Scholarly article on topic 'Meningococcal carriage in Dutch adolescents and young adults; A cross-sectional and longitudinal cohort study'

Meningococcal carriage in Dutch adolescents and young adults; A cross-sectional and longitudinal cohort study Academic research paper on "Clinical medicine"

CC BY-NC-ND
0
0
Share paper
Academic journal
Clinical Microbiology and Infection
OECD Field of science
Keywords
{Adolescents / Carriage / "Diagnostic tests" / Longitudinal / " Neisseria meningitidis " / "Risk factors"}

Abstract of research paper on Clinical medicine, author of scientific article — Mariëtte B. van Ravenhorst, Merijn W. Bijlsma, Marlies A. van houten, Veerle M.D. Struben, Annaliesa S. Anderson, et al.

Abstract Objectives Current information on rates and dynamics of meningococcal carriage is essential for public health policy. This study aimed to determine meningococcal carriage prevalence, its risk factors and duration in the Netherlands, where meningococcal C vaccine coverage is >90%. Several methods to identify serogroups of meningococcal carriage isolates among adolescent and young adults were compared. Methods Oropharyngeal swabs were collected from 1715 participants 13–23 years of age in 2013–2014; 300 were prospectively followed over 8 months. Cultured isolates were characterized by Ouchterlony, real-time (rt-) PCR or whole-genome sequencing (WGS). Direct swabs were assessed by rt-PCR. Questionnaires on environmental factors and behaviour were also obtained. Results A meningococcal isolate was identified in 270/1715 (16%) participants by culture. Of MenB isolates identified by whole genome sequencing, 37/72 (51%) were correctly serogrouped by Ouchterlony, 46/51 (90%) by rt-PCR of cultured isolates, and 39/51 (76%) by rt-PCR directly on swabs. A sharp increase in carriage was observed before the age of 15 years. The age-related association disappeared after correction for smoking, level of education, frequent attendance to crowded social venues, kissing in the previous week and alcohol consumption. Three participants carried the same strain identified at three consecutive visits in an 8-month period. In these isolates, progressively acquired mutations were observed. Conclusions Whole genome sequencing of culture isolates was the most sensitive method for serogroup identification. Based upon results of this study and risk of meningococcal disease, an adolescent meningococcal vaccination might include children before the age of 15 years to confer individual protection and potentially to establish herd protection.

Academic research paper on topic "Meningococcal carriage in Dutch adolescents and young adults; A cross-sectional and longitudinal cohort study"

Accepted Manuscript

Meningococcal carriage in Dutch adolescents and young adults; A cross-sectional and longitudinal cohort study

Mariette B. van Ravenhorst, Merijn W. Bijlsma, Marlies A. van houten, Veerle M.D. Struben, Annaliesa S. Anderson, Joseph Eiden, Hao Li, Kathrin U. Jansen, Hal Jones, Nicholas Kitchin, Louise Pedneault, Elisabeth A.M. Sanders, Dr Arie van der Ende

PII: S1198-743X(17)30096-4

DOI: 10.1016/j.cmi.2017.02.008

Reference: CMI 856

To appear in: Clinical Microbiology and Infection

Received Date: 11 October 2016

Revised Date: 2 February 2017

Accepted Date: 4 February 2017

Please cite this article as: Ravenhorst MBv, Bijlsma MW, van houten MA, Struben VMD, Anderson AS, Eiden J, Li H, Jansen KU, Jones H, Kitchin N, Pedneault L, Sanders EAM, van der Ende A, Meningococcal carriage in Dutch adolescents and young adults; A cross-sectional and longitudinal cohort study, Clinical Microbiology and Infection (2017), doi: 10.1016/j.cmi.2017.02.008.

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

1 Meningococcal carriage in Dutch adolescents and young adults; A cross-sectional and

2 longitudinal cohort study

3 Mariette B. van Ravenhorst1,2, Merijn W. Bijlsma3, Marlies A. van houten2, Veerle M.D. Struben2,

4 Annaliesa S. Anderson4, Joseph Eiden4, Li Hao4, Kathrin U. Jansen4, Hal Jones4, Nicholas Kitchin5,

5 Louise Pedneault4, Elisabeth A.M. Sanders1, Arie van der Ende6

6 1. Department of Immunology and Infectious diseases, Wilhelmina Children's Hospital, University Medical Centre

7 Utrecht, Utrecht, The Netherlands. 2. Research Centre Linnaeus Institute, Spaarne Hospital, Hoofddorp, The

8 Netherlands. 3. Academic Medical Center, Centre of Infection and Immunity Amsterdam (CINIMA), Department of

9 Neurology, Amsterdam, The Netherlands. 4. Pfizer Vaccine Research & Development, Pearl River, NY, USA. 5.

10 Pfizer Vaccine Research & Development, Maidenhead, UK. 6. Academic Medical Center, Centre of Infection and

11 Immunity Amsterdam (CINIMA), Department of Medical Microbiology and the Netherlands Reference Laboratory for

12 Bacterial Meningitis, University of Amsterdam, Amsterdam, The Netherlands.

14 Running title: Meningococcal carriage in Dutch adolescents.

15 Keywords: Neisseria meningitidis; Carriage; Adolescents; Risk factors; Longitudinal; Diagnostic tests

17 Corresponding author:

18 Dr. Arie van der Ende, Department of Medical Microbiology, Netherlands Reference Laboratory for

19 Bacterial Meningitis, Academic Medical Centre, University of Amsterdam, PO Box 22660, 1100DD

20 Amsterdam, The Netherlands. E: A.vanderende@amc.uva.nl, T: +31205664862, Fax: +31 20 5669606

22 Text word count: 3499/3500 words

23 ABSTRACT

24 Objectives: Current information on rates and dynamics of meningococcal carriage is essential for public

25 health policy. This study aimed to determine meningococcal carriage prevalence, its risk factors and

26 duration in the Netherlands, where meningococcal C vaccine coverage is >90%. Several methods to

27 identify serogroups of meningococcal carriage isolates among adolescent and young adults were

28 compared.

