Aspirin Use for the Prevention of Colorectal Cancer: An Updated Systematic Evidence Review for the U.S. Preventive Services Task Force

BACKGROUND: Colorectal cancer (CRC) is the third most commonly diagnosed cancer in both men and women in the United States (U.S.). Aspirin may inhibit CRC development and related mortality. PURPOSE: We conducted this systematic evidence review on aspirin use for the prevention of CRC to support the U.S. Preventive Services Task Force (USPSTF) in updating its previous recommendation. Our review addressed four key questions in adults without a history of CRC, familial adenomatous polyposis, or Lynch Syndrome: 1) Does regular aspirin use reduce CRC mortality or all-cause mortality? 2) Does regular aspirin use reduce the incidence of CRC? 3) Does regular aspirin use reduce the incidence of colorectal adenoma? 4) What are the harms of regular aspirin use for the prevention of colorectal cancer? DATA SOURCES: We performed a search of MEDLINE, PubMed, and the Cochrane Collaboration Registry of Controlled Trials for studies published from January 2004 through May 2014. We supplemented searches by examining bibliographies from previous systematic reviews, retrieved articles, and the previous USPSTF review. We searched federal agency trial registries for ongoing and/or unpublished trials. STUDY SELECTION: We conducted a dual review of 865 abstracts against prespecified inclusion criteria. We retrieved 149 potentially relevant articles, which two reviewers independently evaluated using well-defined inclusion/exclusion criteria and critically appraised for risk of bias. Discrepancies were resolved by discussion with a third reviewer. DATA EXTRACTION AND ANALYSIS: For all fair-quality and good-quality studies, a single investigator extracted study characteristics and outcomes into structured tables and a second investigator verified accuracy. Elements abstracted for each study included study design, population characteristics, sample sizes, exposures, outcomes, and measures of association. We created summary evidence tables to capture key study characteristics and sources of heterogeneity. In addition to the overall results for each included study, we also presented results by dose, duration, latency, and adenoma history where possible. We used forest plots stratified by potentially important exposure and study characteristics to visually identify patterns in the study results and help determine if pooling across studies was appropriate. We used the Mantel-Haenszel fixed effects model to estimate the combined effect and confidence interval; for very rare events (incidence less than one percent), we calculated the Peto odds ratio. RESULTS: Daily or alternate-day aspirin at ≥75 mg was associated with a small reduction in all-cause mortality risk in the first 10 years after randomization (summary relative risk, RR, 0.94, [95% confidence interval, CI, 0.89 to 0.99]) in 11 randomized controlled trials (RCTs) among persons in the general population (i.e., selected without considering their adenoma history). Over a 20+ year period, aspirin appeared to reduce the risk of CRC mortality by approximately 33%. However, long-term data on CRC mortality may have limited applicability, particularly from the perspective of a low-dose aspirin benefits in a primary CVD population addressing women as well as men. Two of four trials were in those with pre-existing cardiovascular disease and two involved dosages of 500 mg or greater daily, with no longer-term mortality results available for alternate-day regimens. Data on mortality among persons with a prior colorectal adenoma were also sparse. Six RCTs of aspirin for primary and secondary CVD prevention provided data on the effect of regular aspirin use on invasive CRC incidence in the general population. In this population, aspirin had no effect on CRC incidence in the first 10 years following randomization, but reduced CRC incidence by approximately 40 percent after a latency of 10 years (summary RR, 0.60 [95% CI, 0.47 to 0.76]). Over a 20+ year period, aspirin appeared to reduce the risk of CRC incidence by approximately 20 to 24%. Data on aspirin use and CRC incidence in persons with a prior adenoma were limited and represented only short-term followup (fewer than 5 years) and could not, therefore, provide sufficient information on the effect of aspirin use on CRC incidence. In persons with a prior adenoma, data were conflicting, but there was some suggestion of a decreased risk of adenoma incidence over a 3- to 4- year period. Data on aspirin and adenoma risk in the general population were sparse. Data from RCTs suggested that aspirin increased the risk of serious gastrointestinal bleeding (summary OR, 1.94 [95% CI, 1.44 to 2.62]), intracranial bleeding (summary OR, 1.53 [95% CI, 1.21 to 1.93]), and hemorrhagic stroke (summary OR, 1.47 [95% CI, 1.16 to 1.88]), but not fatal gastrointestinal bleeding (summary OR, 1.00 [95% CI, 0.43 to 2.36]). LIMITATIONS: Limited data were available to address differences in possible effects of aspirin in subgroups (e.g., age, sex, race) or to compare daily vs. alternate-day aspirin use. Long-term followup data were not identified for persons with a history of adenoma. CONCLUSIONS: Aspirin appears to reduce the risk of CRC incidence after an induction and latency period of approximately 10 years, with a similar effect on CRC mortality. The applicability of data for long-term effects of low-dose aspirin on CRC mortality, however, is limited, particularly in the context of a population selected for primary CVD prevention. Aspirin does not appear to have a strong effect on all-cause mortality within 10 years of initiating use, and data on long-term cumulative risk of all-cause mortality were sparse.