- Email: [email protected]
Impact of 2 + 1 pneumococcal conjugate vaccine program in the province of Quebec, Canada
Vaccine 32 (2014) 1501–1506
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Impact of 2 + 1 pneumococcal conjugate vaccine program in the province of Quebec, Canada Philippe De Wals a,b,c,∗ , Brigitte Lefebvre d , France Markowski e , Geneviève Deceuninck c , Fannie Defay c , Monique Douville-Fradet b , Monique Landry e a
Department of Social and Preventive Medicine, Laval University, Quebec City, Canada Quebec National Public Health Institute (Institut national de Santé publique du Québec), Quebec City, Canada c Quebec University Hospital Research Centre, Quebec City, Canada d Quebec Public Health Laboratory (Laboratoire de santé publique du Québec), Quebec National Public Health Institute (Institut national de Santé publique du Québec), Montreal, Canada e Quebec Ministry of Health and Social Services (Ministère de la Santé et des Services sociaux du Québec), Montreal, Canada b
a r t i c l e
i n f o
Article history: Received 27 March 2013 Received in revised form 17 October 2013 Accepted 6 November 2013 Available online 31 January 2014 Keywords: Pneumococcal conjugate vaccine Immunization program Effectiveness Herd immunity Replacement
a b s t r a c t Background: Quebec was the first jurisdiction in the world to recommend a 3-dose (2 + 1) pneumococcal conjugate vaccine (PCV) schedule. The program was implemented in December 2004 with a catch-up for children <5 years. PCV-7 was first used and replaced, respectively, by PCV-10 in 2009 and by PCV-13 in 2011. Methods: Cases of invasive pneumococcal disease (IPD) notified to public health authorities and isolates submitted to the provincial reference laboratory during the period 2000–2011 were analyzed. Results: IPD incidence in children <5 years was 67/100,000 in 2001–2004, and decreased to 32/100,000 in 2007–2009 following PCV-7 implementation (p < 0.01). A further decrease to 24/100,000 was observed in 2010–2011 following PCV-10 introduction (p < 0.01). PCV-7 serotypes represented 82% of the total IPD cases in 2000–2004 and elimination was achieved in 2011. Main emerging serotypes were 19A and 7F. Children exposed to the PCV-10 experienced lower IPD rates and all serotypes contributed to the decline, mainly 7F and 19A. In adults, a decrease of low magnitude was observed in 2005–2006 but rates in 2007–2009 were higher than in the prevaccination period. Conclusions: A 3-dose PCV schedule with high uptake is highly effective and should be recommended worldwide. Serotype replacement eroded benefits especially in adults. PCV-10 introduction had an effect and the impact of PCV-13 use remains to be evaluated. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Invasive pneumococcal disease (IPD) constitutes a public health problem worldwide [1]. In Canada, the 7-valent CRM197 pneumococcal conjugate vaccine (PCV-7) was licensed and available on the private market in May 2001 [2]. Publicly funded immunization programs are a provincial/territorial jurisdiction and there are variations in the timing of program implementation and recommended schedule [3]. In the province of Quebec, PCV-7 was offered free of charge to high-risk children and those living in two northern
∗ Corresponding author at: Département de médecine sociale et préventive, Université Laval, 1050 Avenue de la Médecine, Québec, QC, Canada G1V 0A6. Tel.: +1 418 666 7000. E-mail addresses: [email protected], [email protected] (P. De Wals). 0264-410X/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.vaccine.2013.11.028
