Aspirin in the prevention of colorectal cancer recurrence.

Abstract

578 Background: Aspirin use reduces the incidence of colorectal adenomas, and of colorectal cancer (CRC) in patients with hereditary non-polyposis colorectal cancer. There is limited data on the effect of aspirin in the secondary prevention of CRC, currently the subject of multiple prospective randomized trials. We aimed to test the hypothesis that aspirin reduces disease recurrence in patients diagnosed with early CRC at a high volume, single institution. Methods: A retrospective analysis was conducted of all patients (pts) treated with at least one cycle of adjuvant chemotherapy for stage II or III CRC over a 5yr period (2009–13). Patients with synchronous CRC were excluded. Aspirin use at the onset of adjuvant chemotherapy was sourced from the universal electronic recording of concomitant medications. Kaplan-Meier curves and the log-rank test were used to compare crude relapse free survival (RFS) between pts who were using aspirin to those who were not. Propensity score analyses was used to balance the groups according to potentially confounding variables: tumor site (colon vs rectum), tumor stage (II vs III), and adjuvant chemotherapy (oxaliplatin-based vs no oxaliplatin). The propensity scores were estimated through a multivariable logistic regression model, and a Cox proportional hazards model used to assess the effect of aspirin use on RFS in the sample weighted according to the inverse probability of receiving aspirin. Results: A cohort of 231 pts met eligibility criteria; median age 64yr; 50% colon, 50% rectal; 19% stage II, 81% stage III; 35% received oxaliplatin; 13% (n=31) were using aspirin and 26% of pts (n=61) developed CRC recurrence in median follow-up of 2.9yr. Unadjusted RFS was not significantly different for those using aspirin (log rank p value 0.88). There was good overlap in propensity scores for the two groups. The hazard ratio for RFS with aspirin use from the weighted Cox regression model was 1.2 (95% CI 0.62-2.29), indicating no statistically significant effect of aspirin on CRC relapse. Conclusions: In this single institution series, we did not find evidence that aspirin use at onset of chemotherapy had an effect on CRC relapse free survival. These results do not exclude an aspirin effect that is modest or restricted to select CRC subgroups.