Cost-Effectiveness of High-Risk Human Papillomavirus Testing With Messenger RNA Versus DNA Under United States Guidelines for Cervical Cancer Screening.

Objectives: To compare the cost-effectiveness of high-risk human papillomavirus (hrHPV) testing using a hrHPV DNA and a hrHPV messenger RNA (mRNA) assay under current US cervical cancer screening guidelines.

Methods: We constructed a Markov model for stochastic cost-effectiveness analysis using published data. We compared screening efficiency using DNA and mRNA testing for the following: (1) cotesting with cytology in women 30 to 65 years, and (2) triage of women with mild cervical cytological abnormalities (atypical squamous cells of undetermined significance [ASC-US]) in the United States. Screening end point is histologically confirmed high-grade lesions (cervical intraepithelial neoplasia grade 2, 3, or invasive cancer). Sensitivity and specificity estimates of DNA and mRNA testing to detect cervical intraepithelial neoplasia grade 2, 3, or invasive cancer were obtained from 2 published trials: the US Clinical Evaluation of APTIMA mRNA (CLEAR) study for ASC-US triage and the French APTIMA Screening Evaluation (FASE) study for cotesting. Costs of DNA and mRNA testing were assumed identical. Costs of screening, diagnosis, and treatment of cervical neoplasia and cancer were from previously published estimates, adjusted to 2012 US dollars. Inputs were modeled as distributions for Monte Carlo probabilistic sensitivity analysis. Model outcomes were costs per life-year saved for each strategy, discounted at 3% annually.

Results: For both cotesting and ASC-US triage, mRNA testing cost less than DNA testing, whereas life expectancies were widely overlapping. There was a 100% probability that DNA testing was not cost-effective at $100,000/life-year saved threshold for ASC-US triage and a 55% probability that DNA testing was not cost-effective at the same threshold for cotesting.

Conclusions: Based on the available evidence, mRNA testing for cotesting or ASC-US triage is likely to be more efficient than DNA testing under current US cervical cancer screening guidelines.