Athens, Greece - The largest meta-analysis to date investigating statins and cancer risk has found no decrease in the risk of cancer and concludes that these drugs do not appear to offer a protective effect [1].

This latest meta-analysis, published online in the Journal of Clinical Oncology on September 25, 2006, was performed by Dr Stefanos Bonovas (University of Athens, Greece) and colleagues. Based on data from 109 143 patients involved in 35 randomized clinical trials, it is "the most extensive and inclusive meta-analysis of its kind," comments an accompanying editorial [2].

The finding of a lack of effect of statins on cancer risk ties in with all other meta-analyses that have previously been carried out on this subject, including a large one published earlier this year in the Journal of the American Medical Association [3]. All have reported a negative association, the editorial notes.

In contrast, several epidemiological studies have found a positive effect of statins on reducing the risk of cancer, the editorial points out, with a decrease either in the overall cancer incidence, or in site-specific cancers. One study that attracted a lot of attention, published last year in the New England Journal of Medicine [4], found a 47% decrease in the risk of colorectal cancer, based on a case-control study in Israel.

However, other epidemiological studies have found no effect of statins on cancer risk, and a handful have even found an increase in the risk of cancer, notes the editorialist, Dr Kyungmann Kim (University of Wisconsin, Madison).

"The only sensible conclusion one can draw at this stage appears to be (1) that the association between statin use and cancer risk is inconclusive at best, and (2) at worst, there are no effects at all," he writes.

Where does this leave the clinical community? Kim asks. He notes that there are several early-stage trials already exploring the use of statins, both as treatment (in multiple myeloma and metastatic colorectal cancer) and as chemoprevention (for breast cancer, colorectal neoplasia, and a trial about to start in dysplastic nevi).

"If - and a big if at that - these studies provide reliable preliminary data, the obvious and simple solution for the dilemma would be to conduct a definitive randomized controlled trial or two," Kim continues. However, the window of opportunity to conduct definitive randomized trials appears to be narrowing rapidly, because the widespread use of these drugs will make it difficult to identify suitable patient populations and will pose challenges of probable contamination. "Perhaps the genie is already out of the bottle," he muses.