Increase of invasive meningococcal serogroup W disease in Europe, 2013 to 2017

Background The total incidence of invasive meningococcal disease (IMD) in Europe has been declining in recent years; however, a rising incidence due to serogroup W (MenW), predominantly sequence type 11 (ST-11), clonal complex 11 (cc11), was reported in some European countries. Aim The aim of this study was to compile the most recent laboratory surveillance data on MenW IMD from several European countries to assess recent trends in Europe. Methods In this observational, retrospective study, IMD surveillance data collected from 2013–17 by national reference laboratories and surveillance units from 13 European countries were analysed using descriptive statistics. Results The overall incidence of IMD has been stable during the study period. Incidence of MenW IMD per 100,000 population (2013: 0.03; 2014: 0.05; 2015: 0.08; 2016: 0.11; 2017: 0.11) and the proportion of this serogroup among all invasive cases (2013: 5% (116/2,216); 2014: 9% (161/1,761); 2015: 13% (271/2,074); 2016: 17% (388/2,222); 2017: 19% (393/2,112)) continuously increased. The most affected countries were England, the Netherlands, Switzerland and Sweden. MenW was more frequent in older age groups (≥ 45 years), while the proportion in children (< 15 years) was lower than in other age groups. Of the culture-confirmed MenW IMD cases, 80% (615/767) were caused by hypervirulent cc11. Conclusion During the years 2013–17, an increase in MenW IMD, mainly caused by MenW cc11, was observed in the majority of European countries. Given the unpredictable nature of meningococcal spread and the epidemiological potential of cc11, European countries may consider preventive strategies adapted to their contexts.


Introduction
The total incidence of invasive meningococcal disease (IMD) in Europe has been declining in recent years; however, a rising incidence of IMD caused by serogroup W (MenW) was reported in the United Kingdom (UK) in 2009 [1], then the Netherlands [2] and Sweden [3], following the spread of MenW IMD in South America since 2004 [4]. The isolated strains predominantly belong to the multilocus sequence typing (MLST) defined hypervirulent clonal complex 11 (cc11).
Some patients with MenW:cc11 infection have been reported to present with gastrointestinal symptoms, which is uncommon for patients with IMD caused by meningococci of other serogroups [5]. The case fatality rate (CFR) of MenW IMD was described to be more than twice that of IMD caused by other serogroups, and the proportion of MenW:cc11 in all IMD cases was found to be higher in adults than in infants [6].
In response to the rapid expansion of hypervirulent MenW:cc11 in the UK, a MenACWY conjugate vaccination programme was introduced in August 2015 for all adolescents 13-18 years of age, as well as university students up to 25 years of age [7]. In Italy, the programme was introduced for those 12-18 years of age in 2017 [8] and in the Netherlands for those 13-14 years of age in 2018, with a catch-up vaccination up to the age of 18 [9,10]. MenACWY vaccination has been recommended in Greece since 2011 for those 11-16 years of age [11] and in Austria since 2012 for those [10][11][12][13] year of age [12]. In contrast to earlier IMD epidemics where vaccines were not available or became available near the end of the epidemic, such as in New Zealand in the 1990s, several vaccines against MenACWY have been approved and are available in Europe [13,14].
The aim of the present study was to compile the most recent laboratory surveillance data on MenW IMD from several European countries to assess recent trends in Europe.

