Is 58% sensitivity for detection of cervical intraepithelial neoplasia 3 and invasive cervical cancer optimal for cervical screening?

Abstract

Recent Food and Drug Administration (FDA) approval of a Roche cobas human papillomavirus (HPV) test application as a first line primary cervical screening tool in women 25 and older introduces a new era of complex cervical screening choices. Perhaps the most surprising findings in Roche's supporting ATHENA trial data were the unexpectedly low verification bias-adjusted CIN3+ sensitivities documented by the FDA for both the proposed cobas HPV testing algorithm (58.26%) and Pap testing algorithm (42.63%). These unexpectedly low sensitivity estimates suggest intuitively that there is still considerable room for improvement in cervical screening, and available data from large systems point to routine cytology and HPV co-testing as offering the greatest protection against development of cervical cancer. Observational studies of large populations screened over time remain essential to document actual protection from development of cervical cancer with any new cervical screening options, as natural history studies and available data from large systems indicate that most CIN2/3 cases detected in short term clinical trials would not progress to invasive cervical cancer. Interpretation of ATHENA trial data and its application to routine clinical practice is further limited by published studies which document that a significant proportion of CIN2/3 biopsy diagnoses in the ATHENA trial could not be confirmed as accurate when evaluated with p16 immunohistochemistry and that cytology laboratory performance in the trial was notably suboptimal.