29 Methods: Oropharyngeal swabs from 1715 subjects 13-23 years of age were collected in 2013-2014;

30 300 were prospectively followed over 8 months. Cultured isolates were characterized by Ouchterlony, rt-

31 PCR or whole-genome sequencing (WGS). Direct swabs were assessed by rt-PCR. Questionnaires on

32 environmental factors and behavior were also obtained.

33 Results: A meningococcal isolate was identified in 270/1715 (16%) participants by culture. Of MenB

34 isolates identified by WGS, 37/72 (51%) were correctly serogrouped by Ouchterlony, 46/51 (90%) by rt-

35 PCR of cultured isolates, and 39/51 (76%) by rt-PCR directly on swabs. A sharp increase in carriage was

36 observed before the age of 15 years. The age-related association disappeared after correction for

37 smoking, level of education, frequent attendance to crowded social venues, kissing in the previous week

38 and alcohol consumption. Three subjects carried the same strain identified at three consecutive visits in

39 an 8 month period. In these isolates, progressively acquired mutations were observed.

40 Conclusions: WGS of culture isolates was the most sensitive method for serogroup identification. Based

41 upon results of this study and risk of meningococcal disease, an adolescent meningococcal vaccination

42 might include children before the age of 15 to confer individual protection and potentially to establish herd

43 protection.

44 Abstract word count: 250/250 words

45 INTRODUCTION

46 Neisseria meningitidis (Nm) is a common colonizer of the upper respiratory tract in asymptomatic carriers,

47 but may occasionally cause invasive meningococcal disease (IMD) [1]. Of the 12 Nm serogroups, six

48 serogroups (A, B, C, X, W and Y) cause the majority of IMD worldwide. Whereas the overall carriage

49 prevalence in the population is estimated to be ~10%, the incidence rate of IMD per 100,000 population

50 varies between 0.5 in North America up to 1000 in epidemics settings [2, 3]. While IMD incidence is

51 highest in infants under 5 years of age, meningococcal carriage prevalence is low at this early age.

52 Carriage increases during childhood and peaks at around 24% among young adults and declines to

53 around 8% in older adults [4].

54 Experience with serogroup C (MenC) polysaccharide conjugate vaccines demonstrated herd protection

55 through reduced colonization and transmission of MenC after vaccination of children and adolescents [2,

56 5]. In the Netherlands, a single MenC conjugate vaccine was offered to all children aged 1-18 years in

57 2002. Routine vaccination at 14 months was subsequently introduced [6]. The vaccine coverage of the

58 population was estimated at 94% in 2002 [7]. At present, meningococcal serogroup B (MenB) is the

59 predominant cause of IMD in Europe [8, 9]. Only recently, MenB vaccines have been made available

60 through the licensure of two protein-based multicomponent vaccines: bivalent rLP2086, Trumenba®,

61 Pfizer and 4CMenB, Bexsero®, GSK [10, 11]. A recent study among University student showed a

62 reduction in overall carriage after a second dose of 4CMenB, but no impact was demonstrated for MenB

63 carriage [12]. Currently, European carriage data are mostly based on studies from the UK and France

64 [13], however, carriage characteristics change over time and may differ by country.

65 In this study, we compared different methods to identify meningococcal serogroups by culture and

66 Ouchterlony, rt-PCR or whole-genome sequencing (WGS) of meningococcal carriage isolates among

67 Dutch adolescent and young adults. Furthermore, we determined the prevalence of meningococcal

68 carriage, its risk factors and duration.

69 METHODS

70 Design and participants

71 The study was conducted in the Netherlands between January 2013 and March 2014. Healthy

72 participants aged 13-23 years were recruited randomly and solely based on age from 15 educational

73 institutions. Exclusion criteria included previous MenB vaccination, antibiotic use during the month prior to

74 enrolment, or participation in any other clinical trial with an investigational drug. Written informed consent

75 was obtained from both parents/guardians of participants aged <18 years and from all participants.

76 Amongst all participants, all subjects in the last year of secondary school who were included in the year

77 2013 were enrolled in the longitudinal study cohort and had study visits 3 times within 8 months after

78 enrolment, twice before and once after starting tertiary school or universities. The study was conducted in

79 compliance with ethical principles originating from the Helsinki Declaration, within the guidelines of Good

80 Clinical Practice and the study was registered at the Dutch Trial register (www.trialregister.nl; NTR3785).

82 Clinical procedures

83 Trained research nurses visited schools and universities to collect swabs, demographic data and

84 provided questionnaires on environmental factors and behavior based on previously identified risk factors

85 [14-16] . Two oropharyngeal swabs, each of the tonsils/the tonsillar fossa and the posterior pharynx were

86 simultaneously collected to compare different methods for identification of meningococcal serogroups.

