Can Statins Lower The Risk Of Cancer?
14 CommentsBy Ed Silverman // January 9th, 2008 // 8:00 am
Statins may cut the risk of developing cancer by as much as 25 percent, according to research reported in the Journal of the National Cancer Institute. Veterans taking statin drugs had a 9.4 percent cancer incidence, compared to 13.2 percent for non-statin users.
“Our findings support the hypothesis that statins may reduce the risk of cancer, in particular lung and colorectal cancers,” Wildon Farwell of the Veterans Affairs Boston Healthcare System, who led the study, tells Reuters. “The risk reduction appeared to be around 25 percent.”
For their study, Farwell and colleagues looked at the health records of nearly 63,000 veterans in the Veteran Affairs New England Healthcare System between January 1997 and December 2005. The veterans were divided into groups that had used either statin drugs, including Lipitor and Zocor, or blood-pressure meds for at least one year.
These groups were chosen because patients on both types of drugs have similar health risks and are likely to get about the same amount of access to the healthcare system. After adjusting for age, prior cancer screenings, smoking, lung disease and other conditions, the researchers found statin users had a reduced risk of all cancer types compared with those not taking statins.
The researchers also looked at five of the most common types of cancers in the study group: prostate, lung, colorectal, bladder cancer and melanoma. And they found “signiificant risk reduction for prostate, lung and colorectal cancer,” says Farwell. Moreover, there was an intereting correlation - the higher the statin dose, the lower the incidence of cancer.
However, the study didn’t show why statins seemed to lower the risk of cancer and had a few limitations - participants were mostly white males, which could skew results. “We don’t want to give the impression that this is the definitive study that proves statins reduce the risk of cancer,” Farwell says, but he believes the findings are compelling enough to warrant further study.
Mark
High cholesterol levels are known to be associated with a reduced cancer mortality, e.g.
http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=1836205
Higher cholesterol levels are also associated to increased statin usage. So where’s the beef?
In controlled clinical trials, there was either no association between statin use and cancer or even an increased cancer risk with statins. We have seen enough of this marketing crap.
Hank
As reported last year, two larger studies - including one meta-analysis - suggested no correlation between statin use and reduced colon cancer (one study) and other cancers as well (other study). See
http://www.cancer.org/docroot/NWS/content/NWS_1_1x_Studies_Statins_Dont_Cut_Risk_of_Colon_Other_Cancers.asp
These were partly in response to an earlier study which was reported in much the same way as the most recent.
Further precincts. As far as I know, the absolute risk change has been quite small in all studies.
nipsey russell
“High cholesterol levels are known to be associated with a reduced cancer mortality….Higher cholesterol levels are also associated to increased statin usage. So where’s the beef?”
um, huh???
Atlex
Mark, do you have any clue what you are talking about? Statins lower cholesterol and don’t raise them. Moreover, if you think this is marketing crap, consider the source of the study–the VA. The VA has little motivation to support the marketing needs of pharma companies. And, since the association is not related to any single statin, but to the entire class, the biggest benefit will likely be to the generic products and not brands since no brand will be able to promote this unless they get it in thier label (which is a very unlikely event).
Mark
@nipsey: Imagine that people with high cholesterol levels get prescribed pink, triangular sugar pills. Now perform the study described above. Result: Pink, triangular sugar pills lower the risk of cancer. This also works with medications that lower the cholesterol level, as long as you don’t assume that there is a direct causal relationship between the current cholesterol level and the cancer risk. Consider a common cause for high cholesterol and low cancer risk, for example.
@Atlex: Pfizer is making 13 billion dollars a year with Lipitor. Pushing the class will help Pfizer sell Lipitor, and it will help Merck sell Zocor etc. This type of observational study (Do Statins lower the risk of XXXX cancer, Alzheimer’s, Parkinson’s, fill in whatever you like) has been in the news every two months for maybe ten years now, although there is overwhelming evidence from a large amount of controlled clinical trials (not observational studies!) that show that there is no protectice effect of statins against cancer, Alzheimer’s, Parkinson’s etc.
Hank
Agree with most points made. Still, “interesting” the ways potential good news gets reported, and the countervailing studies virtually never are. This for those folks who think reporting about pharma is always one-sided in the other direction.
Mark
Let me add that the pink, triangular sugar pills will also provide you with the “interesting effect” observed in the study: The higher the pink triangular pill dose, the lower the incidence of cancer. Simply because doctors will tend to prescribe a higher dose of pink, triangular sugar pills to people with higher cholesterol levels.
