Purpose: To reevaluate current US cervical cancer screening guidelines in the context of better data on the performance of human papillomavirus (HPV) DNA testing and recent FDA approval of a prophylactic vaccine against two of the most common cancer-causing types of HPV, types 16 and 18.

Methods: We used a first-order Monte Carlo simulation model of HPV infection and cervical cancer to estimate cancer incidence reduction, quality-adjusted life-year gains (QALYs), and lifetime costs (2004 US dollars) associated with different cancer control strategies. The model was calibrated to targets from North American studies of HPV type-distribution and age-specific prevalence, precancerous lesions, and invasive cervical cancer. For many randomly generated parameter sets, model outputs were compared to targets using a likelihood-based acceptance threshold for goodness of fit. Strategies included: No screening; Cervical cytology with triage HPV testing; HPV testing with triage cytology; Cervical cytology with triage HPV testing switching to HPV testing with triage cytology for older women; Cervical cytology with triage HPV testing switching to cervical cytology and HPV testing used in combination for older women. Strategies differed by: concurrent vaccine use; screening frequency (every 1, 2, 3, or 5 years), start age (18, 21, or 25), and test switch age (25, 30, or 35). We conducted cost-effectiveness analyses from a societal perspective using 20 good-fitting parameters sets.

Results: Starting 3-year screening intervals at 21-25 with cervical cytology and triage HPV testing and switching at 30-35 to HPV testing with triage cytology had incremental cost-effectiveness ratios (ICERs) of $35,000-80,000/QALY compared to the next best strategy. With concurrent vaccination, the preferred screening strategy remained constant, but the screening interval widened to 5-years ($21,000-83,000/QALY). In two-way sensitivity analyses of HPV test characteristics, the preferred strategy was more sensitive to reductions in specificity. As vaccine efficacy was reduced (less protective, more costly, or administered in the late teens or early 20s), the use of more intensive HPV testing in older women became more attractive.

Conclusions: Every 3-year HPV testing for older women has a cost-effectiveness ratio comparable to those deemed acceptable for other medical interventions and outperforms current screening recommendations under a range of assumptions. Provided vaccination is at least 80% effective, every 5-year HPV testing for older women had a cost-effectiveness ratio under $100,000/QALY.