The current prevention of cervical cancer and elimination of its precursors is predicated on the identification of cervical cytologic abnormalities and their histologic confirmation. This strategy, although effective, depends on both sensitivity and specificity of cytology and precise histologic distinction between precursor lesions and their mimics during biopsy interpretation. The effective application of diagnostic criteria is operator dependent and varies as a function of experience and training. However, because human papilloma viruses (HPV) are causative agents and alter the cell cycle in cervical neoplasms, host genes interacting directly or indirectly with HPV oncoproteins have been identified in vitro. Recent research has centered on identifying the host genes upregulated in association with HPV infection, determining their suitability as "surrogate markers" for HPV infection, and using these markers to identify HPV-associated epithelial lesions in tissue or cytologic specimens. This review surveys recent advances in this field, summarizing the advantages and limitations of several candidate biomarkers, including PCNA, Ki-67, cyclin E, p16ink4, MN antigen, carcinoembryonic antigen (CEA), and telomerase in the recognition of preinvasive cervical neoplasia, and discusses their future potential in cervical cancer screening. Based on current studies, the strongest candidates for diagnosis and screening are p16 and cyclin E (squamous) and MN (glandular) biomarkers. As new genes are identified and tested, the concept of biomarkers as tools in primary screening and lesion recognition will continue to mature.