87 One was inoculated on a Thayer-Martin agar and incubated at 37°C with 5% CO 2 immediately after

88 swabbing and the second ("Direct swab" only collected at baseline from participants who were included

89 in the year 2013) was placed directly into storage transport media (STM, DIGENE, Biomerieux), and was

90 kept at 4°C on study location. Within five hours from collection, the samples arrived at the Netherlands

91 Reference Laboratory for Bacterial Meningitis (NRLBM, Amsterdam, the Netherlands) for microbiological

92 processing.

94 Bacterial identification

95 Meningococcal isolates were isolated at the NRLBM as previously described [17]. Up to a maximum of 5

96 colonies of isolates identified as N. meningitidis were frozen in Microbank tubes and stored at minus

97 80°C. Direct swabs were also stored at this temperature. Direct swabs and frozen colonies were shipped

98 on dry ice to Pfizer Vaccine Research and Development for further identification by rt-PCR (both swabs)

99 and WGS (cultured swabs only) (Pearl River, NY, USA).

100 Phenotyping cultured isolates. Serogrouping of cultured isolates was performed by means of

101 microprecipitation in a modified Ouchterlony assay for serogroup A, B, C, E, W, X, Y and Z as previously

102 described [18].

103 Real-time polymerase chain reaction (rt-PCR) was performed on cultured isolates in the subset of

104 participants from whom two swabs were collected as previously described [19]. Isolates negative for

105 group-specific assays but positive for porA or ctrA were defined as non-groupable meningococcal isolates

106 (NG). Direct rt-PCR assays were performed on swab transport medium (without culture) of the second

107 swab to detect pork, ctrA and genetic targets diagnostic for MenB [19].

108 WGS of cultured isolates and sequence analysis. All meningococcal isolates were analyzed by WGS

109 as previously described [20], and genogroup was identified by capsule locus genes [21]. Isolates without

110 the capsule locus but positive for porA were identified as NG meningococcal isolates. Phylogenomic

111 analysis using core genome sequences was conducted in Harvest Suite [22]. Genomes of carriage

112 isolates were compared and ML phylogenetic trees were built through Parsnp, a fast core-genome multi-

113 aligner. Then the alignment and trees were visualized in Gingr and iTOL [23].

115 Statistical analyses

116 This study was a descriptive epidemiological Nm carriage study. On the basis of confidence interval (CI)

117 estimates, an overall sample size of 1700 subjects was chosen. Using the Clopper-Pearson exact

118 method, a conservative estimate for MenB carriage prevalence of 2% would result in a 95% CI between

119 1.4% to 2.8% within a sample size of 1700 individuals [24]. Continuous variables were presented as

120 means with standard deviation (SD), and categorical variables as numbers and percentages. Differences

121 in proportion were tested with the Fisher's exact test. Statistical tests were 2-sided and a P-value below

122 0.05 was considered statistically significant. Data were analyzed using SPSS statistics 22 (IBM).

123 Serogroup specific meningococcal carriage was defined by WGS of cultured isolates. For laboratory

124 methods comparison analysis, WGS of cultured isolates was used as reference. Possible risk factors for

125 meningococcal carriage were determined using results of the first visit based on previous publications

126 [14, 15, 25]. The association of age and carriage was evaluated with logistic regression analysis. We

127 estimated the univariate association of age with carriage as well as an estimate corrected for all identified

128 confounders. A confounder was defined as a variable that had a significant association with both age and

129 carriage, and that altered the correlation coefficient of age by at least 10% if added to the model. A

130 multivariate logistic regression model that included all evaluated risk factors for carriage simultaneously

131 was also developed. Model stability was evaluated by performing stepwise backward model selection

132 based on likelihood ratio testing with a P-in of 0.05 and P-out of 0.1. Acquisition rate was defined by the

133 number of new carriers for the 8 months study period divided by the total person-time of all initially

134 negative subjects. Acquisition was assumed to have occurred at the time point halfway between

135 assessments.

136 RESULTS

137 Participants' characteristics

138 At baseline, 1727 subjects were assessed for eligibility and 1715 (99%) were enrolled (Figure 1). Median

139 age at inclusion was 16.9 years (±SD 2.0) and 1056/1727 (62%) of the subjects were female. Of all

140 subjects, 1310/1715 (76%) subjects were recruited from secondary school and 405/1715 (24%) from

141 tertiary/University school. The number of enrolled participants was 906/1715 (53%) and 809/1715 (47%)

142 for the years 2013 and 2014, respectively. At the first visit, a meningococcal isolate was identified by

143 culture of oropharyngeal swabs in 270 (16%) of 1715 participants.

145 Characterization of Meningococcal serogroups of cultured isolates at the first visit

146 WGS results were available for 267/270 (99%) cultured isolates (Table 1). The most prevalent serogroups

147 identified by WGS were MenB, with 72/267 (27%) isolates, followed by MenX (38/267; 14%) and MenY

148 (34/267; 13%). Of isolates identified by WGS, 82/267(31 %) isolates did not harbor the capsule locus in

149 the genome sequence, and were therefore identified as capsule null.

150 Ouchterlony was available for 268/270 (99%) cultured isolates (Table 1). MenB was the most common

151 serogroup identified by the modified Ouchterlony with 37/268 (14%) isolates, followed by MenY (20/268;

152 7%) and MenE (17/268; 6%). Of isolates determined by Ouchterlony, 180/268 (67%) isolates were

153 identified as Non-Groupable (NG).

