HUMAN PAPILLOMAVIRUS TESTING FOR THE DETECTION OF CERVICAL INTRAEPITHELIAL NEOPLASIA: A SYSTEMATIC REVIEW

Monday, October 25, 2010
Sheraton Hall E/F (Sheraton Centre Toronto Hotel)
Emily Burger, BSBA1, Hege Kornor, PhD2, Marianne Gjertsen, MD, PhD2, Vigdis Lauvrak, ScD2 and Ivar Sønbø Kristiansen, MD, PhD1, (1)University of Oslo, Oslo, Norway, (2)The Norwegian Knowledge Centre for the Health Services, Oslo, Norway

Purpose: The purpose of the study was to compare the diagnostic accuracy of human papillomavirus (HPV) mRNA testing to HPV DNA testing.

Method: We searched electronic databases MEDLINE, EMBASE and Cochrane Library from January 1996 through March 2010 using a predefined search strategy. We included publications in English or a Scandinavian language reporting data allowing the construction of a 2x2 table from studies with >50 participants where: an HPV mRNA test assessing >2 HPV genotypes was compared to an HPV DNA test, and the reference standard was histologically confirmed cervical intraepithelial neoplasia 2+ (CIN2+). Two reviewers independently assessed study eligibility, extracted data, and assessed risk of bias. Sensitivity, specificity, positive and negative likelihood ratios, positive and negative predictive values and diagnostic odds ratios were calculated for each study. In addition, we fitted a series of summary receiver operating characteristics (SROC) curves. 

Result: Out of 3126 potentially relevant citations, ten publications (nine studies) met our inclusion criteria. The included studies were of varying methodological quality, and predominately performed in a secondary screening setting. Six studies investigated the performance of the Pre Tect Proofer/NucliSENS EasyQ, two studies investigated the performance of the APTIMA assay and one study investigated both mRNA tests on the same patient samples. Due to few studies and considerable clinical heterogeneity, pooling of data was not possible. Instead, we compiled a ‘best evidence synthesis’ for E6/E7 mRNA HPV testing. Sensitivities ranged from 0.41 to 0.86 and from 0.90 to 0.95 for the Pre Tect Proofer and APTIMA assay, respectively. Specificities ranged from 0.63 to 0.97 and from 0.42 to 0.61 for the Pre Tect Proofer and APTIMA assay, respectively. The SROC curves for both mRNA tests were to the left of the diagonal and the APTIMA assay performed closest to the DNA tests. 

Conclusion: The synthesized evidence suggests that mRNA tests have diagnostic relevance. The Pre Tect Proofer appears to have a higher clinical specificity in exchange for a lower sensitivity when compared to DNA tests. The APTIMA results indicate that it may perform very closely to the DNA tests with a slightly higher specificity. Additional studies and economic evaluations should be conducted in order to make a solid conclusion regarding the clinical applicability of HPV mRNA testing.

Candidate for the Lee B. Lusted Student Prize Competition