Assessment of risk factors and human papillomavirus (HPV) related pathogenetic mechanisms of CIN in HIV-positive and HIV-negative women. Study design and baseline data of the HPV-PathogenISS study.Eur J Gynaecol Oncol 2004; 25(6):689-98EJ
In women with HIV-associated immunosuppression, HPV infections have an increased risk of progression to high-grade cervical intraepithelial neoplasia (CIN). With the HAART-induced prolonged survival and more protracted clinical course of AIDS, progression of CIN to cervical cancer (CC) has become a clinically relevant issue, and the mechanisms responsible for HIV-HPV interactions need further elucidation. The study design and analysis of the baseline data of our new project are presented.
MATERIAL AND METHODS
This project is a combination of a prospective cohort study of HIV- and HIV+ women, and a retrospective analysis of CIN lesions and cervical cancer. Up to the present, 244 women have been enrolled (17 HIV+) and subjected to epidemiological interview, colposcopic examination, sampling for HPV testing and typing (PCR, InnoLiPA), and HPV serology. The retrospective series of biopsies were analysed for 13 biomarkers (monitoring key molecular events) using immunohistochemistry and tested for HPV by PCR and TaqMan.
HIV- and HIV+ women differ in their exposure status to many of the key epidemiological risk factors of cervical cancer, the most significant ones being number of sexual partners (p = 0.0001), age at onset of sexual activity (p = 0.002), and contraception (yes-no) (p = 0.009). The differences in the baseline clinical observations are less dramatic; HIV-positive women had more frequent HSIL PAP tests (p = 0.040), CIN2 or higher in cervical biopsy (p = 0.049), and external genital warts (p = 0.019). The factors predicting intermediate endpoint markers of cervical cancer, i.e., HSIL PAP smear, ATZ2 in colposcopy, and high-grade CIN in biopsy were analysed in univariate and multivariate regression models. All factors significant in univariate analysis were entered in the multivariate model; HIV-status and Pap smear history maintained their independent predictive power of the HSIL Pap test. The most powerful predictor of ATZ2 colposcopy was HSIL in Pap test. Only the HSIL Pap test and ATZ2 colposcopy remained significant independent predictors of high-grade CIN (p = 0.0001 and p = 0.008, respectively) in the multivariate model.
The three intermediate endpoint markers are closely interrelated, but predicted in part by different covariantes in the causal pathway to cervical cancer. To elucidate whether the increased risk of HIV-positive women to high-grade CIN is due a) to their different exposure status to the risk factors, b) to the direct effects of HIV, or c) to molecular interactions between HIV and HPV, we need to complete these analyses separately in HIV+ and HIV- women.