IARC 60th Anniversary - 19-21 May 2026
Session : Global and local modelling for shaping future cancer control policies
Cervical cancer screening in HPV-vaccinated birth-cohorts in Europe: using public health decision modelling to support recommendations
GINI A. 1, GIORGI-ROSSI P. 2, TAGHAVI K. 1, ARROSPIDE A. 3, ARBYN M. 1,4, BROUTET N. 5, BERKHOF J. 6, DILLNER J. 7, BASU P. 1, BAUSSANO I. 1
1 International Agency for Research on Cancer, LYON 07, France; 2 Azienda USL-IRCCS di Reggio Emilia, Reggio Emilia, Italy; 3 Ministry of Health of the Basque Government, Vitoria-Gasteiz, Spain; 4 Belgian Cancer Centre, Sciensano, Brussels, Belgium; 5 Consultant in Health Sciences Research, Geneva, Switzerland; 6 Amsterdam University Medical Center, Amsterdam, Netherlands; 7 Karolinska Institutet, Stockholm, Sweden
Background:
Despite declining cervical cancer rates in HPV-vaccinated women, the optimal balance between screening benefits, harms, and resource use remains uncertain.
Objectives:
Using public health decision modelling, we estimated future impacts of organized cervical screening to inform the European Commission initiative on Cervical Cancer (EC-CvC) in developing recommendations for HPV-vaccinated birth-cohorts with at least 50% vaccination coverage.
Methods:
We used the IARC METHIS platform to project the impact of screening in European HPV-vaccinated cohorts. Analyses considered variations in pre-vaccination HPV prevalence, vaccination coverage (50%–100%), target (girls only or gender-neutral), and vaccine type (bivalent/quadrivalent or nonavalent). Assuming full adherence to screening pathways, several HPV-based screening strategies were evaluated, differing by starting age (25, 30, 35 years), stopping age (40, 50, 65 years), interval (5, 10, 20 years), and triage algorithm (with vs without immediately referring HPV16/18 positive women to colposcopy). Among approximately 4750 scenarios evaluated, those that were consistently efficient, safe relative to existing population screening, and cost-effective (incremental cost-effectiveness ratio [ICER] < €100,000 per life-year gained) were prioritised. Outcomes were compared with a reference strategy (5-yearly screening from ages 25 to 65).
Results:
Without screening, the expected lifetime number of cervical cancers in a European vaccinated birth-cohort (with at least 50% vaccination coverage) ranged from 90 to 1520 (per 100,000 women), depending on pre-vaccination HPV prevalence and vaccination coverage assumptions. Compared to 10 year interval, we found that repeating HPV-based screening every 5 years was not optimal (ICER > €100,000 per life-year gained) across all simulated scenarios. Prioritised 10-year strategies included: screening at ages 25–45, which missed – on average across all evaluated scenarios - 0.9 cancers (lifetime, per 100,000 women) but avoided 14 preterm births lifetime (per 100,000 women) as post-treatment consequences compared with the reference strategy; and screening only at ages 30 and 40, which minimized harms (−87 lifetime preterm births per 100,000 women) but left 18.3 cancers (lifetime, per 100,000 women) unprevented. While assumptions related to triage affected estimates on screening resources (i.e., colposcopy referrals), they did not alter the identification of the prioritized strategies. Extending 10-year HPV-based screening to age 65 increased colposcopies with negligible impact on cancer or preterm birth. Although initially prioritized, the EC?CvC expert working group did not consider results with 20?year interval as consequence of literature evidence on screening safety, acceptability, and feasibility.
Conclusions:
When vaccination coverage is at least 50%, extending HPV-based screening intervals to 10 years in vaccinated birth-cohorts offers clear benefits, while a 5-year interval is unlikely to be efficient. These findings played a key role in shaping the EC-CvC recommendations for cervical cancer screening in HPV-vaccinated populations.