Johns Hopkins Radiation Oncology Kimmel Cancer Center Baltimore, MD
S. Mao1, S. B. Oruganti2, T. A. Lin1, S. Sehgal1, A. V. Reddy3, and A. Narang1; 1Department of Radiation Oncology and Molecular Radiation Sciences, Johns Hopkins University School of Medicine, Baltimore, MD, 2Sidney Kimmel Comprehensive Cancer Center, Johns Hopkins University, Baltimore, MD, 3Riverside Regional Medical Center, Newport News, VA
Purpose/Objective(s): In patients with resectable pancreatic ductal adenocarcinoma (PDAC) who undergo neoadjuvant therapy, the “A-B-C-D-E” classification was developed to prognosticate patient outcomes based on CA19-9 dynamics. In this schema, type A represents an always decreasing CA19-9 level to normalization; type B, bidirectional CA19-9 dynamics to normalization; type C, a consistently normal level; type D, any decrease without normalization; and type E, a rising CA 19-9. Whether such a system applies to prediction of outcomes in patients undergoing definitive chemoradiation (CRT) after upfront chemotherapy in the unresectable setting is unclear. As such, we herein report CA 19-9 dynamics during CRT and describe the association of CA 19-9 response types with prognosis in patients with unresectable PDAC. Materials/
Methods: Patients diagnosed with localized PDAC who received neoadjuvant chemotherapy followed by consolidative chemoradiation between 2018 and 2023 who were ultimately not resection candidates were included. CA 19-9 measurements from time of diagnosis through completion of CRT were analyzed and grouped into CA 19-9 response types based on the “A-B-C-D-E” system. Overall survival (OS) and progression-free survival (PFS) were measured from completion of radiotherapy. Kaplan-Meier analysis was used to estimate event probabilities. Outcomes were compared using the log-rank test. Cox proportional hazards regression analysis were used for univariable analysis. Results: Thirty-eight patients with BRPC or LAPC treated with neoadjuvant chemotherapy followed by consolidative chemoradiotherapy who ultimately were not resected were eligible for analysis. Most patients (N=36, 95%) received either neoadjuvant FOLFIRINOX or gemcitabine-based chemotherapy and capecitabine monotherapy concurrent with radiotherapy. Median baseline CA 19-9 was 124.1 U/ml. Median OS and PFS were 8.3 months and 2.2 months, respectively. Using the A-B-C-D-E classification, patients were grouped into two categories: CA 19-9 dynamics with normalization (A, B, and C) and without normalization (D, E). Median OS for response types with CA 19-9 normalization was 27.5 months and 6.5 months for response types with failure of CA 19-9 normalization (p=0.017). Median PFS for response types with normalization was 4 months and 1.8 months for response types without normalization (p=0.037). On univariate analysis, response types with CA 19-9 normalization was significantly associated with improved OS (p=0.024, HR=3.126, 95% CI: 1.164-8.397) and PFS (p=0.043, HR=2.238, 95% CI: 1.025-4.886) as compared to response types without CA 19-9 normalization. Conclusion: The clinical utility of CA 19-9 in unresectable PDAC as a prognostic biomarker is not well characterized. By using the A-B-C-D-E classification of CA 19-9 dynamics during neoadjuvant chemotherapy, this study demonstrates that response types with CA 19-9 normalization are associated with improved OS and PFS in this setting.
