Cervix intraepithelial neoplasia (CIN), a precancerous lesion of invasive squamous carcinoma, can be classified histologically as CIN grades I to III. Not all CIN lesions progress to invasive cancer. In fact, most of the CIN I and II lesions spontaneously regress. However, factors that determine regression, persistence, or progression of CIN remain unknown. In this proposed prospective study, women with histopathologically diagnosed CIN I or II will be enrolled in a 12-month non-therapeutic follow-up study. The purpose is to identify biomarkers that predict spontaneous regression and persistence of cervix dysplasia. Defining biomarkers that modulate the biological process of CIN, the intermediate endpoint for invasive cancer, will provide valuable information on the carcinogenesis of cervix cancer. With a 12-month follow-up period, the emphasis of the study is not on progression, but rather on regression/persistence of CIN. Gaining insight about regression will have significant implications for understanding the natural history of the neoplasm as well as the prognosis and clinical management of women with CIN I or II. Women with histopathologically confirmed CIN I or II (N=465) will be recruited from the Bronx Municipal Hospital and the gynecology clinic of the Albert Einstein College of Medicine. They will be followed at 3-month intervals by Pap smear and colposcopy for 12 months. An endpoint biopsy will be performed at 12 months. The outcome of the study will be the presence (persistence) or absence (regression) of CIN lesion at 12 months. At each visit, cervico-vaginal lavage and a cervical swab sample will be collected for human papillomavirus (HPV) DNA analyses by both southern blot and polymerase chain reaction (PCR) techniques. Extracted DNA will also be used for typing of class II HLA-DQB1 and HLA-DRB1 alleles by PCR. Blood and urine samples as well as questionnaire data will be collected at baseline, 6, and 12 months. Plasma beta-carotene, ascorbic acid, and alpha-tocopherol levels will be measured by high pressure liquid chromatography methods, and red cell folate levels by Radioassay. The urinary ratio of 2-hydroxyestrone to 16alpha-hydroxyestrone will be assessed by Enzyme Immunoassay. The questionnaire will collect information on demographic and behavioral factors. The study pathologist will review the initial and endpoint biopsy specimens, and the histologic characteristics of the lesion will be recorded. The association between regression/persistence of CIN and each of these biomarkers will first be examined in univariate analyses by chi-square, t test, and Wilcoxon rank sum test. Whether the outcome of CIN is influenced by interaction between HPV infection and another co-factor will be examined in stratification analyses. Finally, significant variables and interaction terms will be subjected to multiple logistic regression analysis where independent factors will be identified. ROC curve analysis will be used to examine the sensitivity and specificity of this model in discriminating women who will have regression versus women who will have persistent or progressive CIN.
