Low-Dose Aspirin and Pancreatic Cancer Risk—Letter

We very much enjoyed reading the recent article by Risch and colleagues (1), which examined the association between aspirin use and pancreatic cancer risk in a population-based case–control study in Shanghai, China. The authors' conclusions that regular aspirin use may confer chemopreventive benefit certainly reflect a growing literature, as indicated in their meta-analysis of studies of aspirin use (see Fig. 1 in ref. 1), including our own report (2). Parenthetically, the years of our and other published works are incorrectly labeled in the Figures. The authors also concluded that low-dose aspirin, which does not contain significant anti-inflammatory properties (3), may also confer benefit. This latter conclusion is supported by a separate meta-analysis conducted by the authors (see Fig. 2 in ref. 1). However, our article examining aspirin and non-aspirin NSAID use among postmenopausal women in the Women's Health Initiative observational study and clinical trials cohort, including 129,013 women and 378 incident pancreatic cancer cases, was overlooked and may have influenced the latter conclusion (2). In 2014, we published that consistent use of any aspirin was associated with a statistically nonsignificant 31% reduction in pancreatic cancer risk [HR, 0.69; 95% confidence interval (CI), 0.45–1.06]. In a Supplementary Table published online, we showed that this inverse association was restricted to users of regular-strength aspirin (>100 mg; HR, 0.51; 95% CI, 0.22–1.16), whereas use of low-dose aspirin (≤100 mg) was not associated with risk (HR, 1.04; 95% CI, 0.53–2.05). Our article represents one of very few prospective studies that have examined this hypothesis. Had these data been included in their meta-analysis of low-dose aspirin, we believe they may have tempered the authors' conclusions. We suggest that an updated analysis using this data be reported on as an update to their recent article.

See the Response, p. 1155