Brian
If it is better data you seek..It can be obtained with bigger studies. After all, all drug side effects are uncovered eventually and in all cases they are a small small percentage of total patients. When only a few patients in a study show an effect similar to placebo, it is hard to draw any real conclusions. You might want to take a statistics course at your local community college. The trouble with bigger studies is that you don’t want to pay for them and you don’t want to volunteer to participate. Luckily you weren’t around when surgery was invented during the civil war. You owe a debt of gratitude to those patients who were the subjects of very early medicinal intervention.
Mark
@Brian: I don’t think you see the point I am trying to make.
Observational studies never prove causal relationships. If there is no randomized trial available, observational studies might give you a hint where to go with your research.
In this case, there is a huge amount of data from controlled, randomized, double-blind trials at hands, which in this case includes 132.000 patients and clearly shows that the relationship the authors are trying to establish is not there (check Hank’s link above). Therefore it does not make any sense to perform an observational study on this question.
In addition, my pink triangular pill example explains why the results from this study can be expected with any kind of placebo or medication that is prescribed to people with high cholesterol, simply because those are the people with a low cancer risk.
Brian
So the current observation suggests a trial in patients with the correct demographic..That’s progress. I don’t agree that patients with elevated lipids have an decreased incidence of cancer
Atlex
Observational studies such as this are great at identifying hypotheses for further study. Had the observational study turn out the opposite conclusion–for instance that statin users had twice the rate of cancer–this study would have had a huge impact, probably leading to a black box or pulling these drugs off the market. Fortunately, it turned out that statins are associated with lower cancer rates. Could a manufacturer leverage htis in some way…I guess its possible, but highly unlikely. They would need to complete a large scale prospective trial that would take years to complete. Since Zocor is already generic and Lipitor loses exclusivity in 2011 or so, the chances of either Merck or Pfizer benefitting is remote.
As a statin user, I now have even more confidence in my treatment. Not only am I lowering my CV risk, I am more comfortable that these drugs are safe, and perhaps, could even provide some vague other protections. That being said, I don;t think any reader of this study would suddenly take a statin only for its cancer protection.
Hank
For the sake of clarification, I believe at least one of the earlier studies - not suggesting a benefit re: cancer -was also observational, very large and well-controlled, as described. I am also a believer in the importance of well designed observational studies. They have provided important signals - in all directions - over the years.
The latest study makes note of the 2006 studies suggesting no cancer benefit for statins. The authors suggest that the subjects may have been too young, or the follow-up period too short, to show the impact that the latest study suggests.
No investigator, however, is suggesting that we now have “the grail” on this.
Mark
“The authors suggest that the subjects may have been too young.”
By the way: In real life, statins are typically prescribed to elderly patients.
To my knowledge, there was only a single controlled, randomized, double-blind statin trial with elderly patients. Elderly patients have a significantly higher baseline cancer risk than the typical statin trial participants.
Any statistically significant relationship between statin use and cancer that might have been found in this population is much more significant than all observational studies that have been performed on this question so far, wouldn’t you agree?
And guess what? “New cancer diagnoses were more frequent on pravastatin than on placebo (1.25, 1.04-1.51, p=0.020).”
Hank
Mark - My understanding of the statin studies in incomplete. My “impression,” however, is that the evidence for primary prevention for CV is weakest for older patients (65+), perhaps because they die of other things. In older women, there has been no primary CV benefit shown at all, as far as I know.
I also don’t know what the median age is for statin use. Given the changing NCEP guidelines (which is a whole other discussions), it is certainly getting younger. I know a lot of people in their 40s who take them. And by 50 it’s almost “de rigeur.” So I’m not sure what you mean by “real life” and “elderly” in this context.
The VA trial was not an RCT, correct. I do not agree that carefully designed observational trials cannot be extremely important. Jerry Avorn of Harvard has made the argument many times, infinitelely better than I could.
As one example, David Graham’s much kicked-around Vioxx/Celebrex Kaiser study - which certainly scared the hell out of his FDA directors (as evidenced in their emails to each other) - was an observational study. Was it as convincing as VIGOR? Probably not, but VIGOR was spun in a hundred ways, as we all know.
We also know from documents that have come out that Bayer’s own observational study on Baycol showed danger signals that were recognized as serious inside the company. But when some of this data came out, they were dismissed as “merely observational studies.”
So, whatever side one is on re: whatever issue, I don’t think “gold standards” should negate what may be high quality sterling.