Invasive cervical cancer (ICC) is an AIDS defining disease [CDC Update, 1993]. Our studies show that women with HIV-2 or HIV- I infection are at increased risk of ICC and development of cervical intraepithelial neoplasia 3/carcinoma in situ (CIN 3/CIS). Screening programs based upon cytology, detection of human papillomavirus (HPV), and visualization of the cervix have all been proposed for use in resource poor settings, however, none of these methods are practical approaches to cervical cancer control in areas of endemic HIV infection. Novel approaches must be developed to identify and treat women at high risk for ICC in HIV endemic areas. We hypothesize that, in contrast to most HIV seronegative women who are infected with oncogenic types of HPV; HIV seropositive women co-infected with oncogenic types of HPV acquired the molecular changes which permit, and are associated with progression to ICC even before cytologic changes are detected, or when only mild changes such as CIN 1 are present. We further hypothesize that such molecular changes, which are predictive of increased ICC risk can be identified and could serve as the basis for sensitive and specific assays, with high predictive value, for the early identification of HIV, infected women at high risk of ICC. Our approach to developing relevant molecular assays is based on the knowledge that the set of expressed genes, or """"""""expression profile"""""""" of normal cells differs from that of cancer and dysplastic cells of the same tissue type [Fields, 1994]. We propose to biopsy HIV seropositive and HIV seronegative women infected with oncogenic HPV types and with varying grades of cervical lesion abnormalities (CIN 1, CIN 2/3, CIS, ICC) and examine the expression profiles of these different lesions to identify changes in gene expression that are characteristic of malignancy but occur prior to ICC (i.e., in CIS). Once such profiles are identified, we will determine and compare the frequency with which abnormally expressed genes are present in earlier precursor lesions among HIV seropositive and HIV seronegative women. Validation on additional samples, and proof of principle testing on stored longitudinal samples from women who did and did not subsequently develop CIN 3/CIS, will be performed. This will provide the rationale for future development of """"""""low tech"""""""" ELISA assays for detection of the protein products of the over-expressed genes of interest and for longitudinal studies demonstrating the utility of this approach.
