OC 11-11Performance of a New HPV and Biomarker Assay in the Management of High Risk HPV Positive Cases

13. Molecular markers
A. Kocsis 1, Z. Schaff 2, R. Koiss 3, B. Járay 4, G. Sobel 2, K. Pap 5, I. Székely 3, H.C. Lai 6, T. Takács 4, M. Nyíri 1, M. Benczik 4.
1NEUMANN DX Ltd., Cellcall Ltd. (Hungary), 22nd Department of Pathology, Semmelweis University (Hungary), 3Department of Obstetrics and Gynaecology, St. Stephan Hospital (Hungary), 4NEUMANN DX Ltd., Cellcall Ltd., Synlab Hungary Ltd. (Hungary), 5Jósa András Regional Central Hospital (Hungary), 6Shunag Ho Hospital, Taipei Medical University (Hungary)

Background / Objectives

Objectives. The most recent guidelines recommend primary hrHPV screening as an alternative to cytology-based cervical cancer screening. Although hrHPV detection has high sensitivity, it is not specific enough, leading to a substantial false positive rate. In our ongoing multicentric clinical study we investigated an epigenetic molecular biomarker, based on host gene methylation, as a potential triage method of hrHPV positive women. 


Methods

Methods. In the presented multicentric clinical study over 6,000 women were enrolled to assess the performance of the CONFIDENCE™ assay. Cervical specimens were obtained from women aged between 18 and 65 years , who attend cervical sampling at 4 clinical sites among outpatient and colposcopy clinics in Hungary. LBC cytology, hrHPV detection (CONFIDENCE™ HPV, NEUMANN Ltd.; and cobas® HPV, Roche or Full Spectrum HPV, GenoID/Synlab as comparator) and single target (POU4F3) host-gene methylation test (CONFIDENCE™ Marker, NEUMANN Ltd.) were performed using the same LBC sample in all cases independently from the HPV status. In clinically indicated 116 cases,  diagnostic histology was used as a ’gold standard’ reference method. In the absence of histologically confirmed diagnosis, negative histology result was presumed. The current analysis is focused on the baseline cross-sectional clinical data of 5320 LBC samples above 25 years of age. 


Results

Results. Overall 20.46% of the LBC samples was hrHPV positive. CONFIDENCE™ HPV test was compared to cobas HPV on 3270 samples, and to Full Spectrum HPV in 2280 cases. We found 70.4% and 61.4% agreement in the hrHPV positive cases, with overall kappa value 0.78 (95% CI: 0.74-0.81) and 0.70 (95% CI: 0.66-0.74), which is a very good agreement. The sensitivities of POU4F3 in discriminating CIN2+ and CIN3+ among hrHPV positive women were 90.0% (95% CI: 81.2-95.6%) and 92.4% (95% CI: 83.2-97.5%), whereas the specificities were 75.3% (95% CI: 72.7-77.7%) and 74.7% (95% CI: 72.1-77.1%) respectively. Positive and negative predictive value was 20.1% (95% CI: 16.1-24.6%) and 99.1% (95% CI: 98.2-99.6%) for CIN2+. The methylation of POU4F3 showed increased sensitivity with similar specificity comparing to cytology. The range of sensitivity was equivalent to cytology extended with HPV16/18 genotyping. 


Conclusion

Conclusion. Our study provides the largest clinical evaluation of a single host gene methylation biomarker using in hrHPV triage so far. In our subanalysis POU4F3 exceeds the recently described sensitivity and specificity in the field of epigenetic biomarkers of cervical precancer. POU4F3 alone shows comparable or better performance, than other methylation markers alone or in panel.


References