Article Text
PDF
Abstract
Background For several decades, diagnoses of genital warts at genitourinary medicine (GUM) clinics in England had been increasing. In 2008, a national human papillomavirus (HPV) vaccination programme was introduced using the bivalent vaccine (types 16 and 18 only). A decrease in genital warts was not anticipated. However, rates of genital warts in GUM clinics have declined significantly since the introduction of the vaccine.
Methods Using data from GUM clinics across England, we analysed rates of genital warts by age, gender, sexual orientation and estimated vaccine coverage.
Results The reduction in rates of genital warts diagnoses at GUM clinics between 2009 and 2014 was 30.6% among young women aged 15–19 years and 25.4% among same age heterosexual young men. Overall there was an association showing higher warts reduction with increasing vaccination coverage with the largest declines in warts diagnoses observed in young women aged 15 years (50.9%) with the highest vaccination coverage. No such declines were observed in men who have sex with men (MSM) of the same age.
Conclusion The results of these ecological analyses are strongly in keeping with the bivalent HPV vaccine providing modest protection against genital warts.
- HPV
- VACCINATION
- GENITAL WARTS
Statistics from Altmetric.com
Introduction
The UK introduced a national human papillomavirus (HPV) vaccination programme using the bivalent HPV vaccine (Cervarix) in September 2008, providing the three-dose course in schools for all young women aged 12–13 years (routine cohort) alongside a catch-up programme for young women up to 18 years of age.
Coverage for the vaccination programme has been high, with over 80% of the routine cohort and between 40% and 70% of the catch-up cohort receiving all three doses.1 Use of the bivalent vaccine continued until September 2012, at which time it was replaced by the quadrivalent vaccine (Gardasil). Both vaccines have been shown to be highly effective against infection with HPV types 16 and 18.2 ,3 The quadrivalent vaccine has also demonstrated excellent efficacy against HPV types 6 and 113 responsible for the majority of cases of genital warts.4
In the UK, a reduction in genital warts was not anticipated during the initial years of the HPV vaccination programme using the bivalent vaccine. Rates of genital warts diagnoses in genitourinary medicine (GUM) clinics in England had been increasing sporadically since the 1970s. However, between 2008 and 2011, modest decline in the rates of genital warts diagnoses among women were observed at GUM clinics in England, which suggested a potentially cross-protective effect of the bivalent vaccine against genital warts.5 ,6
We analysed more recent data on genital warts diagnoses and bivalent vaccination coverage in England to determine whether this trend had continued and whether the effect was seen in both men and women.
Methods
Data collection
Data from the GUM Clinic Activity Dataset (GUMCADv2) submitted by GUM and integrated GUM/sexual and reproductive health clinics were extracted for 2009–2014. This is a mandatory dataset, providing disaggregate records of all attendances, testing and diagnoses at GUM clinics in England and has been reported to Public Health England since 2008; with full coverage from 2009.7 All records coded as a first diagnosis of genital warts and with an age at diagnosis from 15–24 were selected from the complete dataset; recurrent diagnoses were excluded from the analysis.
Data on three-dose bivalent vaccination coverage by academic year in England were obtained from published reports.1 These data were used to estimate female vaccination coverage by calendar year and single year of age.6 Rates of genital warts were calculated using published population estimates by sex and age for England, with the 2013 population taken as a proxy for 2014.8
Men who have sex with men (MSM) were defined as any man who had reported being either homosexual or bisexual on at least one clinic attendance. The denominator, that is the overall population of MSM, was taken to be 2.9% of the total male population based on data from the Natsal-3 study of sexual attitudes and lifestyles in the UK.9
Statistical analysis
Vaccine effectiveness could not be directly calculated as individual vaccine status was unknown. As there was considerable variation in vaccine coverage across age groups, we used a negative binomial regression model which included a continuous variable representing female three-dose vaccine coverage (as a proportion) for each age and year. This allowed us to estimate the incidence rate ratios (IRRs) for genital warts diagnoses comparing a population with no female vaccination (coverage=0) with a vaccinated population (coverage=1). To provide similar estimates for men, we made the assumption that male genital warts diagnoses were affected by the vaccination coverage among women of the same age. Analyses were conducted using Stata (V.11).