regions in October 2002, and four (3 + 1) doses were recommended in accordance with the guidelines of the National Advisory Committee on Immunization [2]. Following a careful scientific evaluation, Quebec was the first jurisdiction in the world to recommend a 3-dose (2 + 1) schedule (2, 4, and 12 months) for the routine immunization of low-risk infants [4]. This publicly funded program was launched in December 2004, along with a catch-up program consisting of 2 doses for 12–23 month-old children and 1 dose for 2–5 year-old children. Vaccine uptake was low (<20%) up to the end of the year 2004 and increased very rapidly thereafter [5]. Since 2005, more than 97% of infants are vaccinated by age 24 months and 93% are receiving the recommended number of doses [6]. The licensing of new generation vaccines lead to the introduction of the 10-valent protein-D pneumococcal conjugate vaccine (PCV-10) during the summer of 2009, and the introduction of the 13-valent CRM197 vaccine (PCV-13) in January 2011, with no catch-up in both instances [7,8]. The objective of the study, based on notifiable disease registry
1502
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
and laboratory surveillance data, was to assess the impact of the PCV program using different vaccines in succession on the incidence and serotype distribution of IPD in children and adults. This study adds to the existing literature regarding the impact of PCV7 programs using different schedules in different epidemiological contexts [9]. 2. Methods 2.1. Notifiable disease data IPD is a notifiable disease in Quebec since 2000 and is defined as a clinical infection associated with the identification of Streptococcus pneumoniae (Sp) by culture or nucleic acid amplification test (NAAT) in a normally sterile body fluid or site. IPD cases are reported to regional public health authorities by physicians (rarely) and laboratories (systematically). Following notification, a short enquiry is conducted by the public health services to validate the diagnosis. Data are entered into a web-based provincial registry file. 2.2. Laboratory surveillance data In 1996, the provincial reference laboratory initiated a laboratory surveillance program. Sentinel laboratories were invited to submit all their IPD isolates for bacteriologic confirmation and strain characterization. Sentinel laboratories include all major tertiary care university hospitals and a sample of regional hospitals located throughout the province. The size of the population covered by sentinel laboratories (n = 22) is not known precisely but is believed to have remained fairly constant over the period 2000–2011. In 2005, all microbiology laboratories in the province were requested to submit invasive Sp isolates from children <5 years of age to the reference laboratory. Serotype identification was performed using the traditional capsular swelling method (Quellung reaction) and, for selected serogroups, by a monoclonal antibody technique. Polymerase chain reaction (PCR) amplification of ply and Spn9802 genes was also used for some culture-negative specimens and no serotype identification was available for these cases. 2.3. Statistical methods Yearly incidence rates were calculated using population estimates derived from census data in denominators. The 2006 age-distribution was selected as a reference for standardization of rates (2006: population = 7631,552; livebirths = 81,962). Ageadjusted rate ratios (RR) and their statistical significance (p < 0.05; two-sided test) were computed by a Poisson model (SAS 9.2. software; SAS Institute, Cary, NC). No adjustment of p values was made for multiple statistical testing. Age and season-adjusted comparisons of cumulative incidence rates were performed between cohorts of births exposed to PCV-7 and/or PCV-10 for the infant primary immunization series and the booster toddler dose. The main vaccine use in each monthly cohort was estimated from the population-based immunization registry in the Quebec City area. Details on the methodology are provided in a previous publication [10]. 3. Results 3.1. Notifiable disease data During the period 2000–2011, a total of 10,437 IPD cases were reported in the Quebec population representing 94,938,086 person-years of observation (average rate = 11.0/100,000). The
Fig. 1. Age-specific rates of invasive pneumococcal disease as reported in the Registry of notifiable diseases in the province of Quebec, 2000–2011.
