive identification of surgical BTC patients at high risk of recurrence. Furthermore, ctDNA positivity determined brPROPHET by was significantly associated with shorter DFS (log-rank p = 0.003), while that by fixed-panel was not (log-rank p = 0.113), suggesting a stronger association between the patient-specific approach with clinical outcomes. brPROPHET-derived postoperative ctDNA positivity also showed higher negative predictive value (91.7% vs 84.6%) and accuracy (92.9% vs 85.7%) in predicting recurrence. Together, these results supported a better performance of brPROPHET than the fixed panel approach for MRD detection. Conclusions: This prospective study demonstrated the clinical performance of brPROPHET, a patient-specific tumor-informed ctDNA-based MRD test for surgical patients with BTC. Compared with the fixed-panel approach using 520- and 168-gene panels respectively for tumor and blood samples, brPROPHET yields higher sensitivity, accuracy, and negative predictive value in predicting relapse. Citation Format: Tianqiang Song, Qiang Wu, Yunlong Cui, Huikai Li, Wei Zhang, Feng Fang, Qingqing Xiong. A patient-specific, tumor-informed, circulating tumor DNA (ctDNA)-based minimal residual disease (MRD) assay in surgical patients with biliary tract cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 6682.