INVITED ARTICLE Stanley Plotkin, Section Editor

Fast Tracking the Licensure Process to Control an Epidemic of Serogroup B Meningococcal Disease in New Zealand

Diana Lennon,1 Catherine Jackson,1,a Sharon Wong,1,a Maraekura Horsfall,1,a Joanna Stewart,2 and Stewart Reid3,4

1 2 3 4 Community Paediatrics, and Biostatistics, and Immunisation Advisory Center, The University of Auckland, and Ropata Village Medical Centre, Downloaded from Lower Hutt, New Zealand

Epidemics of serogroup B meningococcal disease are rare. Strain-specific outer membrane vesicle vaccines, which are not

marketed, are the only current tool for control. A correlate of protection is ill defined, but published data suggest that cid.oxfordjournals.org measured serum bactericidal antibody levels parallel efficacy. Even infants can mount a strain-specific antibody response to a strain-specific vaccine. New Zealand’s epidemic (1991–2007; peak rate [in 2001], 17.4 cases per 100,000 persons) was dominated by a single strain. After a 5-year search (1996–2001) for a manufacturer for a strain-specific outer membrane vesicle vaccine, a fast-tracked research program (2002–2004) determined the safety and immunogenicity of vaccine in infants (2 age groups: 6–10 weeks and 6–8 months), children (age, 16–24 months), and school-aged children (age, 8–12 years) after

an adult trial. The vaccine was reactogenic, compared with control vaccines (meningococcal C conjugate and routine infant at World Health Organization on March 14, 2011 vaccines), but retention was high. Three vaccine doses produced antibody levels (measured by serum bactericidal assay) that were considered to be adequate for public health intervention. However, in young infants, a fourth dose was required to achieve levels equivalent to those achieved by other age groups. Provisional licensure by New Zealand’s MedSafe was based on serological criteria strengthened by bridged safety data from studies of the parent outer membrane vesicle vaccine, independent assessment of manufacturing quality, and a clear plan for safety monitoring and effectiveness evaluation after licensure.

To facilitate the control of an epidemic of single-strain dom- United States and other developed countries and accounts for inated serogroup B meningococcal disease in New Zealand approximately one-half of invasive meningococcal disease in (NZ), licensure of a tailor-made strain-specific outer membrane infants in the United States [1]. Epidemics of serogroup B vesicle (OMV) vaccine was based on serologic criteria alone disease have occurred in many parts of the world [4–7]. with bridged safety data from the parent vaccine without direct Because outbreaks of serogroup B meningococcal disease evidence of efficacy. This is a collation of published data on tend to be clonal, the development of OMV vaccines that target this process. the outbreak strain has been possible. These vaccines are not is a leading cause of bacterial men- commercially available. ingitis and other invasive bacterial worldwide [1–3]. Beginning in 1991 (figure 1), NZ experienced an outbreak Endemic serogroup B meningococcal disease accounts for a of serogroup B disease; 86% of cases (during 1990–2003) were substantial proportion of cases of meningococcal disease in the caused by the B:4:P1.7–2,4 strain [8–10]. Expression of the P1.7–2,4 PorA gene served as a marker of the ST-41/44 clonal complex [10]. Surveillance of meningococcal disease in NZ Received 19 December 2008; accepted 12 April 2009; electronically published 15 July 2009. a Present affiliation: Waitemata District Health Board, Auckland, New Zealand. entails a combination of notification and laboratory data rec- Reprints or correspondence: Dr. Diana Lennon, Community Paediatrics, School of Popula- onciled at the national level [11]. The relative stability of the tion Health, University of Auckland, Private Bag 92019, Auckland, New Zealand (d.lennon@ auckland.ac.nz). epidemic strain PorA protein supported the use of a strain- Clinical Infectious Diseases 2009;49:597–605 specific OMV vaccine for epidemic control [12]. 2009 by the Infectious Diseases Society of America. All rights reserved. The epidemic peaked in 2001 (rate of notified disease, 17.4 1058-4838/2009/4904-0017$15.00 DOI: 10.1086/603552 cases per 100,000 persons per year; rate of disease due to the