154 rt-PCR was performed on a subset of 177/270 (66%) cultured isolates (Table 2). Of these cultured

155 isolates, 48/177 (27%), 19/177 (10%) and 16/177 (9%) were identified by rt-PCR as MenB, MenX and

156 MenE, respectively. Identified by rt-PCR, 25/177 (14%) isolates were identified as NG (ie. negative for all

157 8 capsular types). rt-PCR was positive for more than one group-specific assay in 37/177 (21 %) isolates

158 and a serogroup was therefore not assigned.

160 MenB identified by direct rt-PCR at the first visit

161 A direct swab for rt-PCR was collected per protocol from 906/1715 (53%) subjects. Direct rt-PCR

162 targeting MenB was available for 902/906 (99.5%) swabs. MenB was detected by direct rt-PCR in 49/902

163 (5%) swabs.

165 Comparison of conventional diagnostic methods for detecting carriage serogroup

166 Of isolates with a MenB capsule locus identified by WGS of cultured isolates, 37/72 (51 %) were correctly

167 serogrouped by Ouchterlony, while 34/72 (47%) were NG (Table 1). Of isolates with genogroup X or Y,

168 Ouchterlony resulted in the corresponding phenotype in 3/38 (8%) and 13/34 (59%), respectively. Of

169 isolates identified by WGS of cultured isolates, only 5/267 (2%) harbored the MenC polysaccharide

170 capsule gene; none of these were detected by Ouchterlony.

171 Of cultured isolates assessed by rt-PCR, 144/177 (64%) were classified in accordance with the

172 WGS assignment (Table 2). Though both WGS and rt-PCR of cultured isolates identified a MenB carriage

173 prevalence of 27%, the detection of MenB by the two methods showed discrepancies, e.g. MenB

174 assignments by rt-PCR that were not confirmed by WGS and vice-versa (Supplemental Figure 1). The

175 sensitivity of rt-PCR to identify MenB among cultured isolates was 90% compared to WGS as reference.

176 Of isolates that did not harbor the capsule locus identified by WGS, 25/52 (48%) were NG by rt-PCR

177 (Table 2). In one isolate rt-PCR assigned a serogroup (MenW) that did not correspond with the WGS

178 result (MenX).

179 Direct rt-PCR had a sensitivity of 76% (39/51) when compared to WGS of cultured isolates

180 derived from the partner swab (Supplemental Figure 1). Of isolates identified by WGS, 12/51 (24%) MenB

181 isolates were not confirmed by direct rt-PCR. In addition, direct rt-PCR was positive for MenB in 10

182 additional samples that were not confirmed by WGS of cultured isolates. Of these samples, 9/10 (90%)

183 samples were negative for meningococci by culture and 1/10 (10%) isolate lacked the capsule locus by

184 WGS.

186 Age, carriage and risk factors for carriage at the first visit

187 The relationship between age and carriage was non-linear (Figure 2A, Supplemental table 1). Overall Nm

188 carriage increased rapidly in early adolescence, from 15/318 (4.7%) in subjects 13-14 years of age to

189 113/501 (22.6%) in 17-18 year old subjects (OR 5.88, 95%CI: 3.66-10.29, P<0.001). MenB carriage

190 prevalence was highest in the 21-23 year age cohort; 5.3%-7.9% depending on the diagnostic test

191 (Figure 2B). Smoking, level of education, the average estimated number of times per week going to

192 crowded social venues, kissing in the last week and drinking alcohol were identified as confounders of the

193 association between age and meningococcal carriage. The significant association between age group

194 and meningococcal carriage was lost when corrected for these confounders. After correcting for all other

195 variables in supplementary table 1, education level, the average estimated number of times per week

196 going out, kissing in the past week, drinking alcohol and smoking 10-20 cigarettes per day showed were

197 statistically independent predictors of carriage. Model stability, using backward stepwise model selection,

198 resulted in the predictors of carriage.

200 Duration of carrier state and acquisition

201 Of all subjects, 300/1715 (17%) subjects in the last year of secondary school at baseline were included in

202 the longitudinal cohort (Figure 1). Median age at inclusion was 17.5 years (±SD 0.9) and 163/300 (54%)

203 were female. Of the subjects in the longitudinal cohort, 67/300 (22%) carriers were identified at enrolment

204 by culture, and 72/289 (25%) and 80/276 (29%) after 3 months and 8 months, respectively. Of all

205 participants who completed the whole study period, 158/276 (57%) subjects did not carry a

206 meningococcal isolate by culture at any of the three sampling visits (Supplementary Figure 2). A

207 meningococcal isolate at all three sampling visits was identified in 21/276 (5%) subjects. The same

208 genogroup was identified by WGS at all three sampling visits in 12/21 (57%) subjects. MenB (5/12; 42%)

209 and MenW (3/12; 25%) were most frequently identified in the cases of persistent carriage.

210 The acquisition rate of colonization by meningococci irrespective of the serogroup was 45.4 per

211 1000 person-months during the whole study period. That of MenB was 9.4 per 1000 person-months.

213 Detailed analysis of carriage isolates

214 The WGS of 419 meningococcal carriage isolates collected during all three visits were compared to

215 evaluate genetic diversity of isolates between individuals and isolate persistence within an individual

216 (Supplementary Figure 3). Isolates from the same genogroup and clonal complex (CC) generally tended

217 to cluster together but a large diversity of isolates that were not clustered by region or age group were

218 observed. However, in several instances we identified isolates within a cluster that had a different

219 genogroup compared to the neighboring isolates in the cluster (Supplementary Figure 3). In addition,

220 isolates without a capsule locus (NG) located in a serogroup-restricted cluster have been noted indicating

221 loss of the capsule locus by deletion.