Setting and data collection
All national reference laboratories active in the European Meningococcal and Haemophilus Disease Society (EMGM) were invited to a workshop on serogroup W and Y meningococci in Europe, taking place in November 2016. This workshop was organised using EMGM's institutional support from Pfizer, though the company did not have any influence on the outline or content of the meeting.
Reference laboratories from 14 countries were asked to participate in this study. The Czech Republic, England, Finland, Germany, Greece, Italy, the Netherlands, Norway, Poland, Portugal, Spain, Sweden and Switzerland provided the following data in an Excel spreadsheet: the number and age distribution of IMD cases (total and MenW), laboratory surveillance rate of the reported cases and data on cc11 as a surrogate for the W lineages from 2013-16. The participating countries represented over two thirds of the total combined European Union (EU)/European Free Trade Association (EFTA) population.
After identifying and correcting missing or inconsistent numbers, the dataset was compiled and analysed. IMD numbers were based on publicly available national reporting, which is mandatory in all 13 participating countries. On average, 90% of the reported IMD cases were laboratory confirmed by the national reference laboratories, ranging from 60-100% depending on the country. To achieve a high quality standard on the serogrouping data, only MenW cases analysed by reference laboratories were included in the analysis, as  serogrouping in peripheral laboratories is not always performed by standardised methods. Reference laboratories reported the number of cases according to the laboratory criteria of the EU case definitions [15]. Typing data were matched at different levels with the statutory notification data. Preliminary 2017 data were collected on IMD and MenW cases, as at the time of writing sequence typing data was not yet available for many countries.
In England, national reporting is based on laboratoryconfirmed cases only; therefore, only such cases were included in this study [16]. Dutch and Polish data used in this study were also based on laboratory-confirmed cases. In the Czech Republic, Finland, Germany and Greece the reported numbers were based on mandatory notification data merged with laboratory surveillance data in a national database. As notification data were not available for the number of Spanish IMD cases in 2016-17, this number was extrapolated from the available laboratory surveillance data, representing approximately 80% of the notified cases in the years 2013-15.
Population and age distribution data from 12 of the 13 countries were obtained from Eurostat [17]. Data on the English population were acquired from the UK Office for National Statistics [18]. As more recent data were not yet available, the English population number from 2016 was used for 2017.

Typing
The serogroup was determined by slide agglutination or PCR. Depending on the country, MLST-either on Sanger sequence or next generation sequencing datawas used to determine if strains belonged to cc11. The Bacterial Isolate Genome Sequence Database (BIGSdb) was used for data matching [19].

Statistics
Average annual percentage change in incidence was calculated for each country using Poisson regression. A p value < 0.05 was considered statistically significant.  , this corresponds to an average annual incidence of 0.58 cases per 100,000 population. Serogroup W was responsible for 1,329 (13%) of the total number of IMD cases ( Table 1).

Incidence of invasive meningococcal diseases cases
The average annual incidence of overall IMD and MenW IMD varied considerably between individual countries. For IMD, the incidence per 100,000 population ranged from 0.31 cases in Italy to 1.34 in England. For MenW IMD, the incidence per 100,000 population varied from 0.008 cases in Portugal to 0.17 cases in the Netherlands and 0.30 cases in England (Figure 1). 23%; p < 0.001). The latter represents 61% of the total number of reported MenW cases in the 13 European countries during the overall study period. Though not statistically significant, there were also trends towards an increasing incidence in some other countries. (Table  1, Figure 2, Figure 3).

Age distribution
In  (Figure 4)

Sequence typing
Of the 936 reported cases of MenW IMD between 2013-16, 767 (82%) strain isolates were analysed by MLST and 615 (80%) belonged to cc11. The proportion of cc11 increased from 64% (64/100) in the year 2013 to 86% (265/307) in the year 2016 ( Figure 5) The proportion of sequence-typed MenW strains that belonged to cc11 varied considerably among countries, ranging from one in six in the Czech Republic to 31 in 34 in Switzerland (Figure 1).