VACCINES • CID 2009:49 (15 August) • 597 Downloaded from Figure 1. Number of cases of meningococcal disease, by serotype and year, in New Zealand during 1990–2008. Notified meningococcal cases includes laboratory-confirmed and probable cases (isolates were tested by polymerase chain reaction starting in 1996). Vaccine coverage (3 doses received) among persons aged 1–19 years was ∼12% in December 2004, ∼77% in December 2005, and ∼83% in December 2006. Serogroups W135, Y, and X are not shown (mean incidence rate during 1998–2008, 8 cases per year). cid.oxfordjournals.org epidemic strain, 10 cases per 100,000 persons per year). Ap- against noncapsular surface antigens [28, 29]. Evidence of in- proximately 80% of cases occurred in persons aged !20 years, volvement of outer membrane proteins (especially PorA) in and ∼50% of cases occurred in persons aged !5 years. The immunity is accumulating [30–32]. However, unlike serogroup rates of disease were higher among indigenous Maori and chil- C meningococcal disease, no clear correlate of protection is dren of Pacific Island descent (both groups were more likely known [33–35]. Studies evaluating age-specific serum bacte- to be socio-economically deprived) than among NZ children ricidal activity in parallel with field efficacy estimates in Brazil at World Health Organization on March 14, 2011 ! of European origin, especially among infants aged 1 year (rates and Chile support the use of serum bactericidal activity as a in 2001: Maori, 429 cases per 100,000 persons per year; Pacific correlate of protection [16, 30, 36]. A landmark Chilean study Island descent, 640 cases per 100,000 persons per year; Euro- involving infants and children that was performed in 1996 dem- pean, 57 cases per 100,000 persons per year [11, 13]. The mean onstrated excellent activity against the strain against which the case fatality rate (during 1991–2003) was 4.1%. infants and children had been vaccinated but had no activity A population-based carriage study was conducted among against other strains [37]. This led to NZ’s decision to use a teenagers in a high-risk area before vaccine administration (in “designer” vaccine to control an epidemic that occurred largely 2004) and after the epidemic had begun to wane. One hundred among young children and that was dominated by a single twenty-three (11%) of 1115 teenagers were carriers of any me- ningococci; 10 (0.9%) of these teenagers carried strain P1.7– strain. The search for a vaccine manufacturer took 6 years 2,4 [14]. There was no carriage study conducted after vaccine (1996–2001). A public-private partnership (2002–2006) was uptake. Studies in Norway and Chile were unable to find an forged between Chiron Vaccines (now Novartis) and the NZ effect of serogroup B outer membrane protein vaccines on Ministry of Health, with technology transfer from the Nor- nasopharyngeal carriage of N. meningitidis [15, 16]. The low wegian Institute of Public Health (NIPH). and unsustained antibody levels with OMV-outer membrane Evaluation of the efficacy of OMV vaccines has been dogged protein vaccines may be less likely to have an effect on naso- with inconsistencies. Efficacy was established in older children pharyngeal carriage, leading to a postulated absence of a herd in 2 randomized controlled trials [5, 38]. Rapid decrease of immunity effect (figures 2–4). vaccine-induced antibody responses in the Norwegian 2-dose trial led to efficacy results of 57% after 29 months [5]. A later VACCINES estimated efficacy of 87% at 10 months suggested potential To date, vaccines evaluated in detail for serogroup B menin- short-term usefulness of this vaccine [39]. To ameliorate the gococcal disease have been outer membrane protein vaccines rapid decrease in antibody levels, a third or fourth dose was [25, 26]. Protection against meningococcal disease has been studied in additional trials [37, 39–42]. The proven efficacy of demonstrated to correlate with bactericidal antibodies [27], but OMV vaccines in children aged !5 years was unclear [6, 7, 36, in serogroup B infections, bactericidal antibodies are directed 38, 43, 44].

598 • CID 2009:49 (15 August) • VACCINES Table 1. Seroresponders and the Percentage of Participants with a Titer у1:4 to the New Zealand Candidate Vaccine Strain NZ98/ 254 (B:4:P1.7b,4;a Intention-to-Treat Analyses) after Administration of New Zealand Outer Membrane Vesicle (OMV) Vaccine (MeNZB).

Age Variable 6–10 weeks 6–8 months 16–24 months 8–12 years (cohort A) 8–12 years (cohort B) No. (%) of seroresponders [95% CI]b After dose 3c 239 (53) [46–59] 163 (74) [68–80] 254 (75) [69–80] 220 (74) [67–79] 230 (79) [73–84] After dose 4d 45 (69) [54–80]d … 42 (100) [90–100]e …… No. (%) of participants with a titer у1:4 [95% CI] After dose 3c 239 (76) [70–81] 202 (92) [88–95] 254 (92.2) [88–95] 224 (92) [88–95] 230 (95) [91–97] After dose 4c 45 (82) [68–91] … 42 (100) [90–100]e ……

NOTE. Data are from [17, 18, 20–22]. CI, confidence interval. a Current notation: P1.7–2,4. b Seroresponders were defined as participants who had a у4-fold increase in serum bactericidal antibody titer, compared with prevaccination baseline interpolated titers. A baseline titer of !1:4 was required to reach a titer of у1:8 (seroresponse). c

Four to 6 weeks after administration of a vaccine dose. Downloaded from d The fourth dose was given 5 months after dose 3. e For the infants aged 16–24 months who seroresponded to a 3-dose priming series, the fourth dose was given 17 months after dose 3. For 30 children aged 16–24 months who did not serorespond to a 3-dose priming series, the fourth dose was given 11 months after dose 3 and resulted in seroresponse in 100% (86%–100%) and in a titer у1:4 in 100% (86%–100%).

