Smokers at Higher Risk for Undetected Antibody for Oncogenic Human Papillomavirus Type 16 Infection

Abstract

Objective: To determine the association between tobacco smoking and serologic evidence of human papillomavirus type 16 (HPV16)–specific antibodies among HPV16 DNA–positive women.Design, Setting, and Participants: Baseline health history, physical examination, and laboratory data for 205 HPV16 DNA–positive women with no prior cytologic evidence of squamous intraepithelial lesions who were enrolled subsequently in a randomized clinical trial.Main Outcome Measure: HPV16-L1 antibody (anti-HPV16 antibody) detected from serum using RIA or ELISA.Results: Eighty-seven percent (179 of 205) of women tested positive for HPV16 DNA using cervicovaginal swabs or lavage specimens, and 26 women showed similar results using swab specimens of external genitalia alone. HPV16-infected women who reported increasingly greater levels of daily cigarette smoking were less likely to test positive for anti-HPV16 antibodies than nonsmoking women (P = 0.02). Smokers were twice as likely as nonsmokers to test negative for anti-HPV16 antibodies, even after controlling for the effects of other covariates in the analyses (adjusted odds ratio, 0.5; 95% confidence limits, 0.2-0.9). Although Papanicolaou test findings and smoking characteristics were poorly correlated (r2 = 0.01), women who showed atypical cells of unknown significance or squamous intraepithelial lesion were twice as likely to test anti-HPV16 antibody positive as women who showed normal Papanicolaou tests (adjusted odds ratio, 2.0; 95% confidence limits, 1.1-3.7).Conclusion: These data suggest that smoking may influence the long-term risk for cancer by perturbing early immune responses to the virus and may increase the likelihood of persistent infection. Patient education messages should alert women to this additional risk of smoking. A clinical trial of smoking cessation should be explored as a therapeutic intervention for primary HPV16 infection. (Cancer Epidemiol Biomarkers Prev 2006;15(5):915–20)