NO RISK OF CANCER WITH STATIN USE

Lead author Dr Julia Hippisley-Cox (University of Nottingham, UK) told heartwire : "The take-home message of our study is that it is very reassuring, that there aren't increased risks of cancer with statins." And she notes that this large observational study of more than 350 general practices in England and Wales "is reassuring in a way that clinical trials cannot be, because they are not powered to look at adverse effects and don't go on for long enough."

The study, published online May 20, 2010 in BMJ, looked at all unintended effects of statins at a population level, and in a companion paper in Heart [2], also published online May 20, the researchers have developed algorithms to help healthcare providers predict which individuals are most likely to suffer side effects with these drugs [3]. The latter "is a novel look at how to predict who might get serious adverse reactions from statins," Hippisley-Cox says.

In an accompanying editorial in BMJ [4], cardiologists Dr Alawi Alsheikh-Ali (Sheikh Khalifa Medical City, Abu Dhabi, UAE) and Dr Richard H Karas (Molecular Cardiology Research Institute, Tufts University School of Medicine, Boston, MA) say the work of Hippisley-Cox and her coauthor Dr Carol Coupland (University of Nottingham) "is an important addition to the existing literature on statins because it is the best available 'real-world' estimate of the risk of adverse events with statin treatment."

Per Guidelines, Statin Benefits Outweigh Risks

In the BMJ paper, Hippisley-Cox and Coupland estimate the number of additional events potentially caused or prevented by statins per 10 000 patients treated in the general-practice QResearch database from January 2002 to June 2008.

They found no significant association between use of individual statins and risk of Parkinson’s disease, rheumatoid arthritis, venous thromboembolism, dementia, osteoporotic fracture, or many cancers; in fact, there was a significantly reduced risk of esophageal cancer associated with statin use.

They did find an increased risk associated with statin use for moderate or serious liver dysfunction, acute renal failure, moderate to serious myopathy and cataracts, and evidence of a dose response for acute renal failure and liver dysfunction, with higher doses being associated with greater risk. These risks were similar across all the statins, said Hippisley-Cox, with the exception of increases in liver enzymes, which occurred more often with fluvastatin (Lescol, Novartis).

Overall, for every 10 000 high-risk women treated with statins, there would be approximately 271 fewer cases of cardiovascular disease, eight fewer cases of esophageal cancer; 74 extra patients who experience liver dysfunction; 23 extra patients with acute renal failure, 307 extra patients with cataracts, and 39 extra patients with myopathy. Similar figures were found for men, except rates of myopathy were higher.

Alsheikh-Ali and Karas say that by using a large database, Hippisley-Cox and Coupland "overcome the problem of the low incidence of statin-associated adverse events, and they provide information on adverse effects that clinical trials are not adequately powered to estimate."

Alsheikh-Ali told heartwire the findings "are reassuring. The muscle and liver side effects are largely reversible, and when balanced against heart attacks and strokes--which are prevented by statins--the equation favors using statins in people at risk as recommended by current professional guidelines. And statins do not cause cancer." However, the association between statins and some of the side effects reported in the study "needs to be interpreted with caution," because of residual confounding and ascertainment bias, he noted. The analysis does not prove that statins cause cataracts, for example.

Good Algorithms to Predict Renal Failure, Myopathy, and Cataracts

Hippisley-Cox says her BMJ data are useful at a national level "for policy and planning purposes. Our study may also be useful for informing guidelines on the type and dose of statins."

But she feels the Heart paper is more important in terms of determining "absolute risk estimates" for statin use based on individual patient characteristics. This will help answer the individual's question of "What's my risk?" she noted.

In this paper, she and Coupland developed risk-prediction scores using the results from the QResearch database to determine five-year risk of moderate/serious myopathic events, moderate/severe liver dysfunction, acute renal failure, and cataracts.

Doctors will be able to input into the algorithm different patient characteristics such as age, gender, ethnic group, comorbidities, and concomitant medications in order to get an absolute risk assessment of each individual's likelihood of certain adverse events from statins.

They found the performance of the algorithms was "very good" for three of the four they developed--acute renal failure, myopathy, and cataracts--and could be used to identify patients at increased risk of these adverse events, "enabling patients to be monitored more closely." But further research is needed to develop a better algorithm for the fourth one, to predict liver dysfunction, they conclude.

Commenting on the Heart paper, Alsheikh-Ali told heartwire this "elegant work" by Hippisley-Cox and Coupland "refines our understanding of adverse events with statins but is not likely to change clinical practice.

"While we are confident that the estimated/predicted CVD risk reduction is causally related to statins, we are not at all confident that the predicted increases in the four outcomes . . . are causally related to statins, for the reasons acknowledged by the authors, most notably residual confounding and ascertainment bias," he concluded.