NZ’S APPROACH 38], including the Norwegian parent vaccine (strain: B:15: cid.oxfordjournals.org P1.7,16), an OMV vaccine previously developed by NIPH; stan- In the face of a well-characterized, ongoing epidemic domi- dardization of the serum bactericidal assay across 3 laboratories nated by a single strain of meningococcus, a rapid clinical de- (led by NZ) [45]; у3 vaccine doses shown to produce antibody velopment plan (figure 5) was evolved to estimate immuno- levels considered to be adequate for public health intervention genicity, safety, and . After production of a vaccine age groups, as measured by serum bactericidal assay; and ran- based on the NZ epidemic strain by NIPH and enrollment of domized controlled trial safety data from other OMV vaccines, at World Health Organization on March 14, 2011 health care adult volunteers in a phase I/II trial, trials involving including the NIPH parent vaccine (171,800 students aged 14– children aged 8–12 years, 16–24 months, 6–8 months, and 6– 16 years) [16, 38, 46]. A study from Chile used purified outer 10 weeks followed in quick succession. Trials proceeded in membrane protein strain-specific vaccine in a randomized con- school children as soon as the NZ strain vaccine was known trolled trial design involving 40,800 participants [16]. The Cu- to be immunogenic and not to be associated with serious ad- ban randomized controlled trial involved 106,400 students aged verse events in adults (n p 49 ). Similarly, a small trial to con- firm safety in adults (n p 10 ) was performed using the Chiron- 10–14 years and used a strain-specific (B:4:P.15) OMV vaccine produced NZ strain vaccine after technology transfer from mixed with a C polysaccharide vaccine [38]. Other elements NIPH before the trial of this product in school children and essential for licensure were reactogenicity and immunogenicity was followed by a trial involving infants aged 6–10 weeks. Ap- data, by age group, available from other published trials [16, proximately 1500 participants were enrolled into clinical trials 37, 39, 46]; reactogenicity and immunogenicity data, by age over ∼24 months. Supported by the knowledge that measurable group, available from the NZ trials of the NZ OMV vaccine antibodies parallel efficacy and to facilitate rapid progress to [17–22] (the size of these trials was increased beyond the size licensure and epidemic control, a randomized efficacy trial was needed to demonstrate immunogenicity, to produce more reac- omitted. togenicity data); and manufacturing quality, as overviewed in- A number of elements were considered to be essential for dependently by the National Institute of Biological Standards licensure of the NZ meningococcal OMV vaccine (MeNZB) (United Kingdom), which allowed safety data from NIPH (par- without a phase III efficacy trial to support the licence appli- ent) vaccine studies to be bridged to support the NZ vaccine. cation. These elements included a well-characterized clonal out- NZ OMV vaccine (MeNZB) safety monitoring was pre- break [10], leading to the development of a strain-specific OMV planned postlicensure, as required for licensure [47]. Before the vaccine; the assumption that serum bactericidal antibody mea- MeNZB initiative, vaccine safety surveillance in NZ was solely surements are likely to reflect potential vaccine effectiveness, a standard passive reporting system conducted by the Centre with an increase in antibody titer (as opposed to absolute titer for Adverse Reaction Monitoring. There was no national im- cutoff) being a reasonable approach in the absence of a clear- munization register. The prelicensure safety database comprised cut correlate of protection [16, 30]; proven efficacy by ran- ∼3300 doses. A comprehensive phase IV safety monitoring sys- domized controlled trial of OMV vaccines in older children [5, tem ( Safety Strategy, Data Management

VACCINES • CID 2009:49 (15 August) • 599 Downloaded from cid.oxfordjournals.org at World Health Organization on March 14, 2011

Figure 2. Interpolated antibody titers to New Zealand (NZ) strain (NZ98/254) in participants vaccinated with 3 doses of NZ strain outer membrane vesicle (OMV) vaccine (MeNZB), as measured by serum bactericidal assay (intention-to-treat analyses) [17–22]. Boxplot was plotted on the log 2 scale. The box represents the interquartile range, the center line represents the median, the black dot represents the mean values, and circles represent the individual values beyond the whiskers, which extend to the last value within 1.5 times the interquartile range. Blood samples were obtained at 4–6 weeks after administration of a vaccine dose. The trials for persons aged 8–12 years were run in 2 cohorts (A and B) at different times. Cohort A received the Norwegian Institute of Public Health-produced NZ strain OMV vaccine at 0, 5, and 12 weeks, and cohort B and infants received the Chiron (now Novartis) NZ strain OMV vaccine at 0, 6, and 12 weeks. Among infants aged 6–10 weeks, after 4 doses of MeNZB, the geometric mean titer increased to 21.3 (95% confidence interval, 12.3–36.7).

Group 2004) was required. The key features of this system [48] the Canadian Impact system [50]. This was established by 4 were establishment of a National Register; use district health boards considered to be at highest risk of disease of several data sources comprising several elements (the existing and involved monitoring of hospitalizations for ∼200,000 vac- passive reporting system, an intensive vaccine monitoring pro- cinees and ∼600,000 vaccine doses. A list of adverse events gram in which all clinical events were reported to Centre for possibly attributable to meningococcal OMV vaccine admin- Adverse Reaction Monitoring for a 6-week period after vac- istration was established from experience with other OMV vac- cination [49], and hospital based monitoring [described be- cines and was expanded to include other vaccine-related adverse low]); separation of data collection and analysis from data as- events (e.g., neurological symptoms). A priori determined case sessment (an independent safety monitoring board [ISMB] was definitions and background disease rates were established. established by the NZ Health Research Council to assess all Nurse monitors screened admission lists daily for events of safety data); staggered progress of the campaign, interest and reported to a clinical review committee of pedi- which required that the ISMB assess predetermined safety da- atricians to aid interpretation. A cluster investigation protocol tasets to permit progressive extension of the vaccination cam- was available. paign from the highest-risk areas offered immunization to the Vaccine effectiveness evaluation was preplanned postlicen- rest of the country; and hospital-based monitoring, based on sure [44, 51–53]. In the absence of a population-based ran-

600 • CID 2009:49 (15 August) • VACCINES ningococcal vaccines, and the experience in the United King- dom with conjugate serogroup C meningococcal vaccines) [55].

