Performance of p16 INK4a-cytology, HPV mRNA, and HPV DNA testing to identify high grade cervical dysplasia in women with abnormal screening results

Abstract

Objective The prognostic value of dysplastic lesions of the uterine cervix cannot be adequately determined by Pap cytology alone. Detection of HPV DNA increases the diagnostic sensitivity. However, due to the very high prevalence of transient HPV infections, HPV DNA testing suffers from poor diagnostic specificity. Biomarkers that highlight the shift from self limited transient to potentially dangerous transforming HPV infections may improve the accuracy of cervical cancer screening. We evaluated HPV E6/E7 mRNA detection (APTIMA), p16 INK4a-immunocytology (CINtec), and HPV DNA testing (HC2) to identify women with high grade cervical neoplasia in a disease-enriched cross-sectional cohort. Methods Liquid based cytology specimens were collected from 275 patients. All assays were performed from these vials. Detection rates of each test were evaluated against conventional H&E based histopathology alone and stratified by p16 INK4a-immunohistochemistry (IHC). Results All assays yielded a high sensitivity for the detection of CIN3+ (96.4% (95% CI, 90.4–98.8) for HC2, 95.5% (89.2–98.3) for APTIMA and CINtec) and CIN2+ (91.5% (85.8–95.1) for HC2, 88.4% (82.3–92.7) for APTIMA, 86.6% (80.2–91.2) for CINtec). The specificity to detect high grade dysplasia was highest for CINtec p16 INK4a-cytology (60.6% (52.7–68.0) in CIN3+ and 74.8% (65.5–82.3) in CIN2+), followed by APTIMA (56.4% (48.4–64.0) in CIN3+ and 71.2% (61.7–79.2) in CIN2+) and HC2 (49.1% (41.3–56.9) in CIN3+ and 63.4% (53.7–72.1) in CIN2+). All tests had higher sensitivity using p16 INK4a-IHC-positive CIN2+ lesions as endpoint. Conclusions Biomarkers that detect HPV induced dysplastic changes in the transforming stage are promising tools to overcome the current limitations of cervical cancer screening.