Author + information
- Jean-Luc Prétet, PhD⁎ (, )
- Mariette Mercier, PhD,
- Didier Riethmuller, MD, PhD,
- François Aubin, MD, PhD,
- Dominique Vuitton, MD, PhD and
- Christiane Mougin, MD, PhD
- ↵⁎Laboratoire de Biologie Cellulaire et Moléculaire, Bâtiment Diaclone, Hôpital J Minjoz, 25000 Besançon, France
We read with great interest the report by Hsu-Ko Kuo and Ken Fujise (1) titled “Human Papillomavirus and Cardiovascular Disease Among U.S. Women in the National Health and Nutrition Examination Survey, 2003 to 2006,” recently published in the Journal.
The investigators report an association between the presence of human papillomavirus (HPV) deoxyribonucleic acid in the vagina and the risk for cardiovascular disease (CVD) in a population of women who participated in a national health and nutritional survey. They made the assumption that after HPV infection, the 2 major viral oncoproteins, E7 and E6, through their pRB and p53 degradation properties, may favor the development of atherosclerosis. The investigators argued that knock out for either p53 or pRB atherosclerosis development was accelerated in mice models. However, we would like to make some points that should be discussed in greater depth.
First, the overall HPV prevalence was high in the cohort. Kuo and Fujise (1) reported HPV deoxyribonucleic acid in 48% of women, 25% being infected by high-risk HPV, with a mean age of 35 years. Large epidemiological studies have reported a high HPV prevalence (up to 30% to 40%) in young women (age 20 to 24 years), while the prevalence decreases to 10% or less in women older than 35 years (2,3).
Second, women who were considered “nonresponders” were reportedly different from women included in the analysis. The investigators indicated that they were younger and thus more likely to harbor HPV deoxyribonucleic acid but probably less likely to present CVD than the “responders.”
Third, of the 44 women with cervical cancer listed in the their Table 1 (1), 22 were HPV negative. This is not consistent with the fact that HPV is the etiologic agent for cervical cancer.
Fourth, in their Table 3, Kuo and Fujise (1) presented the odds ratios (ORs) for CVD comparing women with cancer-associated HPV types or other HPV types with those who are negative for HPV. In the first model, the OR was 2.87 for women with cancer-associated HPV types. The OR was 2.13 for women with other HPV types, namely, low-risk HPV types that are not able to induce p53 and pRB degradation. If the physiopathology hypothesis is correct, we would expect a nonsignificant OR. As for models 2, 3, and 4, we would also expect lower ORs for CVD in women with other HPV types.
Finally, the overall physiopathology hypothesis is very unlikely. Indeed, HPV fails to produce classic viremia. Rather the virus replicates locally until it is cleared by an efficient immune response. In case of viral persistence, the virus remains localized to the site of infection, and associated lesions can occur after an increased expression of E6 and E7. The virus does not disseminate throughout the body. It is therefore very unlikely that E6 and E7 have a systemic effect leading to atherosclerosis.
- American College of Cardiology Foundation