IMMUNOLOGIC ASSESSMENT OF MENZB

Early discussion between the manufacturer and Medsafe in- dicated that the licensure of the NZ strain vaccine (MeNZB) would be considered on the basis of immunogenicity data and age group, without direct evidence of efficacy. Because the cor- relate of protection was uncertain, a threshold of antibody titer increase that was acceptable for vaccine use in epidemic control was agreed upon before commencement of the trials (see next paragraph). An outline of this strategy was peer reviewed in- ternationally and nationally before the trials commenced [56].

The primary immunogenicity outcome measure for all trials Downloaded from was an serum bactericidal antibody response against N. men- ingitidis serogroup B strain NZ98/254 (the vaccine strain). Sero- response was defined as a у4-fold increase serum bactericidal antibody titer from that measured before vaccination, with an

increase to at least 1:8 if the prevaccination titer was !1:4 [20]. cid.oxfordjournals.org Studies had 80% power for the lower 95% confidence limit of seroresponders being у40%, given a true value of at least 50% (the agreed minimum acceptable threshold). The studies also Figure 3. Persistence of immune response geometric mean titers, as measured by serum bactericidal assay, against the vaccine strain NZ98/ 254 in participants vaccinated with the New Zealand (NZ) outer membrane vesicle (OMV) vaccine (MeNZB) [18]. Top, Titers in persons aged 8–12 at World Health Organization on March 14, 2011 years (62 in cohort A [black] and 91 in cohort B [white]). Bottom, Titers in infants (gray) aged 6–8 months (n p 40 ). Squares indicate geometric mean titers, circles indicate median values, and whiskers indicate inter- quartile ranges. Cohort A received the Norwegian Institute of Public Health–produced NZ strain OMV vaccine at 0, 5, and 12 weeks, and cohort B and infants received Chiron (now Novartis) NZ strain OMV vaccine at 0, 6, and 12 weeks. domized controlled trial of vaccine efficacy of the MeNZB vaccine, application to the licensure authority included a postlicensure effectiveness evaluation. A suite of observational methods included disease surveillance, modeling of effective- ness, and a case-control study of vaccine effectiveness. The staged vaccine delivery throughout NZ allowed regression mod- eling techniques using current and historical data to evaluate the effectiveness of the program, reducing confounding by the natural epidemic path. A case-control study of vaccine effec- Figure 4. Geometric mean serum bactericidal antibody titers after ad- tiveness among children aged !5 years was planned, because it ministration of 3 doses of meningococcal (MCC; three 10-mg doses were given at 2, 3 and 4 months of age [23]), New Zealand could reduce confounding influences. Funds were withdrawn outer membrane vesicle (OMV) vaccine (MenZB; 3 doses were given at for this 2-year study after the first year. 6 weeks, 3 months, and 5 months of age; serum bactericidal antibody NZ’s legislation allowing “licensure with provisional con- titer against the vaccine strain NZ98/254 [22]), or recombinant menin- sent” [54] in special circumstances, such as an epidemic, was gococcal B vaccine with NZ98/254 OMV (rMenB + OMV; 3 doses given at 2, 4, and 6 months of age; serum bactericidal antibody titer against a key element. Earlier licensure decisions made in other juris- the vaccine strain NZ98/254 [24]) in infants. Error bars show the 95% dictions on the basis of immunologic data alone were noted confidence interval of the geometric mean titer and not the distribution (influenza vaccines annually, quadrivalent polysaccharide me- of serum bactericidal antibody titers.

VACCINES • CID 2009:49 (15 August) • 601 Downloaded from cid.oxfordjournals.org

Figure 5. Clinical development plan for the New Zealand meningococcal B outer membrane vesicle vaccine (MeNZB) trials, 2002–2004 [17–22]. Seroresponders (R) were defined as those who had a у4-fold increase in serum bactericidal antibody titer, compared with prevaccination baseline interpolated titers. A baseline titer of !1:4 was required to reach a titer of у1:8 (seroresponse). A, cohort A, for which the Norwegian Institute of Public Health–produced New Zealand strain outer membrane vesicle vaccine was used; B, cohort B, in which Chiron (now Novartis)–produced New Zealand strain outer membrane vesicle vaccine was used; NR, nonresponder. at World Health Organization on March 14, 2011 increased the reactogenicity database. The serum bactericidal adverse events. Despite frequent mild to moderate injection site assay used was validated by 4 individual laboratories (led by reactions, which lasted 2–3 days, tolerability was high, with NZ) [45, 57]. excellent trial retention. Severe reactogenicity was uncommon, Randomized, controlled, observer-blinded trials were con- and there were some self-limited systemic reactions. Follow-up ducted in Auckland with use of 3 doses of 25 mg of MeNZB studies of antibody level decrease revealed rapid decreases in administered intramuscularly 6 weeks apart; doses were added all age groups, especially the youngest (figure 3). Only 27.5% in extension studies (table 1, figure 2, and figure 3). Timing of of infants aged 6–8 months who were tested 7 months after administration for very young infants was in accordance with the 3-dose priming series had an serum bactericidal antibody the NZ infant immunization schedule (at 6 weeks, 3 months, titer у1:4. Therefore, most infants and children are likely to and 5 months of age). be unprotected not long after immunization (figure 4) [30, 31] In a phase I/II trial involving adults, there was a 4-fold in- if a titer of у1:4 is considered as a putative correlate of pro- crease in antibody titer (geometric mean titer [GMT], 49) in tection [31]. A vigorous booster response suggesting memory, all of the participants after the third dose of NIPH-produced although with modest GMTs, was seen both in children aged NZ strain vaccine [19]. Preassigned targets from MeNZB im- 16–24 months who had initially responded to 3 doses of vaccine munogenicity trials in childhood were met or surpassed in all (GMT, 259; 95% confidence interval [CI], 184–363) and in age groups (8–12 years, 16–24 months, 6–8 months, and 6–10 those who had not (GMT, 69; 95% CI, 46–106) [18]. weeks) [17, 20–22]. However, a fourth dose was required in 6–10-week-old infants to reach antibody levels achieved after LICENSURE AND EPIDEMIC CONTROL 3 doses in older children [22]. There was an age gradient in GMT, as was seen in previous OMV vaccine evaluations [16, Provisional licence was granted by the NZ licensing authority 30, 37]. Titers were modest, compared with those measured (Medsafe) for persons aged 6 months to 19 years in July 2004 after meningococcal C and A conjugate (figure 4) and for persons aged у6 weeks in February 2005. In January [23, 58, 59], but were comparable to those found in other 2006, the inclusion of a fourth dose in the infant priming series published meningococcal B OMV trials involving children [16]. was recommended on the basis of additional trial data. Overall, MeNZB was given to ∼1300 children, with no serious Vaccination commenced in July 2004, and by June 2006, 13

