Determining the optimal cervical cancer screening exit strategy in the Netherlands

Abstract

Objective: In the Netherlands, women are currently invited for human papillomavirus (HPV) screening until age 60 (or age 65 for HPV-positive women). New data from HPV screening implementation in the Netherlands improved understanding of its longer-term protective effects and risk differences between population subgroups. With this, our aim was to optimize screening exit strategies.

Methods: The microsimulation model MISCAN-Cervix was used to simulate a population of unvaccinated women born between 1962 and 1992 over their lifetime. We simulated 20 different exit strategies, varying by screening end ages and screening interval dependent on previous HPV status, with two triage methods (cytology triage or direct colposcopy referral for HPV16/18+ women aged ≥60). Main outcome measures were total and unnecessary colposcopy referrals (i.e. ≤ cervical intra-epithelial neoplasia stage 1), cancer cases and deaths prevented, and (quality-adjusted) life years gained ((QA)LYG). The incremental cost-effectiveness ratios (ICERs) were calculated for scenarios on the cost-effectiveness frontier.

Results: Screening of HPV-positive women at age 65 and age 70, with direct colposcopy referral of HPV16/18+ women aged ≥60 was the optimal exit strategy considering a threshold of €50,000 per LYG (ICER: €20,190/LYG, €46,985/QALY). This resulted in 18 additional cancer deaths prevented and 158 additional unnecessary referrals per 100,000 simulated women compared to the current strategy. Direct colposcopy referral of HPV16/18+ women aged ≥60 improved cost-effectiveness in all scenarios.

Conclusions: In the Dutch HPV screening program, adding screening moments for older HPV-positive women and/or incorporating direct referrals for HPV16/18+ women is cost-effective and could increase the efficiency and effectiveness of screening.

Keywords: Age range; Simulation model; hrHPV-based screening.