Sunday, May 29, 2005
People who took a type of cholesterol-lowering drug for five years had nearly half the risk of developing colon cancer, even when they had a family history of the disease or other risk factors, according to researchers at the University of Michigan Comprehensive Cancer Center.
..."This is an important piece of the puzzle. This piece helps bring together evidence that statins may have the potential to prevent chronic diseases other than heart disease, and helps us consider ways to study these powerful drugs for more than one purpose," says study author Stephen Gruber, M.D., Ph.D., M.P.H., associate professor of internal medicine and human genetics at the U-M Medical School and associate professor of epidemiology at the U-M School of Public Health.
Data was based on the Molecular Epidemiology of Colorectal Cancer study, a population-based case-control study of colorectal cancer in northern Israel. The researchers analyzed 1,953 people with colorectal cancer and 2,015 control subjects who did not have colon cancer. All study participants were asked to recall every medication they had used for at least five years. Statin use was determined based on this list and validated against prescription records from the health care provider.
The people without colon cancer were nearly twice as likely to report taking statins for at least five years, compared to the people with a history of colon cancer, 11.6 percent vs. 6.1 percent.
Researchers analyzed the data taking into account increased or decreased risk associated with demographic factors such as age, sex and ethnic background, and with lifestyle factors such as taking aspirin, participating in sports and eating vegetables. They also considered family history of colon cancer, which increases a person's risk.
Even considering all these additional factors, statins were still associated with a 47 percent reduced risk of colorectal cancer.
That last statement doesn't strike me as valid. It would be valid if the researchers started out with two matched groups of people who had no colon cancer then had one group take statins every day and one group take none. Then, if the incidence of colon cancer in the statin-taking group were 47% less than in the non-drugged group, they could make that claim.
But that's not what they did. Instead, they started out with a group of patients who already had colon cancer and compared them to people who didn't (yet.) Then, they just asked each group a series of questions to see what might be different about them. This in itself is a flawed methodology to find any conclusive evidence of cause and effect. The results depend on the questions the researchers choose to ask, for one thing. There could be a factor that separates the two groups more signficantly than the use of statins but that the researchers never thought to ask.
But there's an even greater flaw in this study. The researchers didn't bother to match their two groups of patients - who all lived in northern Israel - by ethnic group. They matched them by sex and age, and they made sure that they had close to the same number of Muslims, and non-Christian Europeans, but the two groups differ markedly in the ethnicity of Jewish subjects. Not surprisingly, since it was done in Israel, Jewish patients made up the preponderance of the subjects. However, 68.4% of the colon cancer patients were Ashkenazi Jews compared to 62.5% of the patients without colon cancer - a difference of 6 percentage points. Conversely, 24.7% of the patients without colon cancer were Sephardic Jews, compared to 17.7% of those who had colon cancer - a difference of 7 percentage points. By comparison, 6.1% of the cancer patient group took statins for five years compared to 11.6% of the non-cancer group - a difference of 5 and a half percentage points. It would appear that being of Ashkenazi Jewish descent is at least as much of a risk for colon cancer as statin use.
This is significant, because Ashkenazi Jews are one of the most genetically homogenous ethnic groups in the world. They also have a higher than normal risk of colon cancer. To claim that statins reduce the risk of colon cancer based on this study is more than exaggeration - it's just downright absurd.
Now, if the researchers really wanted to see if statins prevent colon cancer, they could start out with two groups of equally matched Ashkenazi Jews, give one group statins and have the other go without then count the number of colon cancer cases in both groups over a period of years. That would require a lot more time and patience on their part, but the results would be more valid than what they have now.
posted by Sydney on 5/29/2005 12:04:00 PM 0 comments