602 • CID 2009:49 (15 August) • VACCINES million doses had been administered to 1 million young persons 4-dose priming series to infants was poor (50%–60% reported aged !20 years as part of a catch-up program. Vaccine contin- by the NZ Ministry of Health in 2008). Despite this, the epi- ued to be offered to infants with the routine schedule until 1 demic continues to wane (in 2008, the rate of disease due to June 2008. In its final report, the ISMB stated that there was the epidemic strain was 1.1 cases per 100,000 population). “an outstanding program of sensitive and objective safety mon- MeNZB vaccine ceased to be offered in NZ after 1 June 2008. itoring” and that “it found no evidence of any significant ad- Active disease surveillance remains in place. A robust booster verse health event associated with the vaccine” (T. Nolan, C. response has been shown with MeNZB in children aged 16– Whitney, C. Grant, P. Reid, C. Frampton, unpublished data). 24 months [18], and the booster, along with reinstatement of NZ’s epidemic of meningococcal disease, which began in infant vaccination (vaccine is stock piled), could be used for 1991, began to wane in the last year of the trials and before epidemic control among high-risk age groups if there is a re- the vaccine program commenced (figure 1). Earlier docu- surgence of disease. New serogroup B vaccines consisting of mented epidemics of serogroup B lasted 10–15 years and then recombinant protein antigens with bactericidal activity that naturally ended [6, 16, 60]. The extent of vaccine efficacy, es- were chosen from the sequenced genome may give hope for

pecially in children aged !5 years, who carried 50% of the the future control of endemic meningococcal disease [24, 65]. Downloaded from disease burden, remains uncertain [44, 52, 53, 56, 61]. However, preliminary data (figure 4) suggest that such a vaccine On the basis of current available evidence on vaccine pre- combined with OMV from the NZ epidemic strain, after 3 and vention of meningococcal disease, persistence of antibody is 4 doses in infants, produced similar low GMTs against the NZ considered to be necessary to provide protection [62]. This has strain in vitro, compared with our studies of MeNZB vaccine been best established for serogroup C, although it is unlikely alone (3 or 4 doses). GMTs against 2 other serogroup B outer cid.oxfordjournals.org to be different for serogroup B [16, 30, 36]. Laboratory studies membrane proteins were more robust. Longer durations be- [63] demonstrated an important delay (∼5 days) before titer tween administration of doses in this infant age group could increase after administration of a booster to already primed have some benefit. infants. Meningococcal disease is known to ensue soon after CONCLUSIONS the organism is acquired in the nasopharynx; thus, 5 days may not be fast enough to prevent invasive disease. The observation OMV vaccines are the only currently available tool for control at World Health Organization on March 14, 2011 of waning immunity to meningococcal C disease in UK infants of outbreaks dominated by serogroup B. The NZ experience in a vaccinated population [64] lends support to the notion suggests a role for OMV vaccines as a “circuit breaker” in an that sufficient circulating antibodies are necessary for protec- epidemic of strain-specific serogroup B meningococcal disease, tion. This led to a dose schedule change for UK infants for with a more rapid than expected decrease in the number of meningococcal C vaccine. MeNZB vaccination was not ex- cases in the first year after mass vaccination (C. Kelly, R. Arnold, pected to provide long-term bactericidal antibody levels. Pub- Y. Galloway, J. O’Hallahan, unpublished data). Because serum lished literature shows a sharp decrease after receipt of 4 doses bactericidal antibody titers (which indicate protection) de- of vaccine, even in adults [40, 42]. creased rapidly, it would be expected that there has been min- In our studies (figure 3), among infants aged 6–8 months, imal vaccine effect for 12 years after the mass campaign. There only 27.5% had a serum bactericidal antibody titer у1:4 (i.e., has been no resurgence of disease, which allowed the safe dis- had antibody likely to protect at 7 months after the third dose continuation of MeNZB vaccine in June 2008. Careful ongoing of vaccine) [18]; among preteens in cohort A, 35.5% (95% CI, surveillance will be essential. Vaccine intervention may have 24.8%–48.0%) had a serum bactericidal antibody titer у1:4 at been more effective (and cost-effective) if it was started earlier 14 months after the third dose (figure 3). Antibody level de- during the epidemic cycle. crease after the fourth dose of the priming series in the youngest Acknowledgments age group in immunogenicity trials (age, 6–10 weeks) was not studied, although only 27.5% had a serum bactericidal antibody We thank all of those involved in the MeNZB vaccination program. у Financial support. New Zealand Ministry of Health and Chiron Vac- titer 1:4 when blood samples were obtained 4.5 months after cines (now Novartis Vaccines). 3 doses of MeNZB; the fourth dose produced only a modest Potential conflicts of interest. M.H. has received funding from the increase in GMT to 21.3 (95% CI, 12.3–36.7) (figure 3) [22]. Foundation UK. D.L. has received travel funds from the New Zealand Ministry of Health, Novartis Vaccines, and Wyeth and serves on Therefore, since the completion of the mass vaccination cam- the Vaccine Advisory Board for GlaxoSmithKline Vaccines (the Asia/Pacific paign in late 2005, most children have been unlikely to have region). C.J. has accepted travel funds from Merck. S.R. has received travel any antibody protection against the NZ meningococcal epi- funds from the New Zealand Ministry of Health, Novartis Vaccines, Wyeth, Merck, and GlaxoSmithKline and serves on the Vaccine Advisory Board demic strain as a result of the rapid decrease in antibody levels for GlaxoSmithKline Vaccines (the Asia/Pacific region). S.W. and J.S.: no after vaccination. In addition, administration of the scheduled conflicts.

