Introduction
Gonorrhoea, caused by
Neisseria gonorrhoeae (
N. gonorrhoeae), is the second most prevalent bacterial sexually-transmitted infection globally [
1]. The infection represents a significant public health problem particularly because
N. gonorrhoeae has rapidly developed resistance to every antibiotic used against it to date [
2,
3] and sporadic case reports of highly antibiotic resistant
N. gonorrhoeae strains have led to increasing concerns about future treatment options [
4‐
7]. In the UK [
8] and US [
9] ceftriaxone monotherapy is currently the recommended first-line regimen for treating uncomplicated gonorrhoea, along with susceptibility testing and careful monitoring for treatment failure. Until recently, dual therapy with ceftriaxone and azithromycin was recommended in these countries in an attempt to combat multi-drug resistant strains, but evidence had suggested that azithromycin resistance was increasing, the azithromycin dose may be insufficient to clear infections, and higher azithromycin doses could accelerate resistance selection [
8]. There have been recent reports, however, of gonorrhoea infections that are resistant to ceftriaxone [
10].The latest WHO treatment guidelines recommend dual therapy over single therapy, unless local resistance data are able to inform the choice [
11]. In England, prior to the SARS-CoV-2 (“COVID-19”) pandemic, the number of cases of gonorrhoea seen in sexual and reproductive health care settings was increasing: in 2019, the number of diagnoses was 70,936, a 26% increase compared to 2018 [
12,
13]. Of those with known sexual orientation, 33,853 (~ 50%) diagnoses were in men who have sex with men (MSM), 15,253 in heterosexual men, and 17,826 in heterosexual women. Symptoms, when present, include burning with urination and genital discharge, among others [
14]. Possible complications include disseminated infection, epididymitis in men, and pelvic inflammatory disease, chronic pelvic pain, ectopic pregnancy and tubal factor infertility in women [
15,
16]. However, infection is frequently asymptomatic, particularly in women, meaning the prevalence of infection is higher than diagnoses alone suggest [
14,
15,
17,
18].
An important potential future control option for gonorrhoea is vaccination [
19]. Until recently, efforts to develop an effective vaccine were impeded by the highly antigenically-variable surface of
N. gonorrhoeae, and a lack of suitable animal models [
20]. However,
N. gonorrhoeae shares 80–90% homology of its primary sequences with
N. meningitidis, a bacterium which can cause meningitis and septicaemia [
21]. Thus, a vaccine against
N. meningitidis could potentially protect against gonorrhoea infection. A retrospective case-control study among individuals aged 15–30 years attending sexual health clinics in New Zealand who were vaccinated with an outer membrane vesicle meningococcal B vaccine MeNZB, found that vaccination provided partial protection (31% efficacy, 95%CI, 24–39%) against gonorrhoea infection [
22]. Two studies using 4CMenB vaccine, a protein-based vaccine designed to protect against group B meningococcal disease routinely offered to infants in the UK, have also shown an impact on gonorrhoea. A retrospective case-control study among 16–23-year-olds in New York City and Philadelphia estimated the vaccine efficacy to be 40% (95%CI 23–53%) [
23] and a case-control study of the same vaccine in adolescents and young adults in South Australia estimated the two-dose vaccine effectiveness against gonorrhoea at 32.7% (95%CI 8.3–50.6%) [
24]. A proof of concept trial of 4CMenB vaccination in adolescents and adults against gonorrhoea is currently underway (
ClinicalTrials.gov: NCT04350138 [
25]).
The public health impact in England of an adolescent national immunisation programme (NIP) with the
N. meningitidis vaccine 4CMenB (trade name Bexsero®, GSK) that also provides partial protection against gonorrhoea infection has not been estimated. Also unknown is the optimal implementation strategy: vaccination with 4CMenB is unlikely to provide life-long protection against gonorrhoea [
22,
26], so it will be important to factor in the age-related incidence of gonorrhoea in determining the optimal age for vaccination, and consider whether catch-up or booster vaccination would be useful. In 2019, incidence was low in 13–14-year-olds, 1.2 and 10.1 per 100,000 in males and females respectively, but jumped to 199.7 and 334.7 per 100,000 in males and females aged 15–19 years respectively. Incidence peaked in 20–24-year-olds for males (588.4 per 100,000) and females (395.9 per 100,000), declining in older age groups [
13].
In England, the existing routine immunisation schedule offers vaccination to teenagers against human papillomavirus (HPV) starting in school year 8 (12–13 years), and against meningococcal disease-causing capsular groups A, C, W and Y (MenACWY) in school years 9 and 10 (13–15 years) [
27]. The MenACWY vaccine is the last vaccine given in the childhood NIP, thus this provides an opportunity, logistically, to offer 4CMenB vaccination to help protect individuals against gonorrhoea, before most have commenced sexual activity, but close in age to when individuals are at increased risk of infection. The MenACWY NIP has around 85% uptake [
28] plus approximately 40% uptake of time-limited catch-up vaccination of 18–20-year-olds through general practice (GP) [
29,
30], with vaccination offered to girls and boys. (The HPV NIP had 88% coverage for the first dose in females before the SARS-CoV-2 pandemic (2018–2019). Coverage is lower in males (71% coverage for the first dose in 2020–2021), but the NIP has only recently been extended to males, with its introduction impacted by the SARS-CoV-2 pandemic [
31].)
