Aspirin and Cancer: Trials and Observational Studies

1. John A. Baron

1. Affiliation of the author: Department of Medicine, University of North Carolina School of Medicine Chapel Hill, NC.

1. Correspondence to:
   John A. Baron, MD, Department of Medicine, University of North Carolina School of Medicine, 4160-A Bioinformatics Bldg, Mason Farm Road, Chapel Hill, NC 27599 (e-mail:jabaron{at}med.unc.edu).

Aspirin recently celebrated its commercial centenary, but the exciting news has been a series of articles by Rothwell et al. (1?6) that report on the follow-up of cardiovascular trials begun in the late 1970s and afterward. Those data suggest striking efficacy of aspirin in reducing cancer mortality. One of the recent reports (2) analyzed short-term, in-trial incidence and mortality data from 51 trials. It showed that aspirin conferred a 31% reduction in cancer mortality 5?10 years after subjects were randomized to study treatment, though there was essentially no reduction during the first 5 years.

Motivated by these striking findings, in this issue of the Journal, Jacobs et al. (7) report on the association of aspirin use with cancer mortality in the large American Cancer Society (ACS) cohort. Taking advantage of the study?s repeated assessment of aspirin use, they attempted to replicate the trial findings by evaluating the association of 5 or more years of reported daily aspirin use with subsequent cancer mortality. Unfortunately, they found only 8% or 16% lower cancer mortality, depending on the analytic approach used. In apparent contrast to the trials, associations were similar among those using daily aspirin for less than 5 years and for those using for 5 or more years.

An observational study such as the ACS cohort describes the association of aspirin use with cancer mortality in real life. Did the trials lead us to an overoptimistic assessment, or is the report by Jacobs et al. off the mark? The answer is important, as it clearly will affect an important clinical decision: whether to recommend aspirin for primary prevention of cancer.

An earlier analysis from the ACS cohort assessed cancer incidence in relation to aspirin use (8). As in the report from Jacobs et al., it showed more modest associations than the corresponding ?

Source: http://jnci.oxfordjournals.org/cgi/content/short/104/16/1199?rss=1&ssource=mfr

olympia snowe davey jones dead boston weather monsanto dr seuss birthday jennifer garner jennifer garner