The association between total and integrated HPV-33 DNA loads and high-grade squamous intraepithelial lesions (HSIL) of the uterine cervix was investigated. Of 5,347 women recruited in 4 studies, 89 (64 without SIL, 7 low-grade SIL (LSIL), 15 HSIL, 3 unknown grade) were infected by HPV-33. HPV-33 E6, HPV-33 E2 and beta-globin DNA were measured with real-time PCR that allowed to assess total (E6), episomal (E2) and integrated (E6-E2) HPV-33 viral loads. HPV-33 E6/E2 ratios >/=>/=2.0 suggesting the presence of integrated HPV-33 were obtained for 28.6% (n = 18) of women without SIL and 21.4% (n = 3) of women with HSIL (p = 0.74). Although median viral loads were similar, there was a trend toward having a greater proportion of women with HSIL in the fourth quartile (>/=>/=10(6.69) copies/mug DNA) of total HPV-33 viral loads compared to normal women. Controlling for age, site, ethnicity and LCR polymorphism by logistic regression, HPV-33 total loads in the fourth quartile {odds ratio (OR) 4.5 [95% confidence interval (CI) 1.2-17.3]; p = 0.03} and episomal loads in the fourth quartile (>/=>/=10(6.64) copies/mug DNA) [OR 3.9 (95% CI 1.1-13.2); p = 0.05] but not integrated HPV-33 load in the fourth quartile [OR 1.0 (95% CI 0.3-3.3); p = 0.50] were associated with HSIL. Controlling for age, study site and SIL grade, HPV-33 episomal load [OR 0.2 (95% CI 0.1-0.5), p = 0.0004] was associated with the presence of HPV-33 integration. High episomal loads in HSIL and the presence of integration in women without SIL are likely to weaken the usefulness of HPV load of integrated forms in clinical practice.
(c) 2007 Wiley-Liss, Inc.