Article
Author(s):
Retrospective analysis reveals that statin use is not associated with increased risk of cancer.
Study results presented at the American Heart Association Scientific Sessions 2010 suggest that statins -- widely prescribed in the US for managing dyslipidemia, atherosclerosis, and cardiovascular event risk reduction -- present no significant increased risk for cancer. Time will tell what effect these results will have on the ongoing debate over whether statins are linked to cancer risk, with high-profile studies showing conflicting results.
Researchers led by Claudio Marelli, of S2 Statistical Solutions Inc. in Cincinnati, OH, sought to determine whether cancer can be attributed to statin use among a general population of older adults in the US with at least three years of follow-up. They used a retrospective cohort analysis of incidence of cancer in older adults who have and have not used statins. More than 11 million analyzable patient records from January 1990 though February 2009 were drawn from the GE Centricity electronic medical records database. Propensity matching found pairs of patients receiving and not receiving statins who shared similar propensities for statin use based on matching variables.
Propensity score methods matched 45,857 comparison pairs of statin vs. non-statin patients. Before matching, cancer occurred in 23,906 of 203,763 (11.7%) of statin users. After matching, incidence of cancer in statin patients fell to 11.37% vs. 11.1% in matched non-statin patients. Pairs were followed for an average of eight years.
Multivariate-matched Cox regression analysis showed a non-significant hazard ratio. Kaplan Meier curves for diagnosis of any cancer up to 10 years also showed no difference between statin and non-statin users.
The study deemed “time zero” the time of the first statin use; for those who had never used a statin, it was the time when the LDL value was recorded, said Paul Cload, PhD, GE HealthCare, UK, who presented the findings Wednesday at AHA 2010. Those with a cancer diagnosis prior to “time zero” were excluded.
Men in the study were 45 and older and women were 55 and older. The average age overall was 64 and about 50% of subjects had a history of smoking, which was determined by asking the patient whether he or she had ever smoked.
“When we started, the database included about 11.2 million patients. About 1.2 million patients had received a statin and about 10 million had never received a statin,” Cload said. The study showed no difference between the two groups as to types of cancer that developed, which you would expect if statins were linked to cancer, Cload said.
Limitations of the study include the fact that it was not a randomized, controlled trial and that data were captured in electronic medical records (the way physicians capture data and what they capture varies from physician to physician). Because one database was used, it is unclear whether the results can be randomized to the greater population.
“We’re seeing increasing adoption of electronic medical records in the United States and across the rest of the world,” Cload said. “As time goes by we’re going to see more and more data sets such as this coming forward, which will raise questions about what questions can and can’t be answered with such databases and what statistical techniques should be used to analyze such data.”