Three-Tiered versus Two-Tiered Classification of Squamous Dysplasia in Cervical Cytology: Results of a Follow-Up Study.

OBJECTIVE: Regarding cytological findings of squamous dysplasia, a comparison was made between a three-tiered classification - low-grade squamous intraepithelial lesion (LSIL), high-grade SIL/cervical intraepithelial neoplasia 2 (HSIL/CIN2), and HSIL/CIN3 - and a two-tiered classification - LSIL and HSIL. The respective risk for CIN2+ and CIN3+ was calculated to make decisions regarding management. METHODS: A total of 2,949 women with first-time cytologic findings of squamous dysplasia (LSIL, HSIL/CIN2, or HSIL/CIN3) between January 2013 and June 2016 were enrolled. Subsequent cytological findings and histological diagnoses were evaluated until August 2018. For each category of findings, the risk for CIN2+ and CIN3+ was determined by Kaplan-Meier estimates. The differences in risk between the cytological categories were checked for significance using the log-rank test. RESULTS: For the categories LSIL, HSIL/CIN2, and HSIL/CIN3, the risk for CIN2+ after 12, 24, and 60 months was 3.4, 9.4, and 23.3%; 35.2, 44.8, and 59.8%; and 95.5, 97.8, and 98.9%, respectively. For CIN3+ the risk was 2.0, 5.5, and 13.5%; 28.6, 35.6, and 48.3%; 91.3, 95.6, and 97.9%, respectively. The differences in risk between the categories are highly significant, respectively (p < 0.001). CONCLUSION: A three-tiered classification of squamous dysplasia such as the Munich Nomenclature III for cytology is suitable for risk-adapted clinical management, especially to avoid overdiagnosis and overtreatment.

p16/Ki-67 Immunocytochemistry in Gynecological Cytology: Limitations in Practice.

OBJECTIVE: Immunochemical detection of the protein p16INK4a in cervical cytology is used in combination with Ki-67. Cells positive for both proteins are certain to have been transformed by high-risk HPV. p16/Ki-67 immunocytochemistry provides a significant improvement in specificity over cytology and HPV DNA testing. However, p16/Ki-67 immunocytochemistry also has its limitations. STUDY DESIGN: The research is based on the follow-up of 1,131 patients for whom p16/Ki-67 immunocytochemistry was performed with cytology. Dependencies on the age of patients with LSIL, number of p16/Ki-67-positive cells, and different results during repeated examinations were analyzed. RESULTS: In LSIL, positive p16/Ki-67 is less specific for ≥CIN2/HSIL for patients younger than 30 years compared to patients aged 30 years or older (61.1 vs. 75.7%, p < 0.013). Using a score of 10 p16/Ki-67-marked cells as a positive result instead of 1 led to significantly higher specificity (89.0 vs. 70.2%, p < 0.001). This modified threshold offers better risk assessment in LSIL. In repeated immunocytochemical investigations, 28.4% of the results deviated from the first examination. CONCLUSION: The abovementioned discrepancies can be interpreted as hints about the molecular biological causes of suboptimal performance of p16/Ki-67. An efficient and reliable application of p16/Ki-67 immunocytochemistry requires knowledge of its methodological limitations.

Predictive value of the combined p16 and Ki-67 immunocytochemistry in low-grade squamous intraepithelial lesions.

OBJECTIVE: The reliability of cytological diagnoses, especially for low-grade squamous intraepithelial lesions (LSIL), is limited. This leads to uncertainty in patient management. The application of adjunctive biomarkers is meant to improve this situation. Therefore, we examined the prognostic value of p16/Ki-67 immunostaining of LSIL cytology specimens. STUDY DESIGN: We analyzed the p16(INK4a) and Ki-67 immunocytochemistry (CINtec® PLUS, dual stain) of 260 patients with LSIL. Cytology and dual-stain results were correlated with histology at the time of treatment or with cytological follow-up. RESULTS: After an average duration of 24.9 months (1-58) and a histology rate of 36.2% [cervical intraepithelial neoplasia, grade 2 or higher (CIN2+) as positive], the statistical evaluation for cytology and dual stain resulted in a sensitivity of 98.3 and 90.0%, respectively, a specificity of 74.5% for dual stain, a positive predictive value (PPV) of 22.8 and 51.4%, and a negative predictive value (NPV) of 96.1% for dual stain. CONCLUSION: The combined immunocytochemical investigation of p16(INK4a) and Ki-67 leads to a significantly better PPV and a very good NPV for CIN2+ in LSIL, especially in women 30 years of age and older. An objective individualized prognosis may not be achieved with p16(INK4a)/Ki-67. Statistical data from our study, however, indicate that patient management can be significantly improved by the application of combined p16/Ki-67 immunocytochemistry as an adjunct to cytology.