Transmission dynamic modelling of the impact of human papillomavirus vaccination in the United Kingdom
Introduction
Human papillomavirus (HPV) infection is necessary for the development of cervical cancer in women as well as anogenital warts in both men and women. Most diagnosed cervical cancers are squamous cell carcinomas; however, the incidence of adenocarcinomas has been increasing rapidly [1] and now accounts for up to 20% of all cancer diagnoses in the United Kingdom (UK). Two HPV types (16 and 18) are responsible for about 70% of squamous cell carcinomas [2] and 85% of adenocarcinomas [3], while another two types (6 and 11) cause over 90% of cases of anogenital warts [4].
Two prophylactic vaccines against HPV have been developed: a bivalent vaccine (Cervarix™) against types 16 and 18, and a quadrivalent vaccine (Gardasil™) that also includes types 6 and 11. In clinical trials, use of either vaccine in HPV-naive females resulted in at least 90% reduction in persistent infection and associated disease during 30 months of follow-up [5], [6]. The quadrivalent vaccine has also been shown to be highly effective at preventing anogenital warts [7].
The vaccines have the potential to reduce the substantial burden of HPV-related disease. However, they are likely to be priced at levels significantly higher than other vaccines in national vaccination schedules, so their epidemiological and economic impact needs to be carefully considered. Because of the complexity of HPV infection and pathogenesis, as well as the long delays between infection and the most serious disease endpoints (cervical cancer), mathematical models are required to estimate the impact of vaccination.
The majority of existing studies (reviewed in Ref. [8]) use static models, which explore the natural history of HPV infection on an individual level, but underestimate the population-wide impact of vaccination. Studies using dynamical models are rarer because of their greater demands in terms of both model complexity and data requirements in order to represent sexual contact patterns. However, they are required to take into account herd immunity and hence estimate the wider impact of vaccination. To date, one theoretical HPV transmission model has been published [9], and transmission models have been applied to assess HPV vaccination in the United States [10], Finland [11], [12], Brazil [13] and the UK [14]. Most of these models assume a fixed structure to represent HPV transmission and progression between stages of disease. However, there are still significant gaps in our knowledge of HPV epidemiology and natural history that make it problematic to assume a single structure for a model of HPV disease and vaccination. In order to capture this uncertainty, we develop here a series of transmission models that represent different possibilities about sexual behaviour, natural immunity, vaccine characteristics, disease progression and screening accuracy. These models are then fitted to available data and used to project the impact of vaccination, providing robust estimates of the possible impact of alternative immunisation programmes. The use of the outcomes of the models to parameterise a cost effectiveness model of HPV vaccination has been described elsewhere [15].
Section snippets
Model structure
We developed a set of compartmental Markov models to represent acquisition and heterosexual transmission of infection, with an imbedded progression model to represent the subsequent development of HPV-related disease (different stages of pre-cancerous cervical neoplasias, squamous cell carcinomas, adenocarcinomas and anogenital warts). The models are stratified by HPV type, age, sex and sexual activity-based risk group. The model time step was 1 month [16].
HPV types in the model are divided
Base-case vaccination programme
Fig. 4 shows the estimated impact of vaccinating 12-year-old girls at 80% coverage on diagnosed HPV 6/11/16/18-related disease for different assumptions about the duration of vaccine protection. Large reductions in incidence of cervical dysplasia, cervical cancer and anogenital warts cases are expected, provided vaccine-induced immunity lasts for 10 years or more, although the reduction in cervical cancer takes much longer to become apparent (note the change of scale).