Results
Women
Between 2009 and 2014, there were substantial decreases in the rates of genital warts diagnoses among young women aged 15–19 years, who would have been eligible for bivalent vaccination as part of the routine programme (30.6% decrease; 685.8–475.7 per 100 000) and also in women aged 20–24 years (9.6% decrease; 698.8–631.8 per 100 000). The largest decline was observed in young women aged 15, 16, 17, 18 and 19 years (50.9%, 46.7%, 37.4%, 29.7% and 18.6%, respectively, all with p=<0.0001) (figure 1).
Rate of first case of genital warts diagnosed at a genitourinary medicine (GUM) clinic among men and women. The grey bars represent estimated human papillomavirus (HPV) vaccination coverage among women while the solid lines represent rates of genital warts among women and the dotted lines among men.
The estimated IRR for genital warts comparing vaccinated with unvaccinated populations under 19 years was 0.66 (95% CI 0.56 to 0.79), 0.65 (95% CI 0.56 to 0.75), 0.65 (95% CI 0.55 to 0.77) and 0.79 (95% CI 0.65 to 0.96) for young women aged 15, 16, 17 and 18 years, respectively (table 1). The IRRs for older women with lower vaccination coverage tended towards 1 or greater.
Incidence rate ratios (IRRs) for the effect of female human papillomavirus (HPV) vaccination on the rate of genital warts diagnoses among women, heterosexual men and MSM
Men
Declines in the rate of genital warts diagnoses were also seen among heterosexual men. There were marked decrease between 2009 and 2014 of 25.4% (from 269.0 to 200.6 per 100 000) in young men aged 15–19 years and by 12.2% (from 832.8 to 731.6 per 100 000) among men aged 20–24 years. Just as in women, the greatest decline in rates of genital warts was among young men aged 15, 16, 17, 18 and 19 years (37.5%, 31.7%, 29.6%, 30.4% and 21.2%, respectively, with p=<0.0001) (figure 1). Among MSM, there was evidence of a decrease in genital warts diagnoses over this period among MSM aged 16, (58.1%), 17 (25.6%) and 18 years (12.0%), although this was based on relatively few cases. However, there was still a small increase in the incidence of genital warts diagnoses between 2009 and 2014 in those aged 15–19 years combined (4.3% increase from 439.4–458.2 per 100 000) and a larger increase among MSM aged 20–24 years (34.4% increase from 1410.0 to 1894.4 per 100 000).
Among heterosexual men, the IRRs for genital warts in relation to vaccination of the female population of the same age, were 0.96 (95% CI 0.63 to 1.46), 0.79 (95% CI 0.64 to 0.97), 0.77 (95% CI 0.66 to 0.88), 0.72 (95% CI 0.64 to 0.82) and 0.86 (0.78 to 0.95) for men aged 15, 16, 17, 18 and 19 years, respectively (table 1). Among MSM, the equivalent IRRs did not reach statistical significance (0.59 (95% CI 0.27 to 1.28), 0.60 (95% CI 0.34 to 1.08), 0.72 (95% CI 0.43 to 1.21) and 0.96 (95% CI 0.64 to 1.46) for MSM aged 16, 17, 18 and 19 years, respectively). It was not possible to calculate an IRR for MSM aged 15 years as the number of cases was too small.
Discussion
This ecological study further strengthens the evidence of an unexpected association between bivalent HPV vaccination coverage and a decline in genital warts diagnoses in women.6 Importantly, the study provides new evidence of a herd-protective effect of HPV vaccination in young heterosexual men.
