Melanoma chemoprevention theory takes step backward

March 1, 2006

It looked at statin use and cancer incidence in general, and found the drugs failed to statistically improve cancer rates across the board.

Denver - Melanoma chemoprevention, the hypothesis that certain medications might decrease the risks of developing melanoma in high-risk patients, was not proven in a new meta-analysis.

"The hope was that we could offer people with a strong family history of melanoma something more than just advice to use sunscreen and stay out of the sun," says lead author Robert Dellavalle, M.D., Ph.D., M.S.P.H., associate professor of dermatology, University of Colorado at Denver and Health Sciences Center.

The trial, according to Dr. Dellavalle, that fueled melanoma chemoprevention thinking was the randomized, double-blind, placebo-controlled Air Force Coronary Atherosclerosis Prevention Study (AFCAPS), published in the Journal of the American Medical Association in May 1998. Down JR, et al. compared lovastatin with placebo for prevention of a first acute major coronary event in men and women, who had not been previously diagnosed with heart disease.

"While there have been lots of statin and fibrate trials, only a few reported the numbers of melanomas that had occurred. So, we identified the big, high-quality trials with statins and fibrates (which have also been reported to lower melanoma incidence), then wrote to the authors of these studies to give us their unpublished data on melanoma," Dr. Dellavalle says.

Authors of 13 trials (six statin, seven fibrate) involving 62,197 participants provided data on incident melanomas. The researchers of these trials reported a total of 66 melanomas among those receiving the experimental drug and 86 in groups receiving placebo or other control therapies. For statin trials, this translated to an odds ratio of 0.90 and for fibrate trials an odds ratio of 0.58.

Dr. Dellavalle and colleagues reported no statistically significant differences in melanoma outcomes when they analyzed subgroups based on gender, melanoma occurrence after two years of trial participation, stage or histology or trial funding. Their analysis of subgroups by fibrate or statin also failed to show statistically significant differences, except for the statin subgroup analysis, which showed reduced melanoma incidence for lovastatin, based on one trial, with an odds ratio of 0.52, according to the abstract.

The authors concluded that their meta-analysis - while it did not exclude the possibility that statins and fibrates could prevent melanoma - did not show statistically significant numbers to prove the hypothesis.

"These studies, collectively, were not as promising as the initial, individual reports," Dr. Dellavalle says.

Another meta-analysis, published January 2006 in JAMA, found similar results. It looked at statin use and cancer incidence in general, and found the drugs failed to statistically improve cancer rates across the board.

"That meta-analysis pretty much put the nail in the coffin of this hypothesis," Dr. Dellavalle says.

He notes that despite the evidence to the contrary, some still believe that statins might play a role in melanoma prevention in some circumstances. Nevertheless, the current bottom line is that dermatologists should continue to preach sun avoidance and protection to patients as melanoma prevention practices, and should not prescribe statins or fibrates solely for melanoma prevention in high-risk patients.

Other unproven candidate medications for melanoma chemoprevention are Accutane (Roche), which Dr. Dellavalle says is very controversial, and imiquimod and imiquimod-related medications, which boost the immune system.

Disclosure: Dr. Dellavalle reports no conflicts of interest relevant to this article.

For more information: This information was published in CochraneDatabase Syst Rev. 2005 Oct 19;(4):DC003697.