UICC World Cancer Congress 2006

Objective: High-risk HPV is the primary cause of cervical cancer. The test for HPV infection is effective but expensive, especially for women in the developing world. We attempted to construct a low-cost risk model for screening and selecting apparently healthy women from high risk rural areas in China for HC2 testing.

Methods: In the 2 screening sites of a multi-center study, 4517 eligible women aged 30–49 were recruited. Participants completed a questionnaire regarding risk factor exposures and underwent the HC-2 test for 13 high-risk HPV types, Liquid-based cytology, Visual inspection with acetic acid (VIA) and colposcopy and biopsy as indicated. Risk factors for high-risk HPV and cervical cancer were examined using univariate and multivariate logistic regression. The utility of the final multivariate model for HPV as a screening tool of cervical cancer was assessed using a Receiver Operating Characteristics curve. Totally 3336 women had complete data including HPV, pathology results and questions of risk factors.

Results: 3442 women had complete HPV results and risk factor questionnaire information. Among them, 638 (18.54%) women had the high-risk HPV infection. The sensitivity for this model is 66.07%, and the specificity is 61.94%. Results of this model cannot be regarded as gold standard and women with negative index should be closely followed up while positive index helps to increase the accuracy of other testing for HPV screening such as Pap smear and LCT. Combined use of this model and other HPV screening methods is applicable especially in low-income women to avoid overuse of HC2 test.