VACCINES • CID 2009:49 (15 August) • 603 References 21. Hosking J, Rasanathan K, Mow FC, et al. Immunogenicity, reactogen- icity, and safety of a P1.7b,4 strain-specific serogroup B meningococcal 1. Rosenstein NE, Perkins BA, Stephens DS, et al. The changing epide- vaccine given to preteens. Clin Vaccine Immunol 2007; 14:1393–9. miology of meningococcal disease in the United States, 1992–1996. J 22. Wong S, Lennon D, Jackson C, et al. Immunogenicity and tolerability Infect Dis 1999; 180:1894–901. in infants of a New Zealand epidemic strain meningococcal B outer 2. Greenwood B. Manson Lecture: meningococcal meningitis in Africa. membrane vesicle vaccine. Pediatr Infect Dis J 2009; 28:385–90. Trans R Soc Trop Med Hyg 1999; 93:341–53. 23. Richmond P, Borrow R, Miller E, et al. Meningococcal serogroup C 3. Trotter CL, Ramsay ME. Vaccination against meningococcal disease in conjugate vaccine is immunogenic in infancy and primes for memory. Europe: review and recommendations for the use of conjugate vaccines. J Infect Dis 1999; 179:1569–72. FEMS Microbiol Rev 2007; 31:101–7. 24. Miller E, Borrow R, Findlow J, et al. Safety and immunogenicity of 4. Diermayer M, Hedberg K, Hoesly F, et al. Epidemic serogroup B me- Novartis meningococcal serogroup B vaccine (MenB Vaccine) after ningococcal disease in Oregon: the evolving epidemiology of the ET- three doses administered in infancy [abstract 133]. In: Program and 5 strain. JAMA 1999; 281:1493–7. abstracts of the European Society for Paediatric Infectious Diseases 5. Bjune G, Hoiby EA, Gronnesby JK, et al. Effect of outer membrane (Graz, Austria). 2008. vesicle vaccine against group B meningococcal disease in Norway. Lan- 25. Frasch C, van Alphen L, Holst J, Poolman JT, Rosenqvist E. Outer cet 1991; 338:1093–6. membrane protein vesicle vaccines for meningococcal disease. In: Pol- 6. Noronha CP, Struchiner CJ, Halloran ME. Assessment of the direct lard AJ, Maiden MC, eds. Meningococcal vaccines: methods and pro- effectiveness of BC meningococcal vaccine in Rio de Janeiro, Brazil: a tocols. New Jersey: Humana Press, 2001:81–107. Downloaded from case-control study. Int J Epidemiol 1995; 24:1050–7. 26. Frasch CE, Concepcion NF. Induction of group 17 specific antibodies 7. Padron FS, Huergo CC, Gill VC, Diaz EM, Valdespino IE, Gotera NG. by pneumococcal type 17F and 17A polysaccharide vaccines. Biologicals Cuban meningococcal BC vaccine: experiences and contributions from 2001; 29:11–6. 20 years of application. MEDICC Rev 2007; 9:16–22. 27. Gotschlich EC, Goldschneider I, Artenstein MS. Human immunity to 8. Martin DR, Walker SJ, Baker MG, Lennon D. New Zealand epidemic the meningococcus. IV. Immunogenicity of group A and group C me- of meningococcal disease identified by a strain with phenotype B:4: ningococcal polysaccharides in human volunteers. J Exp Med 1969;

