A Suggested Approach to Simplify and Improve Cervical Screening in the United States

Abstract

Cervical cancer prevention strategies in the United States have become complicated and even controversial, despite advanced understanding of carcinogenic human papillomavirus (HPV) infection as the necessary causal agent. Twenty years ago, etiologic and methodologic studies had already yielded 2 powerful preventive approaches. There are excellent vaccines to prevent the most carcinogenic types of HPV infection; reduced HPV endemicity will ultimately prevent a large fraction of cervical precancer and cancers. For prevention of cervical cancer in the interim, sensitive HPV tests that target women at risk of cancer, by detection of the DNA/RNA of approximately a dozen carcinogenic HPV types, permit early diagnosis and treatment of precancers.Although HPV vaccines and tests have continued to improve, implementation of these new HPV-based prevention methods has been relatively slow in the United States and in most places worldwide. Increasing vaccination rates is the clearest and most vital long-term priority. But, for decades to come, screening will also be important. To promote useful discussion, this commentary will raise some current critical issues in simplifying and speeding the rational introduction of HPV molecular methods into US cervical screening.

Figure 1. Comparison of Common Test Results and Categories in Cervical Screening Programs To the Stages of Cervical Cancer Development

Cervical screening programs generate a large number of terms and categories that only imperfectly correspond to the now-established stages in cervical carcinogenesis (top row). A focus on diagnosis of each stage, with simplification of terms could usefully organize available test methods and maximize concordant clinical action (rescreening at an extended interval when screening indicates a normal cervix, accelerated retesting for high-risk HPV infection, colposcopic biopsy to diagnose precancer treatable by outpatient procedures, and curative or palliative treatment of invasive cancer.) The figure indicates that no screening or diagnostic test is perfect. For example, histology, our current diagnostic reference standard, tends to overcall precancer because we cannot yet determine which severe intraepithelial microscopic abnormalities indicate that the lesion would invade if untreated (rather than regress or persist). The impact of replacing the CIN scale with LAST criteria incorporating p16 testing to clarify precancer are not yet known. In any case, histology cannot accurately distinguish HPV infection from the normal cervix. Similarly, cytologic categories are prone to misclassification of HPV status and whether infection has progressed to precancer/cancer. HPV testing is most effective at establishing normalcy (i.e., lack of infection implying extremely low risk of precancer/cancer) but it cannot distinguish between benign infection and precancer/cancer; secondary triage tests and extended testing intervals are needed to prevent substantial overtreatment.