Aspirin protective against digestive cancers?

Regular* aspirin use may have a favourable effect on digestive** cancers, a comprehensive meta-analysis has shown.

“There are ~175,000 deaths from bowel cancer predicted for 2020 in the EU, of which ~100,000 will be in people aged 50–74 years. If … regular aspirin use increases from 25–50 percent in this age group, this would mean that 5,000–7,000 deaths from bowel cancer and 12,000–18,000 new cases could be avoided if further studies show that aspirin does indeed [reduce] cancer risk,” explained senior author Dr Carlo La Vecchia from the University of Milan School of Medicine in Milan, Italy.

“Corresponding figures would be ~3,000 deaths each for oesophageal, stomach, and pancreatic cancer, and 2,000 deaths from [liver] cancer … Given the unfavourable prognoses for these cancers, the number of new cases would be only slightly greater … The findings … may have implications for the prevention of these highly lethal diseases,” he continued.

Pooled data from 113 studies (n=156,000 cases) revealed an inverse association between regular aspirin use and the risks of colorectal, squamous cell oesophageal, oesophagus/gastric cardia, stomach, hepatobiliary tract, and pancreatic cancers (relative risks [RRs], 0.73, 0.67, 0.61, 0.64, 0.62, and 0.78, respectively). [Ann Oncol 2020;doi:10.1016/j.annonc.2020.02.012]

“The associations are somewhat stronger in case-control than in cohort and nested case-control studies and are characterized by some between-study heterogeneity. The risk estimates are consistent across sex, geographical areas, and other selected covariates,” said the researchers.

Colorectal cancer

A linear dose-risk relation was seen between colorectal cancer (CRC) and aspirin use (RRs, 0.87–0.90, 0.64, and 0.50 for 75–100, 325, and 500 mg/day). These translate to a 10- and 35-percent CRC risk reduction with the respective low- and regular-dose, while high-dose aspirin cut the risk by half.

“However, the estimate for high-dose aspirin was based on a few studies and should be interpreted cautiously,” noted co-author Dr Cristina Bosetti from the Mario Negri Institute for Pharmacologic Research in Milan, Italy.

In terms of duration of aspirin use, the RRs for CRC dropped by 10 years (RR, 0.71). While the risk levels off for longer aspirin use, this is challenging to interpret given the paucity of data evaluating long-term aspirin use, noted the researchers.

Other digestive cancers

Linear drops in RR were also seen between a 10-year duration of aspirin use and risk of squamous cell oesophageal, stomach, and pancreatic cancers (RRs, 0.77, 0.65, and 0.56, respectively).

“[S]ince aspirin may cause gastrointestinal bleeding … part of the inverse association [could be] due to the avoidance of aspirin use in patients with early symptoms of oesophageal or stomach cancer,” said the researchers.

Chemopreventive role

The findings imply that the chemopreventive effect of aspirin tends to increase with longer duration of use, and with increasing dose for CRC, noted the researchers. From a biologic standpoint, this effect has been attributed to cyclooxygenase (COX) inhibition. “COX-2 isoform is abnormally expressed in many cancer cell lines and is implicated in the process of carcinogenesis, tumour growth, apoptosis, and angiogenesis.” [J Am Coll Cardiol 2016;68:967-976; J Natl Cancer Inst 1998;90:1529-1536; J Natl Cancer Inst 1998;90:1609-1620]

However, La Vecchia cautioned against taking aspirin for cancer prevention. “[This] should only be done in consultation with a doctor who can account [for] individual risk factors.”

While the findings do not confirm a link between aspirin use and head/neck cancers (HNCs), the variable characteristics and aetiological factors of different HNCs should be taken into context. Additional studies are warranted to elucidate the role of aspirin in this setting, they said.

It should also be noted that the bulk of the data is mostly observational hence the potential for bias. Inaccuracies in patients’ accounts may have factored in, and most studies lacked data on other medications that could have influenced the findings. The results should thus be validated by related ongoing trials. [Trials 2017;18:50; J Am Coll Cardiol 2015;66:74-85]