P1.4. J Infect Dis 1998; 177:497–500. 129:1367–84. cid.oxfordjournals.org 9. Martin D, Lopez L, McDowell MC. The epidemiology of meningo- 28. Frasch CE, Chapman SS. Classification of Neisseria meningitidis group coccal disease in New Zealand in 2004. Wellington: Ministry of Health, B into distinct serotypes. J Infect Dis 1973; 127:149–54. 2005. 29. Jones DM, Eldridge J. Development of antibodies to meningococcal 10. Dyet KH, Martin DR. Clonal analysis of the serogroup B meningococci protein and lipopolysaccharide serotype antigens in healthy-carriers. J causing New Zealand’s epidemic. Epidemiol Infect 2006; 134:377–83. Med Microbiol 1979; 12:107–11. 11. Dyet K, Devoy A, McDowell R, Martin D. New Zealand’s epidemic of 30. Milagres LG, Ramos SR, Sacchi CT, et al. Immune response of Brazilian meningococcal disease described using molecular analysis: implications children to a Neisseria meningitidis serogroup B outer membrane pro- for vaccine delivery. Vaccine 2005; 23:2228–30. tein vaccine: comparison with efficacy. Infect Immun 1994; 62:4419–24. at World Health Organization on March 14, 2011 12. Devoy AF, Dyet KH, Martin DR. Stability of PorA during a menin- 31. Holst J, Feiring B, Fuglesang JE, et al. Serum bactericidal activity cor- gococcal disease epidemic. J Clin Microbiol 2005; 43:832–7. relates with the vaccine efficacy of outer membrane vesicle vaccines 13. Martin D, Lopez L, McDowell MC. The epidemiology of meningo- against Neisseria meningitidis serogroup B disease. Vaccine 2003;21: coccal disease in New Zealand in 2001. Wellington: Ministry of Health, 734–7. 2002. 32. Martin DR, Ruijne N, McCallum L, O’Hallahan J, Oster P. The VR2 14. Horsfall M, Lennon D, Martin D, Taylor S, Stewart J. Pharyngeal epitope on the PorA P1.7–2,4 protein is the major target for the im- carriage of meningococcal B in Auckland teenagers prior to mune response elicited by the strain-specific group B meningococcal introduction of a meningococcal B OMV vaccination programme [ab- vaccine MeNZB. Clin Vaccine Immunol 2006; 13:486–91. stract 273]. In: Program and abstracts of the 25th Annual Meeting of 33. Artenstein MS, Gold R, Zimmerly JG, Wyle FA, Schneider H, Harkins the European Society for Paediatric Infectious Diseases (Porto, Por- C. Prevention of meningococcal disease by group C polysaccharide tugal). 2007. vaccine. N Engl J Med 1970; 282:417–20. 15. Bjune G. “Herd immunity” and the meningococcal in 34. Trotter C, Borrow R, Andrews N, Miller E. Seroprevalence of menin- Norway. Lancet 1992; 340:315. gococcal serogroup C bactericidal antibody in England and Wales in 16. Boslego J, Garcia J, Cruz C, et al. Efficacy, safety, and immunogenicity the pre-vaccination era. Vaccine 2003; 21:1094–8. of a meningococcal group B (15:P1.3) outer membrane protein vaccine 35. Borrow R, Balmer P, Miller E. Meningococcal surrogates of protec- in Iquique, Chile. Chilean National Committee for Meningococcal Dis- tion—serum bactericidal antibody activity. Vaccine 2005; 23:2222–7. ease. Vaccine 1995; 13:821–9. 36. de Moraes JC, Perkins BA, Camargo MC, et al. Protective efficacy of 17. Jackson C, Lennon D, Sotutu V. Phase II meningococcal B vesicle a serogroup B meningococcal vaccine in Sao Paulo, Brazil. Lancet vaccine trial in New Zealand infants. Arch Dis Child 2008 (Epub ahead 1992; 340:1074–8 (erratum: Lancet 1992; 340:1554). of print). 37. Tappero JW, Lagos R, Ballesteros AM, et al. Immunogenicity of 2 18. Jackson C, Lennon D, Wong S, et al. Persistence of immune response serogroup B outer-membrane protein meningococcal vaccines: a ran- in New Zealand adults and children after 3 doses of MeNZB and domized controlled trial in Chile. JAMA 1999; 281:1520–7. response to a 4th dose in toddlers [abstract G-1698/324]. In: Program 38. Sierra GV, Campa HC, Varcacel NM, et al. Vaccine against group B and abstracts of the 47th Interscience Conference on Antimicrobial Neisseria meningitidis: protection trial and mass vaccination results in Agents and Chemotherapy (Chicago). Washington, DC: American So- Cuba. NIPH Ann 1991; 14:195–207. ciety for Microbiology, 2007. 39. Perkins B, Jonsdottir K, Briem H, et al. Immunogenicity of two effi- 19. Thornton V, Lennon D, Rasanathan K, et al. Safety and immunoge- cacious outer membrane protein-based serogroup B meningococcal nicity of New Zealand strain meningococcal serogroup B OMV vaccine vaccines among young adults in Iceland. J Infect Dis 1998; 177:683–91. in healthy adults: beginning of epidemic control. Vaccine 2006;24: 40. Rosenqvist E, Hoiby EA, Wedege E, et al. Human antibody responses 1395–400. to meningococcal outer membrane antigens after three doses of the 20. Wong S, Lennon D, Jackson C, et al. New Zealand epidemic strain Norwegian group B meningococcal vaccine. Infect Immun 1995; 63: meningococcal B outer membrane vesicle vaccine in children aged 4642–52. 16–24 months. Pediatr Infect Dis J 2007; 26:345–50. 41. Fischer M, Carlone GM, Holst J, Williams D, Stephens DS, Perkins