An explorative modelling study, not specific to any setting, found that adolescent vaccination against
N. gonorrhoeae could achieve a substantial reduction in gonorrhoea prevalence even with < 50% vaccine efficacy [
32]. There has been limited investigation of the impact of a vaccine with both relatively low efficacy and short duration, and of different vaccination implementation strategies. Moreover, for England, the only public health impact of vaccination that has been addressed by modelling to date is for MSM [
33]. To address these research gaps, we developed a mathematical model to address key considerations for a 4CMenB adolescent NIP from an English health policy perspective, namely, to explore the population impact, and the optimal implementation strategy, of
N. gonorrhoeae vaccination with 4CMenB given plausible ranges for efficacy and uptake.
Discussion
Principal findings
Cohort vaccination of 14-year-olds in England with 4CMenB against gonorrhoea assuming 31% efficacy and 85% uptake could lead to a substantial reduction in new gonorrhoea infections, particularly among adolescents. For example, over 10 years, 50,000 (95%CrI 31,000-80,000) or 10% (95%CrI 8–13%) of heterosexual incident gonorrhoea infections could be averted, rising to 25% (95%CrI 17–33%) or 849,000 (95%CrI 476,000-1,568,000) over 70 years. In the short-term (10 years), around 54% more infections could be prevented with the addition of catch-up vaccination. Booster vaccination, meanwhile, could lead to around 61% more infections being prevented in the long-term (70 years), and could help mitigate a short duration of vaccine protection. For a vaccine with higher efficacy (50%), a third of infections among younger ages (13–18-year-olds) could be averted over ten years, increasing to over half of infections over 70 years.
Strengths and limitations of this study
Our study is the first to our knowledge to explore the impact of an NIP for 4CMenB with partial efficacy on gonorrhoea in the heterosexual English population. The transmission dynamic model incorporates considerable complexity with heterogeneity (and model compartments) by age group, sex and sexual behaviour, to describe the transmission of gonococcal infection. The parameters used are strongly informed by large studies of sexual behaviour and infection rates specific to England. We explored a range of vaccine efficacies that were based directly on study data, and considered a number of possible implementation strategies relevant for policy makers and commissioners. Since our study was conducted, a matched cohort study of the impact of 4CMenB on gonorrhoea infection among teenagers and young adults in Southern California has reported an estimated vaccine efficacy of 46% (95%CI 34–86%), which is in line with that reported by other studies, if slightly higher [
42].
There is uncertainty in the underlying data informing our model parameters, particularly around natural history (e.g., rate of recovery, probability of infection, percentage symptomatic), patterns of sexual mixing, and sexual behaviour (in particular among 13–15-year-olds). We incorporated parameter uncertainty in our estimates of impact by sampling from plausible ranges, and used prevalence data to validate our model, but did need to widen the sampling ranges in some age groups to generate a workable model. We compared the effect of key vaccination parameters on model predictions in our analyses. Reassuringly, the predicted impact in terms of percentage reduction in incidence was not sensitive to the base rate of infections though vaccine impact would be greater if the number of imported infections were fewer. The duration of 4CMenB vaccine protection against
N. gonorrhoeae is unknown. Our base case assumption of an average 6 years’ protection (with waning in an exponential fashion) is based on the observed decline in anti-OMV antibodies over time [
26,
43,
44]. This may be conservative given immunity may involve more than just serum bactericidal antibodies. To explore the impact of this parameter we also considered vaccine impact with longer and shorter protection levels.
As with any modelling study, populations are simulated and there are necessarily simplifying assumptions, such as assuming equal population by age between females and males, and applying the same sexual behaviour parameters (but not transmission probabilities) to both sexes. However, this also enables multiple vaccine scenarios and implementation strategies to be compared easily. We did not specifically consider MSM in our model, although we did stratify the model by sexual activity and include importation of infections which approximates bridging from MSM.
The benefits of vaccination may be higher than we have estimated because we only consider transmission through vaginal intercourse (although diagnosis data used for fitting do not specify anatomical location) and have not included possible co-infections, such as with HIV, that could increase transmission rates. In this model we only considered gonorrhoea infection irrespective of accompanying acute symptoms and possible complications - the natural history of disease, and in particular, the risk of longer-term outcomes, is more uncertain than for infection. Reducing associated complications of gonorrhoea infection and reductions in antimicrobial resistance would add further value to an NIP against gonorrhoea.
