Association of dynamic changes in tumor markers with overall survival in patients with mucinous appendiceal adenocarcinoma.

Abstract

e16477 Background: Mucinous appendiceal adenocarcinoma (MAA) is the most common subtype of AA, but it is difficult to accurately measure disease burden using standard cross-sectional imaging. The aim of this study was to determine the association between tumor marker (TM) dynamics and overall survival (OS) in MAA patients. Methods: This was a single-institution, retrospective analysis of MAA patients. TMs analyzed were CEA, CA 19-9, and CA-125. TM response was defined as stable TM if <25% change from baseline and decreased/increased if ≥25% change from baseline. Distributions of OS were estimated by the Kaplan-Meier method. Cox proportional hazards model was used to estimate Hazard Ratio (HR) and p-value. Results: A total of 875 patients with MAA and TM measurements on at least two separate dates were identified and included in this analysis. Evaluating the prognostic effect of a rise in TM after 5-7 months off treatment, we found 220 (31%), 47 (13%), and 46 (13%) patients had TM increase from baseline for CEA, CA 19-9, and CA-125, respectively. Increase in CEA was associated with worse mOS (172.8 mo increase vs. 281.5 mo stable, HR: 1.96; p=0.0002). Patients within the top decile of CEA increase had even shorter mOS at 124.5 mo (HR 1.79 vs. stable CEA; p<0.0001). This association was true for patients with both low-grade (G1, 195.9 mo vs. 264.2 mo, HR: 1.64; p=0.0299) and high-grade (G2/G3, mOS: 143.1 mo vs. 208.3 mo, HR: 2.70; p<0.0001) disease. Rise in CA19-9 was similarly associated with worse mOS (173.5 vs. 264.0 mo, HR: 2.32; p=0.0018). Rising CA-125 had a trend (mOS: 213.1 vs. 281.5 mo, HR: 1.47; p=0.15) towards worse OS that did not reach statistical significance. Evaluating the prognostic significance of decreasing TM while on systemic therapy, we found that among 82 patients with baseline elevated CEA, 44 (54%) had a 25% decrease within 3 months from start of therapy, which was associated with improved mOS (114.4 mo decreased vs. 79.6 mo stable, HR: 0.54; p=0.049). There were insufficient numbers of patients for similar analysis for CA 19-9 or CA-125. Conclusions: Increases in CEA and/or CA 19-9 during surveillance periods are associated with worse survival for MAA patients. For patients with a positive baseline CEA, decrease on systemic therapy is associated with improved survival. Given these findings, serial measurement of TM is warranted for patients with MAA to identify high risk patients for either initiation or change in systemic therapy. Analysis of TM change during systemic therapy and observation. Off Treatment N HR mOS TM ↑ mOS TM Stable P-value CEA 704 1.96 172.8 mo 281.5 mo 0.0002 Grade 1 433 1.64 195.9 mo 264.2 mo 0.0299 Grade 2/3 249 2.70 143.1 mo 208.3 mo <0.0001 CA 19-9 318 2.32 173.5 mo 264.0 mo 0.0018 CA-125 356 1.47 213.1 mo 281.5 mo 0.15 Systemic Therapy (CEA) mOS TM ↓ All Patients 104 0.69 115.4 mo 94.4 mo 0.2 Positive Baseline CEA 82 0.54 114.4 mo 79.6 mo 0.049