Modelling the choice of high-risk human papillomavirus (HR-HPV) assay in the English cervical screening programme

Aquarius developed a decision tree model to compare the use of the Aptima mRNA HR-HPV assay to a DNA HR-HPV assay in the HPV primary screening algorithm in England. Robust results from the model demonstrate that using the mRNA HR-HPV assay is cost saving and avoids unnecessary HPV recall tests, cytology tests and colposcopies compared to using a DNA HR-HPV assay.

In England, women are invited for cervical cancer screening starting at age 25 and are recalled at regular intervals until age 64. In the HPV primary screening algorithm, cervical samples are tested first for HR-HPV. Positive HPV samples are then tested using liquid based cytology to identify abnormal cells.

Aquarius developed a cost consequence analysis decision tree taking the perspective of NHS England and modelling the HPV primary screening algorithm for one simulated cohort of 2,247,439 women aged 25-64 from baseline screen through recall visits. The model endpoint was discharge to routine recall, loss-to-follow up, or referral to colposcopy. A micro-costing approach was used to estimate costs for screening in England, using published data. Head-to-head comparison of mRNA to DNA tests at baseline and with follow-up was not available for the English population. Therefore, data from the HORIZON study in Denmark with similar positivity to that in England was used.

The model indicated that using an mRNA versus DNA assay in the HPV primary screening algorithm in England can save an estimated £15.4 million and require 28,009 fewer unnecessary colposcopies, 90,605 few HR-HPV tests and 253,477 fewer cytology tests.  Uncertainty and scenario analyses demonstrate cost and resource savings will almost certainly be achieved using the mRNA assay.

PublicationWeston G, Dombrowski C, Harvey MJ, et al Use of the Aptima mRNA high-risk human papillomavirus (HR-HPV) assay compared to a DNA HR-HPV assay in the English cervical screening programme: a decision tree model based economic evaluation BMJ Open2020; 10:e031303. doi: 10.1136/bmjopen-2019-031303