diagnostic laboratory method was recorded in 10,175 cases (97.4% of total) and the information was missing for the remaining cases. Sp was identified by culture in 10,162 cases and by NAAT in 13 cases. Age-adjusted incidence rates by calendar year are shown in Fig. 1 and statistical analyses are provided in Table 1. In children less than 5 years of age, IPD rate was 67.0/100,000 in 2001–2004, a period during which PCV-7 use was low. A marked decrease was observed following implementation of the PCV-7 program in December 2004, and the IPD rate was 31.9/100,000 in 2007–2009 (RR = 0.46; p < 0.01). A further decrease in rate was seen following PCV-10 introduction in 2009. This was only observed in children less than 3 years of age, those exposed to this new vaccine (see Fig. 1). Among persons ≥5-year old, a reduction of small magnitude in IPD rate was observed in 2005–2006 following PCV-7 introduction, but the incidence increased thereafter. 3.2. Laboratory surveillance data From 2000 to 2011, a total of 1.135 IPD cases in children <5 years were reported by sentinel laboratories (Fig. 2). The average yearly number of cases was 141 during the period 2000–2004, and started to decrease in 2005 to reach a minimum of 38 cases in 2006. It increased thereafter to reach 80 cases in 2009, with a second decrease in 2010–2011. During the 2001–2004 prevaccination period, PCV-7 serotypes represented 82% of the total cases (586/714). No such case was reported in 2011. The frequency of 6A cases also decreased: five cases per year in average during the 2001–2004 period and one case per year in average during the 2007–2011 period. The main emerging clones following PCV7 introduction belonged to serotypes 19A and 7F. However, the increasing trend of these two serotypes was reversed following PCV-10 introduction in 2009. As shown in Fig. 3, the pattern was different for persons ≥5 years of age (total of 3642 isolates). The decrease in the frequency of PCV7 serotypes was more gradual than in young children, the emerging serotypes were more diverse, and the magnitude of replacement was much higher. A more comprehensive picture is available for the period 2005–2011, when all laboratories submitted their isolates from children <5 years (Table 2). In 2005, 59 IPD cases in children <5 years of age were reported by the sentinel laboratories and 114 cases were reported by all laboratories (coverage of sentinel laboratories = 51.7%). The lowest IPD rate was observed in
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
1503
Table 1 Invasive pneumococcal disease rates (per 100,000 person-years) computed from the registry of notifiable diseases during four periods: (a) prior to PCV-7 program implementation (2001–2004), (b) during PCV-7 program implementation (2005–2006), (c) during PCV-7 program continuation (2007–2009), and (d) during PCV-10 program implementation (2010–2011). PCV-13 was introduced in 2011. Relative rates (RR) and statistical comparisons are (a) vs (b), (a) vs (c), and (c) vs (d). Period
0 to <5 years Rate RR p Value
2001–2004
2005–2006
2007–2009
2010–2011
(a)
(b)
(c)
(d)
67.0
31.9 0.46 <0.01
23.8 0.75 <0.01
–
26.4 0.39 <0.01
0 to <1 year Rate RR p Value
98.1 – –
37.4 0.38 <0.01
47.9 0.49 <0.01
27.3 0.57 <0.01
1 to <3 years Rate RR p Value
101.5 – –
34.6 0.34 <0.01
41.2 0.41 <0.01
30.9 0.75 0.015
3 to <5 years Rate RR p Value
19.4 – –
12.7 0.65 0.02
13.1 0.67 0.01
14.7 1.12 0.54
3.3
3.2 1.01 0.95
2.6 0.83 0.14
3.3 1.23 0.16
6.2
6.1 0.98 0.69
7.1 1.12 0.01
7.7 1.09 0.09
19.5
17.6 0.91 0.10
20.8 1.08 0.10
21.6 1.04 0.43
45.0
44.1 0.97 0.76
51.1 1.13 0.08
52.8 1.03 0.71
5 to <20 years Rate RR p Value 20 to <60 years Rate RR p Value 60 to <80 years Rate RR p Value 80+ years Rate RR p Value
Fig. 2. Serotypes in cases of invasive pneumococcal disease among children less than 5 years of age as reported by sentinel laboratories to the Quebec Public Health Laboratory, 2005–2011.
1504
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
Fig. 3. Serotypes in cases of invasive pneumococcal disease among persons ≥5 years of age as reported by sentinel laboratories to the Quebec Public Health Laboratory, 2005–2011.