222 There were 5/276 (2%) cases of persistent carriage of MenB in the longitudinal cohort (Figure

223 3A). Of these subjects, 2/5 (40%) acquired a replacement strain during a follow-up visit or were colonized

224 by a mixture of strains (subjects D and E). The isolates from the other three cases of persistent MenB

225 carriage were, in each case, nearest neighbors on the WGS phylogenetic tree (subjects A, B and C;

226 Figure 3A) and identical with respect to genogroup, CC and other fine typing markers. Examination of the

227 WGS data revealed a progressive accumulation of single nucleotide polymorphisms (SNPs) and/or

228 recombination events across the genome over the 8 months surveillance period from each of these

229 subjects (Figure 3B).

230 DISCUSSION

231 Meningococcal disease originates from the oropharyngeal niche through contact with a carrier. In the

232 present study to evaluate Nm carriage and acquisition we found that WGS of cultured isolates was able to

233 provide a precise picture of both genogroup and isolate identity. Specifically, we were able to closely track

234 isolates within the population and could assess that there was a large diversity of isolates that were not

235 clustered by region or age.

236 The overall meningococcal carriage prevalence of 16% observed in this study was comparable to

237 carriage rates reported for UK populations in this age cohort [4, 13]. MenB was the most common

238 serogroup identified by WGS with 72/267 (27%) cultured isolates followed by MenX (38/267; 14%) and

239 MenY (34/267; 13%). MenB is also the most common isolate found in disease isolates [9]. MenY disease

240 has traditionally been rare in Europe, however the incidence has been rising in the Netherlands and at the

241 time of the study it was the second most common serogroup causing disease in adolescents and young

242 adults [26]. We identified 6 isolates with the MenC loci, but expression of capsule was not detected. The

243 low prevalence may reflect high uptake and use of MenC conjugate vaccines in the Netherlands since

244 2002 [7]. However, low rates of MenC carriage were also detected in the UK prior to vaccination when

245 MenC disease rates were as high as 0.2 per 10,000 population suggesting that MenC carriage might be

246 transient and therefore difficult to detect [27, 28].

247 It has been demonstrated that when conjugate vaccines targeting MenC and MenA were

248 implemented on a large scale, the transmission of these meningococci was prevented leading to herd

249 protection [5, 16]. To monitor the impact on carriage after vaccine implementation, accurate diagnostic

250 approaches are required. We found that WGS was the most sensitive method for serogroup identification

251 compared to Ouchterlony and rt-PCR. The discrepancies between the two techniques can be explained

252 due to the fact that rt-PCR interrogates a small portion of one gene essential for capsule synthesis;

253 whereas, WGS provides a view of the entire cps operon. Direct rt-PCR has potential logistical be

254 advantages over traditional culture. However, direct-PCR was less sensitive to identify MenB compared to

255 WGS likely due to the lack of culture enrichment [29].

256 The impact of protein based vaccines, such as the MenB vaccines, on meningococcal carriage

257 and transmission needs to be established. Therefore, it is important to determine meningococcal carriage

258 rates by age, specifically in which age group carriage is initially detected, so that subjects can be

259 vaccinated prior to the period of maximum risk for acquisition. In a small study with 4CMenB of young

260 adults in the UK, impact on MenB carriage was not observed [12]. In the current study, we demonstrate a

261 notable increase in carriage between the age 13-14 and 15-16 years. For future studies evaluating the

262 impact on carriage acquisition by vaccination, the age group of 13-14 years would be the preferred target

263 group. The increase in carriage in our study was strongly linked to life style rather than age. Also

264 observed in a UK study, it was found that social behavior, rather than age or sex, was associated with

265 higher frequency of meningococcal carriage among teenagers [14]. Since life style changes over time and

266 between countries, information regarding the risk factors for carriage can aid in Public Health decisions

267 about target ages for vaccination campaigns. In addition, it is important to understand carriage dynamics

268 in the specific trial population before a vaccine interventional study is initiated.

269 In this evaluation, we demonstrated that WGS of cultured isolates provided data sufficient for

270 genogroup assignment and isolate characterization. Clustering of the isolates by their genetic markers

271 (e.g. CC) is useful to identify where capsule switching may have occurred within the study population.

272 Capsule switching has relevance for disease outbreak surveillance as the virulent CC11 strain has been

273 involved in capsule switching from capsule group C to group W through recombination of the entire cps

274 cluster [30, 31]. In this study, we have identified several isolates that appear to have capsule switching or

275 capsule deletion events compared to closely related strains within the cluster, highlighting the genetic

276 plasticity of the meningococcal bacteria and the ability of WGS to identify these types of genetic changes

277 within a population. Longitudinal carriage studies are important for understanding the dynamics of

278 transmission. Acquisition rate and duration of carriage are important variables for models to predict the

279 epidemiological and economic impact of vaccination. One strength of our longitudinal study cohort is the

280 high-attendance for sampling visits achieved in this particular age cohort. Our longitudinal cohort showed

281 that carriage can persist for at least 8 months. A previous longitudinal carriage study observed a high

282 turnover rate of meningococcal carriage among UK students and indicated that carriage of a particular

