Accuracy of GynTect® Methylation Markers to Detect Recurrent Disease in Patients Treated for CIN3: A Proof-of-Concept Case-Control Study.

Abstract

Post-treatment follow-up in women with CIN3 is mandatory due to relapse in up to 15% of patients within 2 years. Standard follow-up care based on hrHPV-DNA/cytology co-testing has high sensitivity but limited specificity. The aim of our proof-of-concept case-control study was to evaluate the performance of the methylation test GynTect® for the detection of recurrent CIN2/3 during follow-up. Residual clinical material from a recent, prospective, multicenter, observational study was available for further analysis. We studied a sample of 17 cases with recurrent CIN2/3 diagnosed within 24 months of follow-up and 31 controls without recurrence. DNA from cervical scrapes at baseline (immediately before CIN3 surgery) and up to three follow-up visits were analyzed for hrHPV and GynTect® methylation status. Cytology data were available from the previous study. Overall, 12 cases and 21 controls were GynTect-positive at baseline. In these subgroups, single test sensitivity at first follow-up was 67% (95% CI 39-87%) for GynTect® compared to 83% (95% CI 55-96%) for hrHPV (p = 0.50). Single test specificity was significantly higher for GynTect® (90%, 95% CI 71-98% vs. 62%, 95% CI 40-80%) (p = 0.03). In a co-testing setting, both hrHPV/cytology and GynTect®/cytology detected all recurrences. Specificity for GynTect®/cytology was higher than for hrHPV/cytology, but this difference was not statistically significant. In conclusion, for initially GynTect-positive patients, both hrHPV and GynTect® tests detected recurrent disease with similar sensitivity, but the GynTect® assay has a higher specificity. Incident hrHPV infection and/or persisting multifocal hrHPV infections without clinical disease are most likely responsible for the poorer specificity of the hrHPV test. A future prospective validation study will have to show whether GynTect®/cytology co-testing can outperform hrHPV/cytology co-testing in post-treatment surveillance.