2006 (20.2/100,000) and the rate increased steadily thereafter until 2009 (32.9/100,000). Following PCV-10 introduction in 2009, the frequency of 7F isolates decreased markedly in children 6–23 months of age exposed to this vaccine. An early indication of the impact of PCV-13 introduction in January 2011 is the low frequency of 19A isolates in children 12–23 months of age, a proportion of them having been primed with PCV-10 and boosted with PCV-13.
The impact of PCV-10 introduction in the summer of 2009 can be explored in an ecological comparison between a cohort of children in which PCV-7 was used both for the primary immunization series and the booster dose and a subsequent cohort of children born during the same months and in which PCV-10 was used (Table 3). PCV-10 use was associated with a significantly lower IPD incidence (p = 0.02), which was attributable to a lower frequency of all categories of serotypes.
Table 2 Cases (and rate per 100,000 person-years) of invasive pneumococcal disease in children less than 5 years of age reported to the Quebec Public Health Laboratory, 2005–2011. Age-group
Serotype
2008
2009
2010
0 to <6 months
PCV-7 1 7F 19A Others Total Rate
2005 7 0 2 0 3 12 32.1
2006 4 0 3 2 3 12 30.9
2007 1 0 3 1 7 12 29.00
1 0 1 3 5 10 23.4
2 0 2 5 3 12 27.1
1 0 3 6 2 12 27.2
2011 0 0 2 4 5 11 25.0
6 months to <1 year
PCV-7 1 7F 19A Others Total Rate
8 0 0 3 3 14 37.5
3 0 2 3 7 15 38.6
5 0 1 7 14 27 65.2
1 0 2 13 10 26 60.9
1 0 5 11 7 24 54.2
0 0 0 2 0 2 4.5
0 0 0 3 6 9 20.4
1 to <2 years
PCV-7 1 7F 19A Others Total Rate
27 0 0 5 13 45 60.2
7 0 1 6 13 27 36.1
1 0 4 15 19 39 49.7
1 0 2 22 19 44 52.6
0 1 9 27 19 56 64.9
2 0 1 26 20 49 54.8
0 0 2 9 22 32 36.0
2 to <5 years
PCV-7 1 7F 19A Others Total Rate
25 0 1 4 13 43 19.2
6 1 3 8 4 22 9.8
5 3 2 5 16 31 13.7
1 1 2 20 17 41 17.7
3 0 4 18 20 45 18.6
1 3 2 22 10 38 15.0
0 1 3 16 15 35 13.2
All <5 years
Total Rate
114 30.6
76 20.2
109 28.2
121 30.2
137 33.00
101 23.4
87 19.6
23 (30.8)a 11 (14.8) 10 (13.4) 1 (1.3) 1 (1.3) Rate ratio = 0.55; p = 0.02. Other comparisons are not statistically significant. a
June–October 2009
1505
4. Discussion
PCV-10 + PCV-10 38,829 (74,557)
40 (56.3) 20 (28.1) 15 (21.1) 4 (5.6) 1 (1.4) PCV7 + PCV7
6–30 months (December 2007–December 2009) 6–30 months (December 2009–December 2011) June–October 2007
37,086 (71,081)
PCV-7 types
Main vaccine primary series + booster Persons at risk (person-years) Age observed (time period) Birth cohort
Table 3 Cases and rate of invasive pneumococcal disease in cohorts of children exposed to different pneumococcal conjugate vaccines.