283 meningococcal strain may not necessarily protect against colonization by homologous or heterologous

284 strains [32]. WGS was also useful for describing carriage dynamics. We found long-term carriage of

285 MenB and were able to identify whether the same strain was present at multiple visits or a new strain was

286 acquired during the follow-up. We found 5 cases of persistent carriage of which three cases most likely

287 carried the same strain retrieved in three consecutive visits over an 8 month period. Most interestingly,

288 the isolates showed progressively acquired mutations or recombinations across the genome, indicating its

289 potential relevance in long-term adaptation to the host. Nm has long been recognized to have highly

290 dynamic population structure due to frequent homologous recombination and horizontal gene transfer

291 [33]. Yet very little is known about within-host evolution of Nm. Accumulation of mutations in the genome

292 from this carriage isolates may potentially increase virulence. We believe that the present methodology

293 allowed for the first time to investigate meningococcal carriage with such high accuracy and this is highly

294 relevant for better understanding of the dynamics of carriage and transmission within populations.

295 A limitation of the current study is that the vaccination history of the subjects was not known.

296 However, a single MenC-TT vaccination in 2002 can be assumed in all subjects of the current study due

297 to the high vaccine coverage of 94% during the catch-up campaign in 2002 [7].

298 In conclusion, data from the current study show that WGS of cultured isolates is recommended to

299 monitor baseline Nm carriage and the impact on carriage after vaccine implementation. Results of this

300 study can be used as a pre-implementation reference, showing that MenB, MenX and MenY were the

301 most common serogroups identified by WGS of cultured isolates. Based upon results of this study and

302 risk of IMD, an adolescent meningococcal vaccination might include children before the age of 15 since

303 this is the age where increased carriage acquisition occurs related to behavioral changes. This study

304 provides useful information for conducting a clinical trial to look at the effect of meningococcal vaccine

305 (particularly MenB vaccine) on carriage.

306 ACKNOWLEDGEMENTS

307 We thank all the participants in the study. Additionally, we extend our gratitude to the clinical team of the

308 Spaarne Hospital, in particular Jacqueline Zonneveld and Greetje van Asselt. The authors would also like

309 to acknowledge the technical expertise and assistance of Agaath Arends, Ilse Schuurman, Wendy

310 Keijzers and Moniek Feller of the NRLBM. Furthermore, we would like to thank Sabine Wellnitz and Paul

311 Liberator of Pfizer for careful review of the manuscript. This publication made use of the Neisseria Multi

312 Locus Sequence Typing website (http://pubmlst.org/neisseria) developed by Keith Jolley and located at

313 the University of Oxford [21]. The development of this database has been funded by the Wellcome trust

314 and European Union. Phenotyping of cultured isolates was performed at the NRLBM, Amsterdam, The

315 Netherlands; rt-PCR was performed at PPD Vaccine and Biologics Lab, Wayne, PA, USA; WGS was

316 performed at Pfizer Vaccine Research & Development.

318 FUNDING

319 This work was financially supported by Pfizer Vaccine Research and Development.

321 CONFLICT OF INTEREST

322 MBvR, MWB, VMDS and MAvH report no conflict of interest. AvdE declares to have received grants for

323 pneumococcal epidemiology studies support from Pfizer. EAMS declares to have received unrestricted

324 research support from Pfizer, grant support for vaccine studies from Pfizer and GSK and fees paid to the

325 institution for advisory boards or participation in independent data monitoring committees for Pfizer and

326 GSK. ASA, JE, LH, KUJ, HJ, NK and LP were all employees of Pfizer Vaccine Research and

327 Development at the time of the study implementation, conduct, and analysis and may hold stock in the

328 company.

330 NOTE

331 Information of this manuscript was presented during the International Pathogenic Neisseria Conference

332 2016, Manchester, UK.

333 REFERENCES

334 1 Harrison LH, Trotter CL, Ramsay ME. Global epidemiology of meningococcal disease. Vaccine.

335 2009; 27 Suppl 2: B51-63.

336 2 Bijlsma MW, Bekker V, Brouwer MC, Spanjaard L, van de Beek D, van der Ende A. Epidemiology

337 of invasive meningococcal disease in the netherlands, 1960-2012: An analysis of national

338 surveillance data. The Lancet Infectious diseases. 2014; 14: 805-812.

339 3 Greenwood B. Manson lecture. Meningococcal meningitis in africa. Transactions of the Royal

340 Society of Tropical Medicine and Hygiene. 1999; 93: 341-353.

341 4 Christensen H, May M, Bowen L, Hickman M, Trotter CL. Meningococcal carriage by age: A

342 systematic review and meta-analysis. The Lancet Infectious diseases. 2010; 10: 853-861.

343 5 Maiden MC, Ibarz-Pavon AB, Urwin R, et al. Impact of meningococcal serogroup c conjugate

344 vaccines on carriage and herd immunity. The Journal of infectious diseases. 2008; 197: 737-743.

345 6 de Greeff SC, de Melker HE, Spanjaard L, Schouls LM, van Derende A. Protection from routine

346 vaccination at the age of 14 months with meningococcal serogroup c conjugate vaccine in the

347 netherlands. The Pediatric infectious disease journal. 2006; 25: 79-80.

348 7 Kaaijk P, van der Ende A, Berbers G, van den Dobbelsteen GP, Rots NY. Is a single dose of

349 meningococcal serogroup c conjugate vaccine sufficient for protection? Experience from the

350 netherlands. BMC infectious diseases. 2012; 12: 35.

