DUPAN-2 Enables Biomarker Monitoring in CA19-9 Nonexpressors : JAMA Surg | Jun 2026

Introduction:

Introduction: Carbohydrate antigen 19-9 (CA19-9) is the most widely used serum biomarker in pancreatic ductal adenocarcinoma (PDAC), guiding treatment response assessment, prognostication, and postoperative surveillance. However, approximately 5%–10% of patients are CA19-9 nonexpressors, typically due to a Lewis-negative phenotype, resulting in undetectable CA19-9 levels regardless of tumor burden. This leaves a clinically important subgroup without a reliable biomarker for treatment monitoring.

Problem Statement:

Problem Statement: There is currently no established alternative biomarker for patients with CA19-9 nonexpressing pancreatic cancer. Whether Duke Pancreatic Monoclonal Antigen Type 2 (DUPAN-2) can function as a surrogate biomarker and provide clinically meaningful treatment monitoring and prognostic information remains uncertain.

Summary:

Summary: This multicenter cohort study evaluated 2,418 patients with resected pancreatic ductal adenocarcinoma treated across nine Japanese academic centers. Among them, 185 patients (7.7%) were classified as CA19-9 nonexpressors (≤2 U/mL), while 2,233 were CA19-9 expressors.

Importantly, overall survival was comparable between CA19-9 nonexpressors and expressors, confirming that lack of CA19-9 expression does not inherently indicate a different biological prognosis but rather reflects a limitation in biomarker availability.

The investigators examined whether changes in DUPAN-2 levels among nonexpressors paralleled the well-established behavior of CA19-9 in expressors. Remarkably, DUPAN-2 dynamics closely mirrored CA19-9 responses during treatment. Reductions in DUPAN-2 after neoadjuvant therapy and after surgical resection were nearly identical to the decline patterns observed with CA19-9 in biomarker-expressing patients.

Most importantly, achieving normalization of DUPAN-2 levels (≤150 U/mL) after neoadjuvant therapy was strongly associated with improved overall survival and disease-free survival. Similar findings were observed after surgery, where patients achieving normal postoperative DUPAN-2 levels experienced substantially better long-term outcomes than those with persistently elevated values.

These observations suggest that DUPAN-2 may serve not merely as a diagnostic biomarker but as a dynamic treatment-response marker analogous to CA19-9.

From a clinical perspective, this study addresses a major unmet need in pancreatic oncology. Modern treatment strategies increasingly rely on biomarker kinetics to assess response to neoadjuvant therapy, determine surgical timing, identify residual disease risk, and guide postoperative surveillance. Until now, CA19-9 nonexpressors lacked an equivalent tool for biomarker-guided decision-making.

The study provides evidence that serial DUPAN-2 measurement may fill this gap, allowing clinicians to monitor treatment effectiveness and risk stratify patients throughout the disease course.

The findings are particularly relevant in the neoadjuvant era, where biochemical response has become an important adjunct to radiographic assessment when evaluating treatment success and surgical candidacy.

Although prospective validation is required before universal adoption, this large multicenter analysis establishes DUPAN-2 as the first clinically validated surrogate biomarker for CA19-9 nonexpressing pancreatic cancer.

Overall, the study suggests that routine DUPAN-2 monitoring can provide meaningful prognostic and treatment-response information for the previously unassessable subgroup of CA19-9 nonexpressor PDAC patients, enabling a more personalized and biomarker-driven management approach.