IPD cases (rate per 100,000 person-years)
PCV-10 types
19A
Other types
All types
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
In Quebec, the introduction of a 2 + 1 dose PCV program with high uptake was followed by the complete elimination of vaccineserotypes and a marked reduction in all-type IPD incidence in children. This is congruent with results of studies in European countries where a 2 + 1 schedule is also recommended [11,12]. Elimination or quasi-elimination of vaccine-serotypes in the target population was observed in the US and in the Canadian province of Alberta where 3 + 1 doses PCV programs were implemented [13,14]. Other studies using different methodologies did not show substantial differences in the efficacy of 2 + 1 and 3 + 1 regimens to prevent IPD, lower respiratory tract infections, and otitis media in children [15–17]. The number of doses is a major driver of the acceptability and cost-effectiveness of immunization programs [18]. In view of this study, along with the accruing evidence on direct effectiveness of 2 + 1 schedule [9], PCV schedules relying on three doses instead of four should definitively be recommended worldwide. In recent years, all other Canadian provinces have moved from a 3 + 1 toward a 2 + 1 PCV schedule [19]. The overall impact of PCV programs is influenced by indirect effects including herd protection and serotype replacement. In Quebec, serotype replacement completely eroded the benefits of herd protection in adults. Different levels of replacement have been observed in different countries [9]. In the US, the magnitude of serotype replacement in children and adults was much lower than in Quebec [13]. It may be that the speed at which PCV-7 was introduced with active catch-up in Quebec could have played a role in creating an ecological vacuum that was rapidly occupied by non-vaccine clones with invasiveness potential [20]. Another contributing factor could have been the high proportion of children attending day-care services in Quebec (61% in 2000–2001), thanks to a very generous publicly funded program [21]. Replacement was also rapid and of high magnitude in the Native communities of Alaska [22]. It may be that the frequency and intimacy of contacts among children could influence the speed and magnitude of serotype replacement. This is the first experience analyzing the transition from PCV7 to PCV-10 for the routine immunization of children. Children exposed to the new vaccine experienced lower IPD rates and all serotypes contributed to the decline. Preliminary results published in a previous publication are thus confirmed [10]. Protection against 7F and some level of cross-protection against 19A were expected in the light of results of immunogenicity studies with PCV-10 showing the induction of functional antibodies [23]. At this time, there is no good explanation for the low rate of other serotypes not included in the vaccine. In the US (excluding Alaska), the introduction of PCV13 following PCV-7 use was associated with a decline in overall IPD rates among children, a marked decline for the six additional serotypes and a low level of replacement [24]. It will be interesting to analyze the transition from PCV-10 to PCV-13 in Quebec. There are limitations in the surveillance system for IPD in Quebec. Mandatory notification of IPD was enforced in 2000, and improvement in reporting can be suspected during the first year. For this reason, the year 2000 was excluded from statistical analyses. Another weakness is the poor quality of information regarding the clinical presentation and the immunization status of patients. For these reasons, a case–control study was initiated in 2005, and preliminary results confirmed the high effectiveness of the 2 + 1 schedule [25]. The direct protection provided by PCV can be measured in a case–control study, although ecological analyses are needed to capture indirect effects, including herd protection and replacement. Sentinel laboratories provided an unrepresentative albeit constant sample of total IPD cases diagnosed in the province. The population covered by this network cannot be defined precisely and no incidence rate could be computed. The quality of
1506
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
