Prognostic significance of preoperative and postoperative CA 19-9 normalization in pancreatic adenocarcinoma treated with neoadjuvant therapy or surgery first

Abstract

Presenter: Timothy P DiPeri MD | The University of Texas MD Anderson Cancer Center Background: Serum carbohydrate 19-9 (CA 19-9) is a biomarker commonly used to assess preoperative disease burden and postoperative disease clearance for patients with pancreatic adenocarcinoma (PDAC). Preoperative normalization of CA 19-9 following neoadjuvant therapy (NT) is an established positive predictor of survival. However, the implications of normalized CA 19-9 levels after surgery have not been compared between patients treated with neoadjuvant therapy (NT) vs. surgery first (SF) sequencing. We hypothesized that postoperative normalized CA 19-9 after a preoperative non-normal(ized) CA 19-9 would not have similar outcomes as those who were normal(ized) before and after surgery. Methods: Clinicopathologic characteristics for patients with PDAC who underwent resection from July 2011 to October 2018 were abstracted from a prospectively maintained database. Patients with pre- and postoperative CA 19-9 measurements, total bilirubin 1 U/mL were included. Patients were assigned to six groups based on treatment sequence and pre- and postoperative CA 19-9. Groups NT-1 and SF-1 had normal(ized) pre- and postoperative CA 19-9, Groups NT-2 and SF-2 had elevated preoperative CA 19-9 that did not normalize until after surgery, and Groups NT-3 and SF-3 had persistently elevated CA 19-9. Overall survival (OS) and recurrence-free survival (RFS) were compared among groups. Results: Of 373 consecutive patients, 228 were included in our analysis. Estimated median RFS and OS were 17 and 49 months, respectively. In patients receiving NT, those with normal pre- and postoperative CA19-9 (NT-1) had the best median RFS and OS (26 and 77 months), followed by those who did not normalize until after surgery (Group NT-2, 16 and 39 months). For patients undergoing SF, those with normal pre- and postoperative CA19-9 (SF-1) had the best median RFS and OFS (28 and 57 months), followed by those with preoperatively elevated CA19-9 who normalized postoperatively (SF-2, 18 and 49 months). Groups NT-3 and SF-3 had persistently elevated CA 19-9 and were associated with the worst median RFS and OS durations (Figure 1). Conclusion: While a normal(ized) postoperative CA 19-9 may result in similar survival as preoperative normal(ization), postoperative normalization is not guaranteed. Thus, NT should be considered in patients who present with elevated CA 19-9 as preoperative normalization correlates with improved RFS and OS. If considering SF, the ideal patients are those with normal CA 19-9 at presentation.