Discussion
Laboratory surveillance and national health authority data from 13 European countries revealed an increase in MenW IMD in the period 2013-17. The MenW IMD incidence differed between participating countries and was highest in England and the Netherlands, as reported previously [1,2]. Significantly increasing incidences and proportions of MenW IMD among all IMD were observed in the majority of the participating countries, while the case numbers in other countries were too small to assess significance.
The increase seen in Europe follows a rising incidence of MenW in South America since 2004 [4]. More recently, a rising MenW incidence has been observed in Australia since 2013 [6] and in Canada since 2015 [20]. An epidemic spread of this MenW strain can be assumed for regions in the Americas, Australia and Europe. Notably, in 2015 an international outbreak related to the World Scout Jamboree in Japan was caused by the MenW:cc11 strain from the UK [21]. This was observed previously for other meningococci serogroups, including serogroup C:cc11 in the 1990s [22], serogroup A since the 1960s [23] and serogroup B since the 1970s [24]. The dynamics of such intercontinental epidemic waves is hardly predictable and the mechanisms behind different dynamics of spread are still unclear. Typing by next generation sequencing revealed the relatedness of the MenW strains from South America, England [25] and the Netherlands [2]. It is possible that cc11 in general has a particular capacity to spread between countries, as these strains have shown this already in the past [21,[26][27][28]. A possible experimental hypothesis is that cc11 strains are maintained in the population by frequent transmission events, despite a short duration of carriage [29].
It is noteworthy that the majority of cases of MenW IMD in Europe between 2013-16 occurred in those ≥ 45 years of age, with the highest number of cases found in patients ≥ 65 years of age. It is unclear whether this is the result of a circulation of strains in those ≥ 45 years of age or if colonised children and teenagers are the major source of transmission to the older population. Indeed, carriage of meningococci is most frequent in adolescents and young adults, with a peak around the age of 20 that decreases with age; however, studies on carriage of meningococci in adults and the elderly are rare [30]. From the existing evidence, it may be suggested that despite infrequent disease, carriage in adolescents and young adults is much more abundant than in infants and toddlers [30]. As target groups and coverages vary in different countries (e.g. case numbers in some countries are low) and some programs were introduced quite recently, it is still too early to conclude on the effect of the vaccinations. In theory, by vaccinating adolescents, acquisition of carriage should be prevented and herd protection induced. This may be a cost-effective approach, as it might not be possible to immunise all groups. Public health decision-makers should be aware of the rise of MenW and, taking into account the situations in their respective countries, they may consider including quadrivalent meningococcal ACWY in their adolescent vaccination programmes.
The present study demonstrates the power of laboratory networks to provide surveillance data on short notice. The EMGM has already proved to be a functioning laboratory network on earlier occasions. The present survey relied on Sanger sequencing information, as next generation sequencing has not been rolled out to a sufficient extent in all countries. However, the example of the MenW spread shows that for particular questions this technique is indispensable. The causative agent is a particular derivative of the cc11, which can only be elucidated by high-resolution techniques [3]. The function of BIGSdb as a platform allowing the integration of genome sequence data was first validated using meningococcal sequence data as an example [19]. Future European-level meningococcal disease surveillance databases linking public health reporting data with high-resolution sequencing data would be beneficial for outbreak detection. The respective information technology is available; however, clarification of data protection issues and an unambiguous nomenclature is a prerequisite.
One limitation of this study is the incomplete coverage of Europe. Only 13 of the 32 EU/EFTA countries joined this project. However, more than two thirds of the population of the EU/EFTA area was addressed. It should be noted that in the UK, only England participated. Further, because of varying reporting and surveillance structures, the ways that data were obtained differed between countries, as did the laboratory surveillance coverage. Under-reporting may also have occurred on different levels, as cases of IMD could have been misdiagnosed or misclassified [32]. In addition, cases of MenW IMD may have been missed if they were not serogrouped by a national reference laboratory.

Conclusion
In conclusion, this study showed a concerning trend of increasing MenW incidence in several European countries, which needs to be followed carefully. Revision of vaccination schedules that takes into account an individual country's situation may be considered to counteract this trend. Furthermore, this study shows the strength of laboratory networks to combine and evaluate data on short notice. In the case of increasing MenW IMD, serogrouping and basic typing data were indispensable for the evaluation.