Abstract
The aim of this study is to investigate whether use of cyclooxygenase-2 (COX-2) inhibitors as auxiliary drug in colorectal cancer (CRC) patients will lead to better survival outcome. This population-based retrospective cohort study was conducted using the Taiwan National Health Insurance Research Database. The cohort consisted of newly diagnosed CRC adult patients during 2003-2010 with at least one prescription of nonsteroidal anti-inflammation drugs. Analysis groups were defined as users or nonusers of COX-2 inhibitors based on their usage prior to or 1 year after diagnosis of CRC. The outcome measurement was overall survival. The application of propensity scores through the inverse probability of treatment weighting (IPTW) was applied to the study groups. Subgroup analyses included stratification of different cancer site, treatment modalities, and first chemotherapy regimens. Kaplan-Meier estimates and Cox regressions were used to compare survival outcome. We identified 14,688 patients with newly diagnosed CRC. The adjusted hazard ratio (HR) with IPTW was 0.91 [95% confidence interval (CI), 0.86-0.96] in patients using COX-2 inhibitors in before and after diagnosis groups, and statistical significance was not reached for usages at only prior to or only after diagnosis. In subgroup analyses, patients with rectal cancer (adjusted HR with IPTW=0.86; 95% CI, 0.79-0.94) who received surgery followed by chemoradiation (adjusted HR with IPTW=0.57; 95% CI, 0.47-0.77) and with adjuvant chemotherapy of FOLFOX regimen (adjusted HR with IPTW=0.81; 95% CI, 0.67-0.99) had survival benefits in using COX-2 inhibitors both prior to and after diagnosis. Use of COX-2 inhibitors was found to be associated with reduction in mortality for CRC patients when taken both prior to and after cancer diagnosis.