Strengths and limitations compared to other studies
A number of other, previous transmission-dynamic models have considered the role of vaccination against gonococcal infection, though in contrast to our work, these have focused on either vaccination in MSM [
33,
45] or have been in a heterosexual population, but not focused on a particular geographical setting [
32]. Whittles et al. [
33] developed a stochastic transmission-dynamic model in MSM in England, considering both antibiotic resistant and sensitive strains and evaluating a large number of hypothetical vaccine profiles. Based on data available at the time, they considered three scenarios using parameters designed to align with properties of the 4CMenB vaccine (2 to 4 year duration of vaccine protection) and found a 7% (0–23%) reduction in incidence in MSM after 10 years with vaccination before sexual debut, assuming no emergence of antibiotic resistance. This is similar to the incidence reduction we find in our heterosexual model where we implement vaccination at age 14 years. Greater vaccine impact was seen when vaccination was implemented at attendance at sexual health clinics (75%, 40–98%) or at gonorrhoea diagnosis (41%, 18–65% assuming 100% vaccine uptake). The authors noted that the before sexual debut strategy in MSM was ineffective, and the vaccination on attendance strategy was insufficient to meet the WHO target of reducing gonorrhoea incidence by 90% during 2018–2030 [
46] in 75% of the simulations. Hui et al. [
45] also modelled potential vaccine impact of
N. gonorrhoeae vaccines in an MSM population. The model considered infection across multiple anatomical sites and is tailored to an Australian setting. They estimated considerably larger reductions in infection, with a 62% prevalence reduction within 2 years, but assumed the vaccine has 50% efficacy against infection with 80% of MSM attending STI clinics annually (high compared to a UK context) and 30% of these choosing to be vaccinated. The results of these models highlight the considerable reductions that can be achieved when targeting a higher-risk group, however the impact critically depends on accessing this population to offer vaccination. Craig et al. [
32] used an individual based model to estimate the impact of vaccination against gonorrhoea in a general heterosexual population, including complex sexual mixing patterns (regular, short-term, and concurrent partnerships) though not tailored to a specific country. They considered a number of hypothetical vaccine profiles in terms of efficacy and duration of protection with vaccination implemented before sexual debut at 13 years of age. Similarly to our study, the authors found a substantial effect on gonococcal prevalence could occur even with a vaccine that was partially efficacious. They estimated a 40% reduction in prevalence after 20 years with a vaccine of 20% efficacy and 20 years' duration of protection.
Implications for policy makers (meaning of the study)
A partially-effective vaccine against gonorrhoea infection could have an important population impact on gonorrhoea, and could therefore add substantial value to the existing benefit of the vaccine against invasive meningococcal disease and in combating the spread of antimicrobial resistance. The global health sector strategy [
46] called for a 90% reduction in the incidence of
N. gonorrhoeae by 2030, however the latest WHO global progress report highlighted that increased efforts are needed to achieve this [
47]. A sizeable impact on gonorrhoea incidence, particularly among adolescents, is estimated in our models with only 31% vaccine efficacy and a relatively short average duration of vaccine protection. In the UK this programme could be implemented in schools alongside the existing MenACWY vaccine and 3-in-1 Td/IPV booster against tetanus, diphtheria and polio offered to 13–15-year-olds [
10]. The 4CMenB vaccine would offer some added protection against non-B capsular groups, as well as direct protection against meningococcal group B, but is unlikely to replace the MenACWY product given the high vaccine-induced protection with the conjugate MenACWY vaccine, particularly against carriage, which is important in teenagers.
Catch-up and booster vaccination could be considered in order to increase the short- and long-term impact of an NIP designed to protect against gonorrhoea. Targeting older age groups for cohort vaccination may offer increased protection at the ages of highest incidence for gonorrhoea, but at the cost of missing vaccination before sexual debut for some individuals and likely reduction in vaccine uptake in older ages, which has been observed in other programmes. Targeting of higher-risk groups, such as MSM, should also be considered, to offer benefits against disease to this community, but also because MSM who also have sex with women can import infections into the heterosexual population; our model suggests vaccination of a heterosexual population is considerably more effective when imported infections are lowered (which could be achieved through vaccinating MSM).
In this study we only considered the benefit of vaccination in terms of gonococcal incidence reduction, however, there would be additional benefits of an adolescent 4CMenB programme in terms of meningococcal disease reduction. Antibiotic resistance is a serious and urgent threat and high levels of resistance have developed against a succession of antibiotic treatments for gonorrhoea. There is emerging resistance to our current last line treatment option for this infection, thus prevention through vaccination offers a vital tool to combat this threat.
Unanswered questions/future research
Our model suggests that the introduction of adolescent vaccination with 4CMenB would provide an important reduction in gonococcal incidence. To evaluate the full benefit of the vaccine, an extended gonorrhoea model which simultaneously considers the impact of the vaccine on gonorrhoea symptoms, including long-term complications of gonorrhoea infection, and on meningococcal infection would be valuable. Allowing for different sexual behaviour between women and men, whilst still balancing the number of partnerships overall, and further including higher-risk groups such as men who have sex with men, would also be useful. Incorporating the implications for combating antibiotic resistance and exploring the optimum age of vaccination if not aligned with an existing NIP would be beneficial as well. Following the public health impact estimates, a health economic evaluation would be needed to enable the Joint Committee on Vaccination and Immunisation (JCVI) to make decisions regarding recommendations for the vaccine.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.