Predictive value of p16^INK4a, Ki-67 and ProExC immuno-qualitative features in LSIL progression into HSIL

The current nested case-control study was conducted to explore the prognostic value of cyclin-dependent kinase inhibitor 2A (p16^INK4a), marker of proliferation Ki-67 (Ki-67) and immunohistochemical cocktail containing antibodies directed against topoisomerase IIα (TOP2A) and minichromosome maintenance 2 (MCM2) proteins (ProExC) immuno-qualitative features to predict low-grade squamous intraepithelial lesion (LSIL) progression. A total of 92 LSIL patients were followed-up for 2 years, where those with high-grade squamous intraepithelial lesion (HSIL) or persistent LSIL were designated as the case group and those who spontaneously regressed were designated as the control group. The infection status of human papillomavirus (HPV) was evaluated using flow-through hybridization and gene chip, whilst the expression of p16^INK4a, Ki-67 and ProExC were tested in LSIL patient biopsies by immunohistochemistry. All data were collected at the beginning of the follow-up and patient outcomes were diagnosed by histopathological examination. To analyze the risk factors for LSIL progression, sensitivity, specificity, positive-negative predictive value (PPV-NPV), positive-negative likelihood ratio (PLR-NLR), Youden's index (YI) and multinomial logistic regression analysis was performed. The expression rates of p16^INK4a, Ki-67, and ProExC were found to be higher in the progression group compared with those in the persistence and regression groups. Only p16^INK4a expression significantly associated with high-risk HPV infection. With respect to predicting HSIL, p16^INK4a staining was the most sensitive but Ki-67 staining was found to be the most specific. YI was the highest (42.1%) for p16^INK4a expression in the present study, followed by ProExC (39.5%) and Ki-67 (28.3%). However, the expression of ProExC was found to be an independent risk factor for LSIL progression into HSIL. In conclusion, whilst immunohistochemical staining for p16^INK4a, Ki-67, and ProExC can be used to predict HSIL progression, only ProExC expression can be applied an independent risk factor for LSIL progression.