Factors predicting human papillomavirus clearance in cervical intraepithelial neoplasia lesions treated by conization
Silvano Costa1, Pia De Simone, Simona Venturoli
1Department of Obstetrics and Gynecology, University of Bologna, Bologna, Italy. scosta@med.unibo.it
Objective:
The objective was to identify the factors, if any, that may predict long-term results of CIN treatment and HPV clearance/persistence after locally excisional therapy.
Methods:
A series of 252 women with CIN lesions treated by conization were subjected to sequential HPV detection by repeated PCR during the prospective posttreatment follow-up. Factors predicting viral clearance during the follow-up (10.26 months) were elaborated using univariate and multivariate statistical techniques applied on epidemiological, clinical and biological data of the lesions.
Results:
Sensitivity of the PAP test in detecting high-grade lesions was 93.9%, and specificity 27.3%. Odds ratio for having CIN 3/Stage IA1 squamous cervical cancer in the cone with HSIL PAP test was 5.69; 77.8 and 22.2% residual disease were found among PCR-positive and -negative cases, respectively. HPV DNA was negative in 74/252 (29.8%) samples at the first PCR. Multivariate logistic regression analysis showed that HPV 16 was an independent explanatory factor for high-grade CIN (P = 0.0001). HPV clearance increased to 63.5% at completion of the follow-up, corresponding to the monthly clearance rate of 5.27%. In Kaplan-Meier analysis, the highly significant (P = 0.0001) predictors of HPV clearance/persistence were age, lesion grade in the biopsy, lesion grade in the cone, volume of the cone, length of active sexual life, and involvement of endocervical margin (P = 0.0013). In chi-square tests, high-risk HPV type (P = 0.001) was such a predictor. In multivariate (Cox) model, the significant independent predictors of HPV clearance were involved endocervical margin (P = 0.001), lesion grade in the cone (P = 0.004), high-grade lesion in the colposcopic biopsy (P = 0.023), age (P = 0.029), and HSIL in PAP smear (P = 0.029).
Conclusions:
These data suggest that posttreatment follow-up should include both the PAP test and HPV detection techniques for early detection of any patients at increased risk for disease recurrence and progression, because of persistent oncogenic HPV types.