Periodontal Disease Linked to Certain Cancer Types

Jean Wactawski-Wende, PhD

Today we are speaking with Jean Wactawski-Wende, PhD, an epidemiologist at the State University of New York at Buffalo who, along with colleagues, just came out with a study that links periodontal disease to risk of certain types of cancers. The study was published in the Cancer Epidemiology, Biomarkers & Prevention.

-Interviewed by Anna Azvolinsky 

Cancer Network: Prior to your prospective cohort study, what was known about the link between periodontal disease and the risk of certain tumor types?

Wactawski-Wende: Actually, there had been a few studies. One was exclusively in females and had looked at cancer and periodontal disease, and the most important association the authors found was for an increase in pancreatic cancer risk. There had been some additional smaller studies that had a slightly different design, they were case-controlled studies that looked at oral cancers, but this was the first study in older women, and by far the largest study that has been done.

Cancer Network: What was the question you and your colleagues set out to investigate, and can you tell us about the design of the study?

Wactawski-Wende: We were interested in the role of periodontal disease in incidence of cancer. We have an ongoing study called the Women’s Health Initiative that initially enrolled participants in the early- to mid-1990s, and we have been following this cohort of women now for almost 25 years. And as part of that study, we included a series of questions about oral health including a question that we used here, “Have you ever been told by your dentist that you have periodontal disease or gum disease?”

We also asked about frequency of dental visits and whether the participants had all of their teeth or not, but we primarily used that question, “Has a dentist told you that you have periodontal disease?” So we were able to go from the point of knowing that and prospectively following these women on many health outcomes, including asking whether the women had been diagnosed with cancer, and we were able to get medical records for those with cancer to verify the answers. In this group of a little over 65,000 women, there were over 7,000 new incidents of cancer that occurred.

This study then looked at the association between total cancer in women with or without periodontal disease history as well, as we had a large enough group to look at specific cancer types.

Cancer Network: What were the important findings and were there any that particularly surprised you?

Wactawski-Wende: The most important finding was a total cancer increase when we looked at all cancers combined-a 14% increase in women that reported a history of periodontal disease. Although the relative risk of 14% may not seem like a huge amount, when you think about the entire population and the fact that probably half of women over the age of 65 have moderate to severe periodontal disease, that 14% increase would be significant in terms of population health.

We also were able to demonstrate a statistically significant increase for a number of specific cancer sites-lung cancer, breast cancer, esophageal cancer, gallbladder cancer, and malignant melanoma of the skin. We also saw a borderline increase in stomach cancer in this group.

What would be surprising? Well, I think this is the first time that gallbladder cancer has been shown to be increased. We had a previous publication on lung and breast cancer so we knew those would be elevated.

One of the more important things about this study is that we were able to look at these increased risks and control for many other factors that are also associated with periodontal disease that might account for the increased cancer risk. One very important factor is smoking, and so the models-the risk estimates that we put forward in the paper-controlled for smoking, statistically. We were able to look specifically in the women who reported being smokers and those who had never smoked, and the results of the study held even in the never-smokers, so we can say that this association that exists is independent of any exposure to cigarette smoke.

Cancer Network: Your study used at least some self-reported data from individuals. Do you view that as a limitation for the conclusion you reached?

Wactawski-Wende: This is self-reported data because it is difficult to do exams in as many as 65,000 women. But we do have a smaller study (OsteoPerio) at our University at Buffalo where we actually brought in women that participated in this same larger study and did a detailed examination of their oral cavity-assessed pocket depth and bone height, and all the other ways that you get a detailed assessment of periodontal disease-so we were able to use that to help validate the self-reporting. What we found is that when people do report periodontal disease, they are generally accurate in reporting it, but that it may be underreported in women who have more mild and moderate types of periodontal disease. So, if anything, we may have underestimated the findings.

Cancer Network: Just lastly, are you or other colleagues now further addressing this potential link? You mentioned this smaller study you are doing, are you looking to start any new studies to investigate this link further?

Wactawski-Wende: We have some interesting work going on right now. As I mentioned, we have this study called the OsteoPerio Study in Buffalo, and in that study, we have a little over 1,000 women who we followed and did oral exams 17 years ago, 12 years ago, and also over 500 now at 17 years post baseline. We are doing the periodontal exam to look at the extent of periodontal disease, but as part of that, we collected oral plaque. The study we have funded right now is actually looking at the bacterial composition in the oral cavity at those three time points in the women with and without periodontal disease, so we will be able to characterize the type of bacteria. And because we followed cancer in these women, what we are piloting right now-we don’t have funding yet to do the larger study-but we are going to collect the tumor blocks from their breast and lung tissue, and look at the bacterial composition in those tumor blocks. So, we will be able to look at whether those bacteria that are in the oral cavity are similarly found in the tumor blocks. That is our ultimate plan, but we are just actually piloting that study right now so that we can apply for National Institutes of Health funding to help support that grant.

Cancer Network: Thank you so much for joining us today, Dr. Wactawski-Wende.

Wactawski-Wende: Thank you, it was a pleasure to talk to you.