How Skeptical Epidemiologists Figured Out That Cholesterol Drugs Can Help Prevent Prostate Cancer
Alison Mondul
Assistant Professor of Epidemiology
Epidemiologists are trained skeptics. One of the first lessons in introductory epidemiology classes is that we are studying associations and that there are many reasons something might be associated with a disease without actually causing it. We want to see an association present in lots of different populations—and be backed up by other kinds of science—before we believe it is real. Ultimately, the goal of epidemiology research is to find out what causes diseases.
One of my research topics—statins and prostate cancer—illustrates this well. Statins are a type of medication commonly prescribed to treat high cholesterol. About 30 percent of older adults in the US are taking these drugs. They are extremely effective at lowering cholesterol and, thus, preventing cardiovascular disease.
When we first started studying men who used statins, we found that they were less likely to be diagnosed with advanced prostate cancer. This was exciting.
But statins also do lots of things that might prevent cancer. I study prostate cancer, which is the most commonly diagnosed cancer and the second leading cause of cancer death in men in the US. In the early 1990s, a blood test called the PSA test was approved as a screening tool for prostate cancer. This test is very good at helping us detect cancer—almost too good. We now know, after years of research, that many of the cancers we detect using the PSA are not aggressive and would not cause problems for a man in his lifetime if left alone and not treated.
A good deal of research is underway to determine which cancers need treatment and which don't, and it is still a work in progress. However, when we are studying prostate cancer as epidemiologists, we don't want to study those indolent cases—the ones that don't need treatment. We want to know how to prevent cancers that will cause harm. One way we pick out the relevant cases is to study only those diagnosed at a more advanced stage. These cases have a poorer prognosis and are less likely to be the indolent cases that don't need treatment.
But we are skeptics. We realized that statins are prescription drugs, which means all the men taking them must have gone to a doctor.
When we first started studying men who used statins, we found that they were less likely to be diagnosed with advanced prostate cancer. This was exciting. It seemed we could use a medication already known to be safe, generally well-tolerated, and beneficial in treating other chronic diseases to also prevent cancer.
But we are skeptics. We first realized that statins are prescription drugs, which means all the men taking them must have gone to a doctor. Which means they might be more likely to have had a PSA screening test than men who aren't taking a statin. And the whole point of a screening test is to catch cancer early, before it reaches an advanced stage. Men taking a statin might be more highly screened and less likely to have advanced prostate cancer simply because the overlap between getting a statin and a PSA test is based on going to the doctor regularly—not because statins were actually preventing bad prostate cancer.
We became concerned that our findings were not real, that they were the result of what epidemiologists would call a bias—incorrectly thinking a particular exposure causes a disease. At that point, we wanted to study the association between statins and prostate cancer in the calendar period before PSA screening was used. But both statins and PSA screening became available at almost the same time (1987 for statins, 1992 for PSA), so there was no way to do that.
Men who are taking a statin to lower their cholesterol can feel good knowing that their cholesterol medicine is also likely helping their prostate health.
We had to get creative. The PSA test is not widely used in Europe, so our European colleagues studied this same PSA-statin association in their data and got similar results to ours. This was encouraging. Our laboratory science colleagues studied what statins do to cancer cells in a petri dish and found that they appear to stop cancer growth and progression through several biologic mechanisms—also encouraging. Finally, we conducted additional studies that found that longer duration of statin use increased the protective association against prostate cancer and that statins even protect against death due to prostate cancer.
The first paper reporting that statins might protect against prostate cancer was published in 2006. Twelve years later we continue to believe that statin drugs are protective for prostate cancer. Now we must determine how to use this information for prostate cancer prevention. Giving all men a statin is probably not the right answer. Though they are generally safe, some men experience side effects from statins, and not all men need to take one. We are now looking at who might benefit most from statins for prostate cancer prevention and when statins need to be taken to prevent prostate cancer. Could they be given to men after diagnosis to prevent bad outcomes? Or should they be taken before diagnosis to have the most benefit? How long would men need to take them to get the most benefit? Could they be useful in preventing other types of cancer as well?
These are ongoing research questions. In the meantime, men who are taking a statin to lower their cholesterol can feel good knowing that their cholesterol medicine is also likely helping their prostate health.
About the Author
Alison M. Mondul is assistant professor of Epidemiology at the University of Michigan
School of Public Health. She received her MSPH in Epidemiology from Emory University
in 2002 and her PhD in Cancer Epidemiology from Johns Hopkins University in 2009. She then spent five
years as a post-doctoral fellow in the Division of Cancer Epidemiology and Genetics
at the National Cancer Institute before joining the Department of Epidemiology at
Michigan Public Health. Mondul has devoted her scientific career to studying the role
of modifiable risk factors in the etiology of cancer. In particular, she has studied
how lifestyle factors, such as use of common medications, and factors related to diet
and nutrition may influence prostate, bladder, and kidney cancers as well as other
genitourinary conditions. More recently, she has become involved in studying head
and neck cancer, which remains an understudied area in cancer research