OC 09-03DETERMINATION OF THE 5-YEAR LONGITUDINAL NEGATIVE PREDICTIVE VALUE OF THE APTIMA HPV TEST IN A ROUTINE SCREENING POPULATION IN GERMANY

10. HPV testing
T. Iftner 1, S. Becker 2, K.J. Neis 3, A. Castanon 4, A. Staebler 5, M. Henes 2, C.H. Von Weyhern 5, S. Brucker 2, P. Sasieni 4.
1Division of Experimental Virology, Institute of Medical Virology, University Hospital Tübingen (Germany), 2Department of Gynecology and Obstetrics, University Hospital Tübingen (Germany), 3Frauenärzte am Staden, Saarbrücken (Germany), 4Centre for Cancer Prevention, Queen Mary University of London (United Kingdom), 5Department of Pathology and Neuropathology, University Hospital Tübingen (Germany)

Background / Objectives

To date there are no data available on the 5-year longitudinal clinical performance of the RNA-based Aptima HPV test (AHPV) in screening populations. Thus, we compared AHPV and HC2 performances to liquid-based cytology (LBC) in women aged 30–65 attending routine cervical screening and follow both test-positive and -negative women for up to 5 years.


Methods

Women (N=10,040) were screened at office-based gynaecologists. All specimens were collected and tested centrally by LBC, AHPV and HC2. Women were referred to colposcopy if they had an abnormal cytology result and/or were tested positive on either HPV assay. Sensitivity, specificity and positive predictive values were calculated based on review histology. For follow-up, 482 women with a positive test result and no treatment were followed up for 5 years with annual testing. In addition, 5 years after baseline, cervical samples are being collected from a random sample of 4000 study participants who tested triple negative at baseline for determination of the longitudinal negative predictive value (NPV) and HPV related disease after a 5 year period.


Results

Cross-sectional results of 9336 women have recently been published and show comparable sensitivities for CIN2+ and CIN3+ detection, while the positive predictive value (PPV) and specificity for <CIN2+ was significantly higher for the AHPV test compared to HC2. An interim analysis of the 48 months follow-up data shows a cumulative proportion of HPV clearance at month 28 of 78%.  With regard to 15 incident CIN3+ lesions during follow up we calculated a PPV of the HC2 and AHPV base line test result of 3.9% (2.0-6.8) vs 5.0% (2.6-8.6), respectively, while the number of women requiring follow up deduced from the base line result is lower for women tested by the AHPV test. Furthermore, preliminary data from 10% of the target population of the 5-year follow-up cohort of triple-negative women suggest a similar longitudinal NPV of AHPV and HC2. More robust data will be presented at the conference.


Conclusion

In conclusion, our data demonstrate a non-inferior performance of the RNA-based AHPV test in comparison to the gold-standard DNA-based HC2 test.


References