Prospective, controlled study
To conduct their research, Michaud and Izard drew on medical records and preserved blood samples collected by the Imperial College-led European Prospective Investigation into Cancer and Nutrition Study, a massive dataset of more than 500,000 adults in 10 countries. Detailed health histories and blood samples are available from more than 380,000 of the participants.
From that population, the researchers found 405 people who developed pancreatic cancer, but no other cancer, and who had blood samples available. The researchers also selected 416 demographically similar people who did not develop pancreatic cancer for comparison.
The researchers blinded themselves to which samples came from cancer patients and which didn't during their analysis of the blood, which consisted of measuring antibody concentrations for 25 pathogenic and commensal oral bacteria. In their study design and analysis they controlled for smoking, diabetes, body mass index, and other risk factors.
An important element of the study design was that date of the blood samples preceded the diagnosis of pancreatic cancer by as much as a decade, meaning that the significant difference in antibody levels were likely not a result of cancer.
Instead, the underlying mechanisms that link Porphyromonas gingivalis to pancreatic cancer could be causal, Michaud said, although much more research is needed to understand this association.
Meanwhile, the researchers speculate, the association of high levels of antibodies for commensal bacteria and pancreatic cancer, may indicate an innate, highly active immune response that is prote
|Contact: David Orenstein|