Aspirin prevents colon cancer

A recent systematic review suggested that aspirin is associated with decreased colorectal cancer (CRC), an equivalent effect to performing sigmoidoscopy or fecal occult blood testing, FOBT (see doi:10.1111/apt.13857).

Details:

--Databases searched as of November 2015 for randomized trials in average-risk populations that reported CRC mortality, CRC incidence, or both, with a minimum follow-up of 2 years.

--17 publications from 15 studies with about 38K patients (6 trials) in the aspirin arm and 160K in the screening arms (4 trials for FOBT, 5 for flex sig)

Results:

--The effect of aspirin on colorectal cancer (CRC) mortality:

    -- not significantly different from FOBT, RR 1.03, (0.76–1.39)

    -- not significantly different from flexible sigmoidoscopy, RR 1.16 (0.84–1.60).

--No difference in CRC incidence between aspirin and either FOBT or flex sig

--Death from cancer in the proximal colon:

    --aspirin was much more effective than either FOBT, RR 0.36 (0.22–0.59) or flexible sigmoidoscopy, RR 0.37 (0.22–0.62)

-- flexible sigmoidoscopy was superior to FOBT, RR 0.84 (0.72–0.97).

Commentary:​

--of note, in this systematic review including several newer studies, there was a 20% decrease in CRC incidence by flex sig screening, but no significant difference by FOBT. Aspirin vs placebo led to a 14% decrease in CRC incidence

--CRC mortality was reduced by 16% by FOBT and 26% by flex sig. Aspirin reduced CRC mortality by 19%, but this was not statistically significant (the difference here may be related to the lower numbers of participants in the aspirin as opposed to the screening studies)

--aspirin was equally effective to screening in reducing CRC incidence, with apparently superior effectiveness for proximal cancers. This may be very important, since the benefit of colonoscopy appears much less significant for these lesions, further supporting prevention as the most effective initiative. A pretty recent blog (http://gmodestmedblogs.blogspot.com/2017/04/one-time-flex-sig.html  ) highlighted the value of a one-time flex sig leading to significant reductions in colorectal cancer incidence and mortality after 11 years of followup, noting that in a large population-based case-control Canadian study, “excision of colonoscopy-discovered right colonic lesions (i.e. beyond the reach of the sigmoidoscopy) do not affect mortality (eg see Baxter NN. Ann Intern Med  2009; 150:1).” Other studies have found more positive results for screening (see Brenner H. Ann Intern Med 2011; 154: 22), though the protective effect of colonoscopy in this German population-based study was much less in right-sided vs left-sided lesions.

--a more recent population-based study of 2419 patients with invasive CRC from registries in the US, Canada, and Australia found: in those on aspirin all-cause mortality after 10.8 years was reduced by 25% and CRC-specific mortality was reduced by 56% !!!, though benefit was limited to those with KRAS wild-type tumors (about 1/2 of their total). and this associated mortality improvement was even greater in those newly starting aspirin after colon cancer diagnosis (see Hua X. J Clin Onc 2017; 35:2806)

--aspirin also has potential benefit for other cancers. For example, http://gmodestmedblogs.blogspot.com/2012/05/2-articles-on-aspirin-and-colonbreast.html  has 2 articles on aspirin and cancer: one showing 36% decrease in distant metastases and one showing a general decrease in cancer. also, aspirin may decrease ovarian cancer risk, 
see

http://gmodestmedblogs.blogspot.com/2014/02/aspirin-decreases-risk-of-ovarian-cancer.html  , and decrease prostate cancer incidence, metastatic disease​, and mortality (see http://gmodestmedblogs.blogspot.com/2014/02/aspirin-use-and-prostate-cancer.html  )

--there have been many studies, some dating back to the 1970s per my recollection, which document that NSAIDs lead to decreases in adenomas (including some studies finding regression of familial polyposis adenomas), better differentiation of adenomas, and reduced risk of adenoma recurrence (on the order of 45%, see  doi:10.1158/1940-6207.CAPR-11-0107 )

--potential mechanisms: unclear, but several possibilities. there may be a role for COX inhibition (exp COX-2), with some studies finding upregulation of COX-2 in adenomas as well as increased levels of prostanoids in colorectal tumors; likely non-COX related effects associated with NSAID-induced increased apoptosis and reduced cell proliferation (see Gupta RA. Gatroenterol 1998; 114: 1095), and there is also perhaps increased aspirin/NSAID-related bleeding leading to increased colorectal investigation and treatment for earlier lesions (ie, earlier detection).

--caveats: aspirin studies are much smaller than screening studies, aspirin studies were all initiated for CVD protection and assessment of CRC effects were therefore post-hoc analyses, aspirin studies had active comparator (placebo) as opposed to screening studies, and there have been no direct comparisons between aspirin and colorectal screening, (ie, this was effectively a network meta-analysis, which is a statistical approach to approximate the relative effects of different interventions in different studies in lieu of direct comparisons, raising issues about comparing different patient samples/settings, different inclusion/exclusion criteria, etc, which could undercut its validity)

--i wonder if there might be synergy for a screening/aspirin combo. screening colonoscopy for example does not prevent all CRC deaths, which elevates CRC prevention as an important aspect of improving morbidity/mortality. and this is especially true for more proximal lesions, as noted above. CRC prevention includes the usual suspects: decreasing obesity, diabetes/hyperinsulinemia, red meat consumption, processed foods (“avoid foods that contain additives that your grandmother would not recognize”, though maybe now “your great grandmother”), tobacco, and alcohol, but also the use of aspirin. perhaps a reasonable strategy is: only one or perhaps two cycles of screening colonoscopy to pick up prevalent lesions, along with aspirin prophylaxis/colon-healthy lifestyle to minimize new ones.  this would need an RCT to determine the best strategy, but the combination of prevention and some level of detection of prevalent early lesions makes a lot of sense.

so, the reason i bring up this review is:

--it demonstrates, with the above caveats, that the order of magnitude of aspirin-related decreases of colorectal cancer prevention is similar to the documented screening interventions (unfortunately, there are no studies documenting that either colonoscopy or FIT testing leads to decreased mortality)

--the USPSTF did add colorectal cancer protection to their aspirin recommendations, giving the combo of CAD and colon cancer prevention a "B" rating: they recommend it, with high certainty of moderate benefit or moderate certainty that benefit is moderate to substantial

--and, my sense is that many of us are not including cancer prevention in our algorithm for prescribing aspirin therapy, which should probably skew the decision-making at least somewhat towards its use

for prior blogs on colon cancer, see http://blogs.bmj.com/bmjebmspotlight/?s=colon+cancer&submit=Search


to be added to the email list: contact me at gmodest@uphams.org