Combined use of cytology, p16 immunostaining and genotyping for triage of women positive for high-risk human papillomavirus at primary screening

New Technologies for Cervical Cancer Screening Working Group, Jack Cuzick, Rachael Adcock, Francesca Carozzi, Anna Gillio-Tos, Laura De Marco, Annarosa Del Mistro, Helena Frayle, Salvatore Girlando, Cristina Sani, Massimo Confortini, Manuel Zorzi, Paolo Giorgi-Rossi, Raffaella Rizzolo, Guglielmo Ronco

Research output: Contribution to journalArticlepeer-review


Human papillomavirus (HPV) testing is very sensitive for primary cervical screening but has low specificity. Triage tests that improve specificity but maintain high sensitivity are needed. Women enrolled in the experimental arm of Phase 2 of the New Technologies for Cervical Cancer randomized controlled cervical screening trial were tested for high-risk HPV (hrHPV) and referred to colposcopy if positive. hrHPV-positive women also had HPV genotyping (by polymerase chain reaction with GP5+/GP6+ primers and reverse line blotting), immunostaining for p16 overexpression and cytology. We computed sensitivity, specificity and positive predictive value (PPV) for different combinations of tests and determined potential hierarchical ordering of triage tests. A number of 1,091 HPV-positive women had valid tests for cytology, p16 and genotyping. Ninety-two of them had cervical intraepithelial neoplasia grade 2+ (CIN2+) histology and 40 of them had CIN grade 3+ (CIN3+) histology. The PPV for CIN2+ was >10% in hrHPV-positive women with positive high-grade squamous intraepithelial lesion (61.3%), positive low-grade squamous intraepithelial lesion (LSIL+) (18.3%) and positive atypical squamous cells of undetermined significance (14.8%) cytology, p16 positive (16.7%) and, hierarchically, for infections by HPV33, 16, 35, 59, 31 and 52 (in decreasing order). Referral of women positive for either p16 or LSIL+ cytology had 97.8% sensitivity for CIN2+ and women negative for both of these had a 3-year CIN3+ risk of 0.2%. Similar results were seen for women being either p16 or HPV16/33 positive. hrHPV-positive women who were negative for p16 and cytology (LSIL threshold) had a very low CIN3+ rate in the following 3 years. Recalling them after that interval and referring those positive for either test to immediate colposcopy seem to be an efficient triage strategy. The same applies to p16 and HPV16.
Original languageEnglish
Pages (from-to)1864-1873
Number of pages10
JournalInt. J. Cancer
Issue number7
Publication statusPublished - Oct 1 2020


Dive into the research topics of 'Combined use of cytology, p16 immunostaining and genotyping for triage of women positive for high-risk human papillomavirus at primary screening'. Together they form a unique fingerprint.

Cite this