One of the key biological
Discussion
We present here a dynamical model of heterosexual HPV transmission and the development of cervical dysplasia, cervical cancer and anogenital warts in the UK. Model results indicate that vaccinating 12-year-old girls at 80% coverage over 60 years will result in substantial reductions in cervical cancer and anogenital warts incidence. The magnitude of the reduction depends on a number of parameters, particularly the duration of vaccine protection. Catch-up campaigns reduce short-term incidence
Acknowledgments
We thank Kate Soldan and Liz Miller for insightful comments and help throughout this process, and the panel of referees who commented on an earlier version of this work. This work was funded by grants from the Department of Health Policy Research Programme reference numbers DOH 039/0030 and DOH 0039/031. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
References (51)
- et al.
Changing rates of adenocarcinoma and adenosquamous carcinoma of the cervix in England
Lancet
(2001) - et al.
Efficacy of a bivalent L1 virus-like particle vaccine in prevention of infection with human papillomavirus types 16 and 18 in young women: a randomised controlled trial
Lancet
(2004) - et al.
Cost-effectiveness analyses of human papillomavirus vaccination
Lancet Infect Dis
(2007) - et al.
Evidence for the coexistence of two genital HPV types within the same host cell in vitro
Virology
(2004) - et al.
Papillomavirus vaccines in perspective
Lancet
(2007) - et al.
Efficacy of a prophylactic adjuvanted bivalent L1 virus-like-particle vaccine against infection with human papillomavirus types 16 and 18 in young women: an interim analysis of a phase III double-blind, randomised controlled trial
Lancet
(2007) - et al.
Sexual behaviour in Britain: early heterosexual experience
Lancet
(2001) - et al.
Hysterectomy prevalence and adjusted cervical and uterine cancer rates in England and Wales
BJOG
(2001) The serological response to papillomaviruses
Semin Cancer Biol
(1999)- et al.
Costs of running a universal adolescent hepatitis B vaccination programme
Vaccine
(2005)
Induction of immune memory following administration of a prophylactic quadrivalent human papillomavirus (HPV) types 6/11/16/18 L1 virus-like particle (VLP) vaccine
Vaccine
Chapter 21: modelling the impact of HPV vaccines on cervical cancer and screening programmes
Vaccine
Epidemiologic classification of human papillomavirus types associated with cervical cancer
N Engl J Med
Against which human papillomavirus types shall we vaccinate and screen? The international perspective
Int J Cancer
European guideline for the management of anogenital warts
Int J STD AIDS
Quadrivalent vaccine against human papillomavirus to prevent high-grade cervical lesions
N Engl J Med
Quadrivalent vaccine against human papillomavirus to prevent anogenital diseases
N Engl J Med
The theoretical population-level impact of a prophylactic human papilloma virus vaccine
Epidemiology
Model for assessing human papillomavirus vaccination strategies
Emerg Infect Dis
Strategies for the introduction of human papillomavirus vaccination: modelling the optimum age- and sex-specific pattern of vaccination in Finland
Br J Cancer
Epidemiology of HPV 16 and cervical cancer in Finland and the potential impact of vaccination: mathematical modelling analyses
PLoS Med
The value of including boys in an HPV vaccination programme: a cost-effectiveness analysis in a low-resource setting
Br J Cancer
The epidemiological and economic impact of a quadrivalent human papillomavirus vaccine (6/11/16/18) in the UK
BJOG
Economic evaluation of human papillomavirus vaccination in the United Kingdom
BMJ
Estimating progression rates for human papillomavirus infection from epidemiological data
Med Decis Making
Cited by (86)
Age difference between heterosexual partners in Britain: Implications for the spread of Chlamydia trachomatis
2018, EpidemicsCitation Excerpt :Because this approach is simple and flexible it has been used in many STI transmission models, with few attempts to improve this method (Walker et al., 2012). Published models have either used fixed values for the assortativity constant or considered ranges of values (Barnabas et al., 2006; Choi et al., 2010). However, because this constant only summarizes mixing patterns (assuming that the assortativity constant itself does not depend on age), the computed patterns can differ from true mixing patterns.
Effect of HPV vaccination and cervical cancer screening in England by ethnicity: a modelling study
2018, The Lancet Public Health