Impact of differentiating between persistent and new infections on colposcopy referral in HPV-positive triage-negative women: results from the NTCC2 study

Background: In cervical screening, human papillomavirus (HPV)-positive women at 1-year retesting are typically referred to colposcopy. This study, by the use of extended genotyping, estimates the impact of distinguishing persistent from new infections in an effort to reduce colposcopy referral. It also evaluates whether the new infection rate varies according to genotyping, cytology, p16/ki67, and E6/E7 mRNA results. Methods: We analyzed data from HPV-DNA-positive women at baseline in the NTCC2 trial genotyped with the Onclarity HPV assay. Eligible participants were Onclarity-positive women without baseline CIN2+ who had a cervicovaginal sample collected at least 10 months after baseline. Persistent infections were defined as cases with at least one common genotype between baseline and follow-up specimens. New infections were defined as cases positive for different genotypes at follow-up, with no baseline genotypes detected. Results: Among 1,540 women included, 613 were Onclarity-positive at both baseline and 1-year retesting: 488 (79.6%) had persistent infections, while 68 (11.1%) had new ones. All the 11 CIN3 cases identified at follow-up occurred in women with persistent infections. The new infection rate was significantly higher in women with baseline single-channel compared to multiple channels positivity (13.3% vs 5.5%, p=0.001). Stratifying by age, persistence and new infection rates were significantly different among different age groups (respectively p=0.005 and p=0.009). No significant difference in both new infections and persistence rate was observed between baseline cytology positive and negative cases. Stratifying by baseline p16/ki67 results, we found a significantly higher rate of persistent cases among p16/ki67 positive cases and of new infections among p16/ki67 negative cases (respectively p=0.013 and p=0.016). E6/E7 mRNA findings affected only persistence rate showing a significant difference in the proportion of persistent cases between positive and negative cases (p=0.004), but did not affect new infection rate. Conclusion: Extended genotyping classified 11.5% of 1-year HPV-positive cases as new infections. As 1-year HPV positivity accounts for about 60% of first-level colposcopies, this corresponds to about 7% of colposcopies. Baseline single-channel positivity, age, and p16/ki67-negative results may influence this proportion. Trial registration: Clinicaltrials.gov registration number: NCT01837693, New Technology in Cervical Cancer 2 (NTCC2) study.