The introduction of widespread HPV vaccination for pre-adolescents has important implications for the conduct of cervical screening (What is the future of cervical screening in the era of vaccination?/STI/blogs). Ecological studies have shown the potential impact of vaccination on cervical cancer, by using various proxies for cervical cancer prevention, including declines in HPV infections (Garland & Jayasinghe/STI), genital warts (Chow & Fairley/STI; Ali & Donovan/STI; Wilson & Baker/STI), or HPV specific neoplasia or cytological abnormalities (Paavonen/STI ; Kibur & Lehtinen/STI).  Ferris & Das, in a follow-up study of the quadrivalent virus (against HPV types 6, 11, 16, 18) covering nine countries, have recently demonstrated the vaccine’s clinical effectiveness over ten years. In countries that have introduced vaccination programmes, this declining incidence of abnormalities will have an impact on the effectiveness of cervical screening, and will alter the balance of benefit and harm (through false positives). In the UK, where vaccination using the quadrivalent vaccine became widespread in 2007-2009, an initial response to this changing situation was made in January 2016 with the decision of the National Screening Committee to move from a cytology-, to a primary HPV-, based regime of screening. But the question of the best regime of screening – i.e. the number and timing of HPV screening tests – remains to be determined, and the choice between the quadrivalent vaccine and the nonavalent vaccine (against types, 6, 11, 16, 18, 31, 33, 45, 52, 58) may yet be subject to revision.

A recent modelling study (Landy & Sasieni) sets out to offer an answer that is valid for the UK – though the design of the model, which is not calibrated to the UK population but investigates outcomes shown to be robust under diverse model calibrations, makes it relevant to other populations. Another curious feature of this study is that it does not evaluate effectiveness in terms of cost-effectiveness, but of the incremental benefit (as a percentage of cancers prevented) of each additional screen. Alternative scenarios consider the outcomes with q4 and q9 HPV vaccine, assume 100% or ‘realistic’ levels of compliance, and do – or do not – make allowance for some level of cross-protection (q4 HPV protecting against cancer-causing strains other than 16 and 18) or diminishing vaccine effectiveness. Finally – with a view specifically to UK policy development – parameters for ‘worthwhile’ incremental benefit per screen are benchmarked to assumptions of acceptability implied by the UK recommendation of 3(5) year cytology over 6(10) year cytology and 6 (10) year primary HPV over 3(5) primary HPV. This sets worthwhile incremental benefit for each additional screening at≤ ≤ 2% and ≥ 0.9% of cancers prevented.

So what is the outcome? For the quadrivalent vaccine, the simulation favours three screens at age 30, 40, and 55 (as opposed to the current seven screens with primary HPV testing), with an incremental benefit of 2.7% for the third screen, but of only 1% for a fourth. For the nonavalent vaccine, two screens are favoured, with incremental benefit of 1.6% for the second screen, and 1% for a third.

There remains, of course, the question of which of the two vaccines – the quadrivalent or the nonavalent – should be preferred (NHS position (as of 2015); The efficacy of Gardasil nonavalent HPV vaccine/STI/blogs).