Page 44 - Strategies for non-invasive managementof high-grade cervical intraepithelial neoplasia - prognostic biomarkers and immunotherapy Margot Maria Koeneman

Chapter 2
of lesions regression. Two authors studied the prognostic value of both p53 and pRb in HSIL and found consistent results. Baak et al. followed 28 patients with histologically confirmed HSIL during a mean of 8.2 months and found that lower p53 and pRb in the deeper half of the epithelium was associated with disease persistence.(78) The overall regression rate was 43%. Lesions with low expression of both markers persisted in 16/20 cases (80%), whereas none of the eight lesions with high expression of both markers persisted. Ovestad et al. followed 55 patients with histologically confirmed CIN 2/3 over a mean of 12 weeks. Increased expression of pRb and p53 was associated with disease regression and proved to be an independent predictor of disease regression with multivariate analysis.(7)
Ki67
An association between Ki67 and progression of CIN 1/2 was established by Kruse et al.(79) They followed 44 patients for a maximum follow-up period of 18.8 months. Progression occurred in 11%. Test characteristics of their Ki67-staining method showed a sensitivity of 100%, specificity of 56%, positive predictive value of 23% and negative predictive value of 100%. Overall, 61% of cases were correctly classified. The results were validated in a test set of another 35 patients an showed similar results. As was later reviewed by Baak et al., the prognostically strongest Ki67 features were the Ki67 stratification index (threshold 0.57) and percentage of Ki67 positive nuclei in the middle third layer of the epithelium (threshold 30%).(18) The association between Ki67 and natural history of high-grade CIN was studied by Baak et al. A study with a limited number of patients (n=28) found no association between Ki67 expression and regression of high-grade CIN.(78) Given the high prognostic value of Ki67 in low-grade CIN, additional research seems indicated to evaluate the prognostic value of Ki67 in high-grade CIN.
Other markers
Currently, the only biomarker that has been implemented in clinical practice is p16. Its use is recommended by the College of American Pathologists and the American Society for Colposcopy and Cervical Pathology, as part of the LAST project (Lower Anogenital Squamous Terminology project).(80) Evaluation of p16 is intended as an adjunct to morphologic assessment of biopsy specimens for differentiation between HSIL and a disease state mimicking HSIL, or to differentiate CIN 2 into a HSIL and LSIL category. As such, p16 has been implemented as a diagnostic biomarker. The literature on the prognostic value of p16 is inconsistent. Several authors have found an association between p16 staining and progression of LSIL.(81) Recently however, Quint et al. found no correlation between p16 and LSIL progression: they followed 238 patients with histologically proven LSIL during two years. Age and HPV genotyping were identified as predictive for LSIL progression, but not cell cycle markers (pRb, p53, Ki67 and p16).(77) Three authors evaluated the prognostic value of p16 in the natural history of HSIL and found inconsistent results. Omori et al. followed 52 CIN cases during two years. Regression occurred in 28 lesions, progression in 13 cases. The risk of progression was greater with stronger p16 expression and a higher frequency of a punctate nuclear signal.(82) Baak et al. followed 28 HSIL patients during a mean of 8.2 months and found no association between p16 and disease persistence or regression.(78) Ovestad et al. retrospectively studied 55 patients for a mean of 12 weeks and found that p16 staining intensity in the upper epithelium was significantly stronger in the persistent cases. The conflicting results on
42