STOCKHOLM – Data from 170,000 patients in 26 randomized trials may finally dismiss the idea that low cholesterol levels during statin therapy play a role in causing cancer.
Dr. Jonathan Emberson
“The results are really very reassuring. There is no evidence that cancer risk is increased when very low cholesterol concentrations are achieved with high-dose statins,” Jonathan Emberson, Ph.D., said at the annual Congress of the European Society of Cardiology.
“There is no suggestion of an emergence of any hazard with longer duration of treatment, at least within a period of about 5 years. There is no evidence that low cholesterol increases cancer risk at any site or in any group of individuals. I think the question about statins has now probably been answered as well as it can be from the randomized trials,” said Dr. Emberson, a statistician in the clinical trial service unit at the University of Oxford (England).
Follow-up for the patients in the trials ran 5-6 years, producing “extremely reassuring” results for long-term safety, he said in an interview.
Concern that very low serum cholesterol levels – and hence statin therapy – might boost cancer incidence has probably not had a big impact on statin use. But “it’s a hypothesis that’s been raised at various times over the past 20 years. Every now and again, a trial threw out a random result that raised a new hypothesis,” he said. For example, in 1996 results from the CARE (Cholesterol and Recurrent Events) trial, which compared 40-mg pravastatin and placebo for secondary prevention in more than 4,000 patients who were followed for an average of 5 years, showed 12 cases of breast cancer during follow-up in the pravastatin arm, compared with one case in the placebo arm, a statistically significant difference (N. Engl. J. Med. 1996;335:1001-9).
“Random things happen all the time,” Dr. Emberson noted. Instances like the CARE results show “a significant excess, but what’s important is, it wasn’t supported by data from all the other statin trials. Occasionally, trials throw up hypotheses that can be tested. We attempted to systematically test all of those hypotheses using all of the data.”
The 26 statin trials in the meta-analysis included all those that were published through the end of 2009 with at least 1,000 patients who were followed for at least 2 years. In all, 21 trials compared a statin against placebo, and 5 compared a low statin dose with a higher statin dose. The 170,000 patients in all 26 trials developed 10,000 cases of cancer during follow-up, with more than 3,500 cancer deaths.
Statin treatment gave no hint of an influence on cancer rates in all 26 studies together, nor separately in the 21 studies in which it was compared against placebo, nor in the 5 in which high dose was compared against low dose. The results also showed no sign of cancer increase when treated patients started with low serum levels of LDL cholesterol. The findings showed no cancer impact with longer duration of statin use, no impact for various, specific cancer types, and no difference by age or by sex. The analysis showed no suggestion of an increased risk in the elderly. And no increased risk appeared for gastrointestinal cancers, another cancer type that gave a signal for higher risk in one of the trials.
“The value of our analysis is that we were able to systematically test all of the hypotheses that had been raised in a much larger data set than has previously been possible, and the results are very reassuring for the many millions of patients who take statins, at least for 5-6 years’ duration. The meta-analysis could not address the risk faced by patients treated for longer period of time, such as 15 or 20 years, but results from studies that followed patients for longer than 6 years have not suggested any cancer concerns,” Dr. Emberson said.
Dr. Emberson said that he had no disclosures.