Duke Pancreatic Monoclonal Antigen Type 2 for Monitoring Carbohydrate Antigen 19-9 Nonexpressor Pancreatic Cancer
Menée au Japon à partir de données portant sur 2 418 patients atteints d'un cancer du pancréas n'exprimant pas l'antigène CA 19-9 (âge médian : 70 ans ; 58,1 % d'hommes), cette étude évalue la possibilité d'utiliser le niveau sérique de l'antigène DUPAN-2 pour prédire la survie sans maladie et la survie globale après un traitement néoadjuvant et une résection
Importance : Approximately 5% to 10% of patients with pancreatic cancer are carbohydrate antigen 19-9 (CA 19-9) nonexpressors (≤2 U/mL), predominantly due to the Lewis-negative phenotype, leaving them without established biomarkers for treatment monitoring despite the central role of biomarker assessment in modern pancreatic cancer management.
Objective : To evaluate Duke pancreatic monoclonal antigen type 2 (DUPAN-2) as a surrogate biomarker for monitoring treatment response in patients with CA 19-9 nonexpressor pancreatic cancer.
Design, Setting, and Participants : This cohort study included patients with resected pancreatic ductal adenocarcinoma from 9 Japanese academic centers between January 1, 2013, and December 31, 2019. Patients were classified as CA 19-9 nonexpressors (≤2 U/mL) or expressors (>2 U/mL). Median (IQR) follow-up was 55.3 (38.5-74.1) months for expressors and 51.1 (39.3-72.4) months for nonexpressors. Two analysis cohorts were defined among nonexpressors: those with postneoadjuvant DUPAN-2 measurements and those with postresection DUPAN-2 measurements. Data were analyzed from July 2024 to February 2026.
Exposures : DUPAN-2 levels after neoadjuvant therapy and after resection, with normal levels defined as ≤150 U/mL.
Main Outcomes and Measures : Overall survival (OS) and disease-free survival (DFS) calculated from the date of resection.
Results : Among 2418 patients (median [IQR] age, 70 [63-76] years; 1404 [58.1%] male), 185 (7.7%) were CA 19-9 nonexpressors and 2233 (92.3%) were expressors. Nonexpressors and expressors showed comparable OS (adjusted hazard ratio [HR], 1.10; 95% CI, 0.89-1.37). DUPAN-2 dynamics in nonexpressors mirrored CA 19-9 patterns in expressors, with comparable median (IQR) reductions (postneoadjuvant: −64.5% [−82.2 to −37.2] vs −67.1% [−89.0 to −25.9]; postresection: −71.4% [−84.4 to −48.8] vs −73.4% [−90.7 to −35.4]). Achieving normal DUPAN-2 levels (≤150 U/mL) after neoadjuvant therapy was associated with improved OS (adjusted HR, 0.26; 95% CI, 0.07-0.93) and DFS (adjusted HR, 0.15; 95% CI, 0.04-0.53). Similarly, normal postresection DUPAN-2 was associated with better OS (adjusted HR, 0.18; 95% CI, 0.06-0.55) and DFS (adjusted HR, 0.23; 95% CI, 0.08-0.64).
Conclusions and Relevance : In this cohort study, DUPAN-2 served as an effective surrogate biomarker for patients with CA 19-9 nonexpressor pancreatic cancer, with normal posttreatment levels associated with favorable outcomes. These findings suggest that routine DUPAN-2 monitoring may enable biomarker-guided treatment decisions for this previously unassessable subgroup.
JAMA Surgery , article en libre accès, 2026