351 8 European centre for disease prevention and control. Surveillance of invasive bacterial diseases

352 in europe. Available at: Http://ecdc.Europa.Eu/en/publications/publications/invasive-bacterial-

353 diseases-surveillance-2011.Pdf. Accessed april 13, 2015.

354 9 Safadi MA, Bettinger JA, Maturana GM, Enwere G, Borrow R, Global Meningococcal I. Evolving

355 meningococcal immunization strategies. Expert review of vaccines. 2015; 14: 505-517.

356 10 Fletcher LD, Bernfield L, Barniak V, et al. Vaccine potential of the neisseria meningitidis 2086

357 lipoprotein. Infection and immunity. 2004; 72: 2088-2100.

358 11 Esposito S, Principi N. Vaccine profile of 4cmenb: A four-component neisseria meningitidis

359 serogroup b vaccine. Expert review of vaccines. 2014; 13: 193-202.

360 12 Read RC, Baxter D, Chadwick DR, et al. Effect of a quadrivalent meningococcal acwy

361 glycoconjugate or a serogroup b meningococcal vaccine on meningococcal carriage: An

362 observer-blind, phase 3 randomised clinical trial. Lancet. 2014; 384: 2123-2131.

363 13 Soriano-Gabarro M, Wolter J, Hogea C, Vyse A. Carriage of neisseria meningitidis in europe: A

364 review of studies undertaken in the region. Expert review of anti-infective therapy. 2011; 9: 761365 774.

366 14 MacLennan J, Kafatos G, Neal K, et al. Social behavior and meningococcal carriage in british

367 teenagers. Emerging infectious diseases. 2006; 12: 950-957.

368 15 Neal KR, Nguyen-Van-Tam JS, Jeffrey N, et al. Changing carriage rate of neisseria meningitidis

369 among university students during the first week of term: Cross sectional study. Bmj. 2000; 320:

370 846-849.

371 16 Kristiansen PA, Diomande F, Ba AK, et al. Impact of the serogroup a meningococcal conjugate

372 vaccine, menafrivac, on carriage and herd immunity. Clinical infectious diseases : an official

373 publication of the Infectious Diseases Society of America. 2013; 56: 354-363.

374 17 Jeppesen CA, Snape MD, Robinson H, et al. Meningococcal carriage in adolescents in the united

375 kingdom to inform timing of an adolescent vaccination strategy. The Journal of infection. 2015;

376 71: 43-52.

377 18 Slaterus KW. Serological typing of meningococci by means of micro-precipitation. Antonie van

378 Leeuwenhoek. 1961; 27: 305-315.

379 19 Rojas E, Hoyos J, Oldfield NJ, et al. Optimization of molecular approaches to genogroup neisseria

380 meningitidis carriage isolates and implications for monitoring the impact of new serogroup b

381 vaccines. PloS one. 2015; 10: e0132140.

382 20 Jones CH, Mohamed N, Rojas E, et al. Comparison of phenotypic and genotypic approaches to

383 capsule typing neisseria meningitidis using invasive and carriage isolate collections. Journal of

384 clinical microbiology. 2015.

385 21 Jolley KA, Maiden MC. Bigsdb: Scalable analysis of bacterial genome variation at the population

386 level. BMC bioinformatics. 2010; 11: 595.

387 22 Treangen TJ, Ondov BD, Koren S, Phillippy AM. The harvest suite for rapid core-genome

388 alignment and visualization of thousands of intraspecific microbial genomes. Genome biology.

389 2014; 15: 524.

390 23 Letunic I, Bork P. Interactive tree of life (itol) v3: An online tool for the display and annotation of

391 phylogenetic and other trees. Nucleic acids research. 2016; 44: W242-245.

392 24 Pearson CJCES. The use of confidence or fiducial limits illustrated in the case of the binomial.

393 Biometrika. 1934; 26: 404-413.

394 25 Caugant DA, Hoiby EA, Magnus P, et al. Asymptomatic carriage of neisseria meningitidis in a

395 randomly sampled population. Journal of clinical microbiology. 1994; 32: 323-330.

396 26 Netherlands reference laboratory for bacterial meningitis. Bacterial meningitis in the

397 netherlands. Annual reports 2007-2011. Amsterdam: University of amsterdam, the netherlands.

398 27 Maiden MC, Stuart JM. Carriage of serogroup c meningococci 1 year after meningococcal c

399 conjugate polysaccharide vaccination. Lancet. 2002; 359: 1829-1831.

400 28 Miller E, Salisbury D, Ramsay M. Planning, registration, and implementation of an immunisation

401 campaign against meningococcal serogroup c disease in the uk: A success story. Vaccine. 2001;

402 20 Suppl 1: S58-67.

403 29 Trzcinski K, Bogaert D, Wyllie A, et al. Superiority of trans-oral over trans-nasal sampling in

404 detecting streptococcus pneumoniae colonization in adults. PloS one. 2013; 8: e60520.

405 30 Beddek AJ, Li MS, Kroll JS, Jordan TW, Martin DR. Evidence for capsule switching between

406 carried and disease-causing neisseria meningitidis strains. Infection and immunity. 2009; 77:

407 2989-2994.

408 31 Swartley JS, Marfin AA, Edupuganti S, et al. Capsule switching of neisseria meningitidis.

