Prognostic and predictive effect of COX-2 expression in appendiceal adenocarcinomas.

Abstract

526 Background: Cyclooxygenase-2 (COX-2), an inducible isoenzyme, is upregulated in inflammation and involved in cancer progression by promoting angiogenesis, cell proliferation and migration and by inhibiting apoptosis. COX-2 is implicated in colorectal tumorigenesis and by extension COX-2 expression testing and therapeutic inhibition has been considered in appendiceal adenocarcinomas (AAs). However, its role in this setting has never been well studied. The purpose of our study was to investigate COX-2 expression in AAs and to evaluate its prognostic and predictive significance. Methods: We performed a retrospective review of 607 patients with AAs treated at MD Anderson Cancer Center between 2002 and 2010. Immunohistochemistry was performed for COX-2 expression (COX2 mAb Clone CX229, Cayman Chemical) in 49 (8%) patients. Thirty (61%) stained positive and 19 (39%) showed no staining. Kaplan-Meier product limit method and log-rank test were used to estimate overall survival (OS) and to determine association between OS, COX-2 expression and other characteristics. Results: Median age at diagnosis was 47 yrs (range 34-78 yrs). Grade (P = 0.003), completeness of cytoreduction score (P = 0.010) and stage (P = 0.023) were significantly associated with OS. Median OS for patients with COX-2 positive and COX-2 negative tumors was 58.3 and 42.8 months, respectively (P = 0.232). Nine COX-2 positive patients received celecoxib (selective COX-2 inhibitor) in combination with other chemotherapy. Mean treatment duration was 5.5 months with best radiographic response being stable disease in 7 patients. Reduction in tumor markers (CEA or CA19-9) was seen in 7 cases (median reduction of 30%). Similar treatment duration (6.6 months), rate of stable disease (6/8 pts) and tumor marker reduction (7/8 pts) was seen on the preceding non-celecoxib containing chemotherapy in these patients. Median OS, with and without COX-2 inhibition, was 55.7 and 57.6 months respectively (P = 0.57). Conclusions: In this cohort, COX-2 expression is not a significant prognostic factor in AAs. Benefit from COX-2 inhibition in COX-2 expressing AAs is unclear. Current data does not support the routine use of either COX-2 testing or COX-2 inhibition therapy in AAs.