For women and heterosexual men, there were similar patterns of decreases in genital warts diagnoses. Similar decreases were not seen among MSM. While differences in sexual behaviour between heterosexual men and MSM may partly explain these differing trends, these data strongly suggest that heterosexual men are benefiting from high levels of vaccine-related protection against HPV in the female population. Age differences in sexual partnerships may account for the slightly older age distribution of protection observed among men as is seen in age distributions for STI diagnosis more generally.10 ,11
The potential biological mechanisms for cross-protection against genital warts via the bivalent vaccine are unclear. However, these results support a post-hoc analysis from the clinical trials of the bivalent HPV vaccine have shown estimated vaccine efficacy against types 6 and 11 of 34.5% in those who tested HPV-negative at the start of the trial. The authors of this paper hypothesised that a plausible mechanism for this cross-protection could involve cross-reactivity at the T-helper cell (CD4 receptor) level.12 Few studies have considered the reduction in genital warts after introduction of the bivalent vaccine in a population setting. One study conducted in the Czech Republic (where both the bivalent and quadrivalent vaccines have been used) compared genital warts acquisition according to vaccination status ascertained from a self-administered questionnaire. This study showed no evidence of protection against genital warts from the bivalent vaccine.13
We have previously considered in detail alternative factors which may have facilitated the decline in genital warts in the young heterosexual population.6 Changing sexual behaviour has been implicated in both the increase and decline of STIs over time.14 It is worth noting that no similar decline in the rates of gonorrhoea or genital herpes at a national level has been observed for these age groups over the same time period.10 There is no evidence that people are seeking healthcare for warts in different settings such as general practice (GP) surgeries.10 ,15 However, there have been some decline in chlamydia diagnoses over the same period which may be in part due to changes in service provision over this period, with an increasing number of clinics offering integrated GUM and contraceptive services. These changes in service provision may also have had some effect on genital warts diagnoses. However, the age-specific and gender-specific pattern seen for decline in genital warts was not apparent for chlamydia diagnoses, which suggests an effect of higher vaccination coverage rather than entirely a consequence of decreased chlamydia diagnoses.
Furthermore, there is evidence that HPV types 16 and 18 are implicated in a minority of genital warts diagnoses. A recent study found HPV type 16 and/or type 18 in over 50% swabs from warts lesions but laser-capture microdissection of these lesions suggested that HPV types 6 and 11 were the likely causative type for the vast majority of cases.4 While types 16 and 18 may well cause a small number of genital warts, the role of HPV types 6 and 11 is well established and these types would likely have accounted for the majority of cases in this study. It is also unlikely that use of the quadrivalent vaccine outside of the UK's national vaccination programme (ie, private uptake prior to the introduction of the quadrivalent vaccine in 2012) could account for the large decline in genital warts observed. There has been evidence of substantial decline in genital warts diagnoses in countries with a high coverage vaccination programme using the quadrivalent vaccine.16 It is possible that some reduction in genital warts diagnoses in England is due to individuals vaccinated with the quadrivalent vaccine outside of the UK. Among this population of GUM attenders aged 15–24 years, individuals born outside of the UK represented 9% of women, 9% of heterosexual men and 19% of MSM. However, only 4%, 3% and 11%, respectively, of women, men and MSM were born in countries who had adopted a quadrivalent vaccination programme prior to 2012. Furthermore, analyses restricted to individuals born in the UK showed very similar decline in the rates of genital warts diagnoses compared with all individuals (data not shown) so, while it is possible that migration of individuals from countries with a national quadrivalent vaccination programme could have had some impact on genital warts diagnoses, it is doubtful that this could explain the level of declines seen.
The suggestion of a modest cross-protective effect of the bivalent vaccine against genital warts in women and the associated herd-protective effect in heterosexual men could further improve its cost-effectiveness in a vaccination programme. Further research into the mechanisms of cross-protection to improve our understanding of HPV vaccine protection is warranted.
Key messages
The rates of genital warts in genitourinary medicine (GUM) clinics have declined in young women and men since the introduction of the human papillomavirus vaccination programme in England.
The decline is in keeping with the bivalent vaccine providing a modest cross-protective effect against genital warts in women.
Similar patterns of decrease in genital warts diagnoses in heterosexual men suggest a herd-protective effect.
Acknowledgments
The authors thank all clinics that report data to the Genitourinary Medicine Clinic Activity Dataset (GUMCAD V.2). They also thank Claire Tanner for assisting in formatting the figures for publication. They are grateful to Nick Andrews for advice on statistical analyses and to Sarah Woodhall for comments on the manuscript.
References
Footnotes
Handling editor Jackie A Cassell
Contributors This work was initiated and designed by MC, DM and KS. GH was responsible for the data collection and management. MC conducted the statistical analysis. MC and DM wrote the first draft of the manuscript. All authors contributed to and approved the final draft.
Ethics approval PHE has permission to handle data obtained by GUMCADv2 under section 251 of the UK National Health Service Act of 2006 (previously section 60 of the Health and Social Care Act of 2001), which was renewed annually by the ethics and confidentiality committee of the National Information Governance Board until 2013. Since then the power of approval of public health surveillance activity has been granted directly to PHE.
Funding This work was supported by Public Health England.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Data sharing statement Data on diagnoses of genital warts are publicly available. All data sources are cited in the reference list.