409 Proceedings of the National Academy of Sciences of the United States of America. 1997; 94: 271410 276.

411 32 Ala'aldeen DA, Oldfield NJ, Bidmos FA, et al. Carriage of meningococci by university students,

412 united kingdom. Emerging infectious diseases. 2011; 17: 1762-1763.

413 33 Joseph B, Schwarz RF, Linke B, et al. Virulence evolution of the human pathogen neisseria

414 meningitidis by recombination in the core and accessory genome. PloS one. 2011; 6: e18441.

416 Figure legends

417 Figure 1. Flow-chart for inclusion and follow-up. Carriers were identified by culture of oropharyngeal

418 swabs.

419 Note. Oropharyngeal swabs and questionnaires of participants in the longitudinal study cohort were

420 collected at enrolment and 3 months and 8 months after enrolment.

422 Figure 2. Meningococcal carriage prevalence by age cohort at the first visit identified by culture of

423 oropharyngeal swabs.

424 A. Overall meningococcal carriage identified by culture of oropharyngeal swabs from all 1715 participants

425 at the first visit. B. Meningococcal serogroup B carriage prevalence at the first visit identified by each

426 detection method for a subset of participants of whom a second swab was collected; Ouchterlony, rt-PCR

427 and WGS of cultured isolates and direct rt-PCR of the second swab.

428 Note. Error bars indicate 95% confidence intervals.

430 Figure 3. Phylogeny of carriage isolates based on comparative genomic analysis.

431 A. Meningococcal serogroup B carriage (N=105). B. Distribution of SNPs in MenB isolates from three

432 subjects showing persistence of carriage across three longitudinal visits.

433 The visits (ring 1) and clonal complexes (CCs) (ring 2) were annotated in color strips along the circular

434 tree. In ring 2, CCs were labeled accordingly, whereas white strip represents NT. Tree branches were

435 colored to match CCs (inferred from the majority isolates of the cluster). The isolate IDs of three

436 persistent cases (carriage isolates collected from 3 longitudinal visits) were highlighted in different colors.

437 Persistent carriage in three different subjects is accompanied by a progressive accumulation of SNPs

438 and/or recombination events. The WGS of isolates from visit 1 for each individual was used as a baseline

439 or reference data set. Vertical light grey lines for each of the visit 1 WGS data sets represent sequences

440 that are not part of the N. meningitidis core genome and are shared in WGS of carriage isolates from

441 visits 2/3. In each case, WGS of isolates collected at visits two and three revealed an accumulation of

442 SNPs and/or recombination events.

Table 1. Serogroup distribution of isolates detected by WGS and Ouchterlony of cultured isolates at the first visit (N=270).

Isolate WGS, N

Ouchterlony, N B C E W X Y Z Capsule null Negative ND Total

B 37 0 0 0 0 0 0 0 0 0 37

C 0 0 0 0 0 0 0 0 0 0 0

E 0 0 16 0 0 0 0 1 0 0 17

W 0 0 0 8 0 1 0 0 0 0 9

X 0 0 0 0 3 0 0 0 0 0 3

Y 0 0 0 0 0 20 0 0 0 0 20

Z 0 0 0 0 0 0 2 0 0 0 2

NG 34 5 3 4 35 13 2 80 1 3 180

ND 1 0 0 0 0 0 0 1 0 0 2

Total 72 5 19 12 38 34 4 82 1 3 270

Abbreviations: N= number of isolates, WGS = Whole Genome Sequencing, NG = Non-groupable, ND = Not Done, Negative = WGS negative for porA and ctrA. Numbers in bold were concordant between the two assays.

Table 2. Serogroup distribution of isolates according to isolate rt-PCR and WGS of cultured isolates at the first visit (N=177).

Isolate rt-PCRa Isolate WGS

B C E W X Y Z Capsule null ND Total

B 46 0 0 0 0 0 0 2 0 48

C 0 2 0 0 0 0 0 1 0 3

E 0 0 10 0 0 0 0 6 0 16

W 0 0 0 10 1 0 0 2 0 13

X 0 0 0 0 10 0 0 7 2 19

Y 0 0 0 0 0 7 ^ 0 2 1 10

Z 0 0 0 0 0 0 4 2 0 6

NGb 0 0 0 0 0 0 0 25 0 25

Unassignedc 5 1 2 1 14 9 0 5 0 37

Total 51 3 12 11 25 16 4 52 3 177

Abbreviations: N= number of isolates, WGS = Whole Genome Sequencing, NG = Non-groupable, ND = Not Done. Numbers in bold were concordant between the two assays.

a rt-PCR was only performed on cultured isolates from a subset of participants of whom a second oropharyngeal swab was collected. b Non-groupable was defined as positive porA and/or ctrA, but negative for group-specific assays. c rt-PCR positive for multiple group-specific assays.

Figure 1.

Figure 2.

A. Any meningococcus 0.4-i

O 0.2-|

0.0-1-

B. MenB

.1 02H

£ 0.1-1

l l Ouchterlony

1771 Isolate rt-PCR

^ Isolate WGS

^ Direct rt-PCR

M M à

Age group (years)

Age group (years)

Figure 3

Subject C-[

Subject E

^Im CC269

Subject D

Subject D

Subject A

PMB4667 (v1) PMB4939 (v2) PMB6011 (v3)

PMB4677 (v1) Subject B [PMB4942 (v2)

PMB6005 (v3)

. 1 1 ■11 III" 11

Subject C

PMB4681 (v1) PMB5905 (v3) PMB4966 (v2)