No abstract

The purpose of this study was to investigate the correlations between high-risk human papillomavirus (HPV) load and p16 (INK4a) or Ki-67, and to identify biomarkers that may predict residual disease after conization with positive margins. The following samples were analyzed: 49 paraffin-embedded specimens from patients with cervical intraepithelial neoplasia (CIN), including 12 CIN 2 conization specimens and 37 CIN 3 conization specimens. Immunohistochemical analysis was performed with antibodies to p16 (INK4a) and Ki-67. Hybrid Capture II testing was used to detect high-risk HPV DNA. The mean HPV loads within each of the p16 (INK4a)-staining cases were 9.5 (relative light units/positive control) RLU/PC for negative staining, 531.8 RLU/PC for 1+ staining, 140.2 RLU/PC for 2+ staining, and 545.1 RLU/PC for 3+ staining. HPV loads differed significantly according to p16 (INK4a) expression (P = 0.0021). The mean HPV loads within Ki-67 staining cases were 28.2 RLU/PC for 1+ staining, 189.6 RLU/PC for 2+ staining, and 563.3 RLU/PC for 3+ staining. HPV loads differed significantly according to Ki-67 expression (P = 0.0259). The expression of p16 (INK4a) (P = 0.0012) and Ki-67 (P = 0.0006) were significantly associated with the CIN grade. In univariate and multiple logistic regression analysis, age, parity, cytology, lesion grade in the cone, high-risk HPV load, and the expression of p16 (INK4a) and Ki-67 were not significantly associated with residual lesions after conization with positive margins (P > 0.05). In conclusion, high-risk HPV load showed significant differences according to the expression of p16 (INK4a) and Ki-67, while none of the prognostic factors were significantly associated with residual disease after conization with positive margins.