Low-Dose Aspirin, NSAIDs Offer Chemoprevention Options for CRC

Non-aspirin NSAIDs are more effective than other options for colorectal chemoprevention in individuals with previous colorectal neoplasia, according to a new meta-analysis. Low-dose aspirin, however, is a safer option, and offers the best risk-benefit profile.

“Meta-analyses have suggested that aspirin and non-aspirin NSAIDs could be effective in decreasing the risk of advanced metachronous neoplasia … but in the absence of head-to-head trials, their relative efficacy and safety are not known,” wrote study authors led by Mohammad Hassan Murad, MD, of the Mayo Clinic in Rochester, Minnesota.

The new systematic review included 15 randomized controlled trials, totaling 12,234 patients; a Bayesian network meta-analysis was used to provide a relative ranking of 10 different chemoprevention strategies. The strategies were scored using the surface under the cumulative ranking (SUCRA) method, where a score of 1 would indicate a high likelihood a treatment was the best and a score of 0 would mean it was the worst. The results were published online ahead of print in the British Medical Journal.

The strategies included varying combinations and dosages of aspirin, calcium, folate, vitamin D, and non-aspirin NSAIDs, all compared with placebo. Non-aspirin NSAIDs (celecoxib and sulindac) showed the best chemopreventive effect, with an odds ratio (OR) for development of advanced metachronous neoplasia compared with placebo of 0.38 (95% CI, 0.26–0.56). Other non-aspirin NSAIDs, and calcium either alone or combination with vitamin D were also effective. Low-dose aspirin showed a lower OR (0.77), but it did not reach significance (95% CI, 0.58–1.01]).

The non-aspirin NSAIDs were ranked best using the SUCRA method (0.98), followed by low-dose aspirin (SUCRA = 0.67), aspirin plus folic acid (SUCRA = 0.67), and an aspirin/calcium/vitamin D combination (SUCRA = 0.59).

Non-aspirin NSAIDs, however, also carried higher risks than the other options. They had an OR for serious adverse events compared with placebo of 1.23 (95% CI, 1.04–1.45); only calcium also had a significant increase in such events. Non-aspirin NSAIDs ranked second-worst for safety, with a SUCRA score of 0.26. Low-dose aspirin, meanwhile, was safest, with a SUCRA of 0.84.

Taken together, the results suggest low-dose aspirin may be the best option. The authors estimated that non-aspirin NSAIDs would result in 34 more serious adverse events per 1,000 individuals treated compared with placebo, compared with 45 fewer advanced neoplasias in those with previous low-risk neoplasia and 96 fewer in those with previous advanced cases. Low-dose aspirin would yield 35 fewer serious adverse events, along with 20 and 42 fewer advanced neoplasias in the two types of patients, respectively.

Thus, NSAIDs may only be the best option when high-risk neoplasia was previously found. “Shared decision-making with a thorough understanding of patients’ values and preferences in the context of risks and benefits of each agent would be helpful,” the authors wrote.