Comparison of cervical cancer screening results among 256,648 women in multiple clinical practices.

Abstract

Blatt et al1 have presented data regarding the correlations of routine high-risk human papillomavirus (HPV) and Papanicolaou (Pap) “cotesting” results with biopsy diagnosis in approximately one-quarter of million women aged 30 to 65 years. These data complement those from Kaiser Permanente Northern California, in which >1 million women aged 30 to 65 years have been cotested since 2003.2 Unfortunately, the analysis by Blatt et al1 introduces a bias that favors disease detected by Pap testing and, by extension, cotesting. Guidelines recommend that women who have an HPV-positive atypical squamous cells of undetermined significance, low-grade squamous intraepithelial lesion, or more severe cytologic interpretation are referred for immediate colposcopy.3 Women who test positive for HPV but have a negative Pap test (HPV+/Pap−) are recommended to undergo rescreening within a year and sent to colposcopy if they have a cytologic abnormality or evidence of persistent high-risk HPV.3 Thus, including only those women who underwent a biopsy within 1 year excludes a significant percentage of women with HPV+/Pap− results who came back for rescreening, because many women do not come back exactly at 1 year (or less) but return over the course of several months past the one-year anniversary. Restricting to 1-year follow-up likely results in an underattribution of disease detected by HPV screening and missed by Pap testing. It therefore will be important to reanalyze these data with ample follow-up of the women with HPV+/Pap− results to better reflect the real-world performance of HPV testing and cotesting. More importantly, the critical question should be what cervical cancer screening strategy is more effective and cost-effective, and has a better benefits-to-harms ratio over a “screening lifetime.” Although a single cotest will always be more sensitive than either test alone, how does cotesting every 5 years compare with HPV testing alone every 3 or 4 years projected over 20 or 30 years? For example, over a 20-year period, are 4 rounds of cotesting every 5 years more effective than 5 rounds of HPV testing alone every 4 years? Recent data from Kaiser Permanente Northern California2 have demonstrated that the 5-year reassurance after a negative cotest is approximately the same as the 4-year reassurance after a negative HPV test. Cotesting every 5 years would have one fewer HPV test but 4 more Pap tests than HPV testing every 4 years to achieve approximately the same level of effectiveness in that period of time. Over a 30-year period, cotesting every 5 years would have 4 fewer HPV tests but 6 more Pap tests and possibly less reassurance than HPV testing every 3 years. Thus, screening modalities must be projected over the screening lifetime to estimate its relative programmatic benefits, harms, and costs to make an informed decision regarding which approach is best for women. No specific funding was disclosed. Dr. Castle has received commercial human papillomavirus tests for research at a reduced or no cost from Roche, Qiagen, Norchip, Arbor Vita Corporation, BD, and mtm. He has been compensated financially as a member of a Merck Data and Safety Monitoring Board for human papillomavirus vaccines. Dr. Castle has been paid as consultant for BD, Gen-Probe/Hologic, Roche, Cepheid, ClearPath, Guided Therapeutics, Teva Pharmaceutics, Genticel, Inovio Pharmaceuticals, and GE Healthcare. Dr. Castle has received honoraria as a speaker for Roche and Cepheid. Philip E. Castle, PhD, MPH Department of Epidemiology and Population Health Albert Einstein College of Medicine Bronx, New York