Patients with Barrett's esophagus who took statin drugs had a 45% lower risk of developing esophageal adenocarcinoma than did similar patients not taking statins, and patients taking aspirin or NSAIDs also had a reduced risk, Dr. Hashem B. El-Serag and his colleagues reported.
To learn more about the effect of statins, proton pump inhibitors, aspirin, and NSAIDs on progression of Barrett's esophagus (BE) to cancer, researchers at the Department of Veterans Affairs Medical Center and Baylor College of Medicine, both in Houston, used a national VA database to identify 11,823 records of people diagnosed with Barrett's between 2000 and 2002 (Gastroenterology 2010 June [doi: 10.1053/j.gastro.2010.02.045]).
Among these records, they identified 116 cases of esophageal adenocarcinoma and 696 controls who had BE but not adenocarcinoma. The controls and cases were matched for age (within 5 years), race (78% of cases and 73% of controls were white), and sex (97% of both cases and controls were men). Data were subsequently adjusted for frequency of outpatient visits and socioeconomic status.
All of the controls had been diagnosed with BE on the same date (plus or minus 14 days) as the cases. The researchers examined records of prescriptions filled by the cancer patients between their BE diagnosis and cancer diagnosis, and looked at prescriptions filled by controls for 2 years after BE diagnosis.
Filled prescriptions for statins and aspirin/NSAIDs were associated, to different degrees, with a decreased risk of developing esophageal adenocarcinoma in people with BE.
A 36% decrease in cancer risk was found for aspirin/NSAIDs, and this “was largely expected although not previously shown in a cohort of BE patients,” Dr. El-Serag said in an interview. With the statins, however, a hypothesis of chemopreventive effects in esophageal adenocarcinoma was based on in vitro studies, “and we were surprised,” he said, about the 45% decrease in risk.
The investigators could not conclude anything about proton pump inhibitors, because use of those drugs was nearly universal among cases and controls. This, said Dr. El-Serag, was not necessarily a bad thing, as it removed a potential confounding factor.
“The key point of the study is that, unlike any previous studies that examined the effect of NSAIDs on esophageal cancer, all the subjects had BE,” Dr. El-Serag said. “Everyone started on an equal playing field. We had pooled data from 128 hospitals, so to my knowledge this is the largest cohort assembled to date.”
Dr. El-Serag and his coauthors wrote in their analysis that the study's limitations included its relatively brief length, using only 2 years of prescribing data, and the fact that the records of prescriptions filled through the VA system likely presented only a partial picture of medications taken.
For the purposes of the study, the actual number of prescriptions filled did not matter—even one fill of a statin drug was considered statin use. “Biologically, it seems implausible that one fill of a statin could have an effect, but we regard one fill as a potential indicator of other fills,” he said.
Also, since NSAIDs and aspirin are readily available over the counter, patients of lower socioeconomic status would likely be the only users of the VA prescriptions to obtain these, presenting another problem for the researchers. “One cannot capture OTC medication with a study like this,” Dr. El-Serag said.
Disclosures: The study was funded in part by a grant from the National Institutes of Health; none of the investigators declared conflicts of interest.
Unlike previous studies examining NSAIDs' effects on esophageal cancer, all the subjects had Barrett's.
Source DR. EL-SERAG