High risk HPV testing in the triage of repeat ASC-US and LSIL.

Abstract

The purpose of this study was to evaluate the real-life performance of high-risk (HR) HPV testing in the triage of repeat atypical squamous cells of undetermined significance (ASC-US) and low-grade squamous intraepithelial lesions (LSIL). In 2004-2012, 127 women with repeat ASC-US and 118 women with LSIL were triaged with HPV testing using either a Hybrid Capture 2(®) or Abbott RealTime(®) HR-HPV test. The patient charts were retrospectively reviewed for performance of the tests. In the repeat ASC-US group, 40.9% of the women were positive for HR-HPV. The prevalence of cervical intraepithelial neoplasia grade 2 or worse (CIN2+) was 8.7%. The sensitivity of the HR-HPV testing for detection of CIN2+ was 90.9% (95% Cl 58.7-98.5%), with a specificity of 63.8% (95%CI 54.4-72.5%). In women ≥30 years old, the specificity was 70.4%, whereas in younger women it was only 27.8%. The negative predictive value to predict CIN3 was 100% (95% CI 95.2-100.0%). Of the women with LSIL, 68.6% were positive for HR-HPV and the prevalence of CIN2+ was 14.4%. The sensitivity of HR-HPV testing for detection of CIN2+ was 94.1% (95% CI 71.2-99.0%), and the specificity 35.6% (95% CI 26.4-45.8%), respectively. The negative predictive value to predict CIN3 was 100% (95% CI 90.4-100.0%). HR-HPV testing seems to be beneficial in the triage of repeat ASC-US in women ≥30 years old, but a high prevalence of HR-HPV infection combined with poor specificity limit the use of HPV testing in the case of LSIL.