604 • CID 2009:49 (15 August) • VACCINES BA. Neisseria meningitidis serogroup B outer membrane vesicle vaccine 54. MedSafe. New Zealand Medicines and Medical Devices Safety Au- in adults with occupational risk for meningococcal disease. Vaccine thority. Available at: http://www.medsafe.govt.nz. 12 February 2009. 1999; 17:2377–83. 55. Miller E, Salisbury D, Ramsay M. Planning, registration, and imple- 42. Feiring B, Fuglesang J, Oster P, et al. Persisting immune responses mentation of an immunisation campaign against meningococcal se- indicating long-term protection after booster dose with meningococcal rogroup C disease in the UK: a success story. Vaccine 2001; 20(Suppl group B outer membrane vesicle vaccine. Clin Vaccine Immunol 1):S58-67. 2006; 13:790–6. 56. Sexton K, Lennon D, Oster P, et al. The New Zealand meningococcal 43. Costa EA. Effectiveness of meningococcal vaccine in Brazil. Int J Ep- vaccine strategy: a tailor-made vaccine to combat a devastating epi- idemiol 1997; 26:681–4. demic. N Z Med J 2004; 117:U1015. 44. Kelly C, Arnold R, Galloway Y, O’Hallahan J. A prospective study of 57. Borrow R, Aaberge IS, Santos GF, et al. Interlaboratory standardization the effectiveness of New Zealand meningococcal B vaccine. Am J Ep- of the measurement of serum bactericidal activity by using human idemiol 2007; 166:817–23. complement against meningococcal serogroup b, strain 44/76-SL, be- 45. Martin D, McCallum L, Glennie A, et al. Validation of the serum fore and after vaccination with the Norwegian MenBvac outer bactericidal assay for measurement of functional antibodies against group B meningococci associated with vaccine trials. Vaccine 2005; 23: membrane vesicle vaccine. Clin Diagn Lab Immunol 2005; 12:970–6. 2218–21. 58. MacLennan JM, Shackley F, Heath PT, et al. Safety, immunogenicity, 46. Aavitsland P, Bjune G, Aasen S, Halvorsen S. Adverse events following and induction of immunologic memory by a serogroup C meningo- vaccine or placebo injection in an efficacy trial of an outer membrane coccal conjugate vaccine in infants: a randomized controlled trial. Downloaded from vesicle vaccine against group B meningococcal disease in Norwegian JAMA 2000; 283:2795–801. secondary schools 1988–1991. NIPH Ann 1991; 14:133–4. 59. Kshirsagar N, Mur N, Thatte U, et al. Safety, immunogenicity, and 47. Meningococcal Vaccine Strategy Data Management Group. Menin- antibody persistence of a new meningococcal group A conjugate vac- gococcal B vaccine (MeNZB) safety monitoring plan. Wellington: New cine in healthy Indian adults. Vaccine 2007; 25(Suppl 1):A101–7. Zealand Ministry of Health, 2004. 60. Lystad A, Aasen S. The epidemiology of meningococcal disease in 48. McNicholas A, Galloway Y, Stehr-Green P, et al. Post-marketing safety Norway 1975–91. NIPH Ann 1991; 14:57–65.

monitoring of a new group B meningococcal vaccine in New Zealand, 61. Galloway Y, Stehr-Green P, McNicholas A, O’Hallahan J. Use of an cid.oxfordjournals.org 2004–2006. Hum Vaccin 2007; 3:196–204. observational cohort study to estimate the effectiveness of the New 49. Tatley MV, Kunac DL, McNicholas A, et al. The intensive vaccines Zealand group B meningococcal vaccine in children aged under 5 years. monitoring programme (IVMP): an electronic system to monitor vac- Int J Epidemiol 2009; 38:413–8. cine safety in New Zealand. Vaccine 2008; 26:2746–52. 62. Harrison LH. Vaccine prevention of meningococcal disease: making 50. Scheifele DW, Halperin SA; CPS/Health Canada IMPA. Immunization slow progress. Clin Infect Dis 2006; 43:1395–7. monitoring program, active: a model of active surveillance of vaccine 63. Snape MD, Kelly DF, Salt P, et al. Serogroup C meningococcal gly- safety. Semin Pediatr Infect Dis 2003; 14:213–9.

coconjugate vaccine in adolescents: persistence of bactericidal anti- at World Health Organization on March 14, 2011 51. Ameratunga S, Macmillan A, Stewart J, et al. Evaluating the post- bodies and kinetics of the immune response to a booster vaccine more licensure effectiveness of a group B meningococcal vaccine in New Zealand: a multi-faceted strategy. Vaccine 2005; 23:2231–4. than 3 years after immunization. Clin Infect Dis 2006; 43:1387–94. 52. Lennon D, Stewart J, Crengle S. Re: “a prospective study of the effec- 64. Trotter CL, Andrews NJ, Kaczmarski EB, Miller E, Ramsay ME. Ef- tiveness of the New Zealand meningococcal B vaccine.” Am J Epidemiol fectiveness of meningococcal serogroup C conjugate vaccine 4 years 2008; 167:1140–1. after introduction. Lancet 2004; 364:365–7. 53. Kelly C, Arnold R, Galloway Y, O’Hallahan J. Re: “a prospective study 65. Pizza M, Scarlato V, Masignani V, et al. Identification of vaccine can- of the effectiveness of the New Zealand meningococcal B vaccine.” Am didates against serogroup B meningococcus by whole-genome sequenc- J Epidemiol 2008; 167:1141–2. ing. Science 2000; 287:1816–20.

VACCINES • CID 2009:49 (15 August) • 605