laboratory surveillance for IPD in children was improved in 2005, when all microbiology laboratories in the province were invited to participate instead of a limited number of sentinel laboratories. In a record linkage study performed in 2006, the sensitivity of the notifiable disease registry was 94% and was 89% for the reference laboratory, while the concordance between the two sources was 95% [26]. There are plans to extend laboratory surveillance for IPD in adults to all microbiology laboratories in the province but budgetary constraint is an issue. In conclusion, a 2 + 1 PCV schedule with high uptake is highly effective to prevent IPD and should be recommended worldwide. Unfortunately, in Quebec, serotype replacement eroded indirect benefits especially in adults. PCV-10 introduction had a marked effect and the impact of PCV-13 use remains to be evaluated. Funding The study was supported by a research grant from the Quebec Ministry of Health and Social Services (Ministère de la Santé et des Services sociaux du Québec). Acknowledgments The authors thank all health professionals who are contributing to the surveillance of invasive pneumococcal disease in the province of Quebec. Conflict of interest statement: Philippe De Wals received research grants, honoraria and reimbursements of travel expenses from vaccine manufacturers including Glaxo-Smith-Kline, Novartis, Sanofi Pasteur, Merck and Wyeth, as well as from governmental agencies including the Quebec Ministry of Health and Social Services, Health Canada, and the Public Health Agency of Canada. The other authors have no conflict of interests. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vaccine. 2013.11.028. References [1] Butler JC. Epidemiology of pneumococcal disease. In: Tuomanen EI, Mitchell TJ, Morrison DA, Spratt BG, editors. The pneumococcus. Washington, DC: ASM Press; 2004. p. 136–47. [2] National Advisory Committee on Immunization (NACI). Statement on recommended use of pneumococcal conjugate vaccine. CCDR 2002;28(ACS-2):1–32. [3] De Wals P. Optimizing the acceptability, effectiveness and costs of immunization programs: the Quebec experience. Expert Rev Vaccines 2011;10:55–62. [4] Quebec Immunization Committee. Assessment of the appropriateness of an immunization program for pneumococcal infections in children using a reduced number of doses of conjugate vaccine. Montreal, QC: Institut national de santé publique du Québec; 2005. Available at:. [5] De Wals P, Boulianne N, Sévin É, Ouakki M, Deceuninck G, Guay M. Uptake of pneumococcal conjugate vaccine: methodological issues in measurement and impact of publicly funded programs. Can J Public Health 2009;100:413–6. [6] Boulianne N, Bradet R, Audet D, Ouakki M. Enquête sur la couverture vaccinale des enfants de 1 an et 2 ans au Québec en 2010 [Survey on vaccination coverage of children one and two years of age in Quebec, 2010]. Quebec City, QC: Institut national de santé publique du Québec; 2011. Available at: .
[7] Comité sur l’immunisation du Québec (CIQ). Évaluation de la pertinence d’un nouveau vaccin antipneumococcique conjugué au Québec [Evaluation of the appropriateness of a new pneumococcal conjugate vaccine in Quebec]. Quebec City, QC: Institut national de santé publique du Québec; 2009. Available at:. [8] Comité sur l’immunisation du Québec (CIQ). Évaluation de deux nouveaux vaccins pneumococciques conjugués pour l’immunisation des enfants au Québec [Evaluation of two new pneumococcal conjugate vaccines for the immunization of children in Quebec]. Quebec City, QC: Institut national de santé publique du Québec; 2010. Available at: . [9] Myint TTH, Madhava H, Balmer P, Christopoulou D, Attal S, Menegas D, et al. The impact of 7-valent pneumococcal conjugate vaccine on invasive pneumococcal disease: a literature review. Adv Ther 2013, http://dx.doi.org/10.1007/s12325-013-0007-6. [10] De Wals P, Lefebvre B, Defay F, Deceuninck G, Boulianne N. Invasive pneumococcal diseases in birth cohorts vaccinated with PCV-7 and/or PHiD-CV in the province of Quebec, Canada. Vaccine 2012;30:6416–20. [11] Miller E, Andrews NJ, Waight PA, Slack MPE, George RC. Herd immunity and serotype replacement 4 years after seven-valent pneumococcal conjugate vaccination in England and Wales: an observational cohort study. Lancet Infect Dis 2011;11:760–8. [12] Harboe ZB, Valentiner-Branth P, Ingels H, Rasmussen JN, Andersen PHS, Bjerre CC, et al. Pediatric invasive pneumococcal disease caused by vaccine serotypes following the introduction of conjugate vaccination in Denmark. PLoS One 2013;8:e51460, http://dx.doi.org/10.1371/journal.pone.0051460. [13] Pilishvili T, Lexau C, Farley MM, Hadler J, Harrison LH, Bennett NM, et al. Sustained reductions in invasive pneumococcal disease in the era of conjugate vaccine. J Infect Dis 2010;201:32–41. [14] Leal J, Vanderkooi OG, Church DL, MacDonald J, Tyrrell GJ, Kellner JD. Eradication of invasive pneumococcal disease due to the seven-valent pneumococcal conjugate vaccine serotypes in Calgary, Alberta. Pediatr Infect Dis J 2012;31:e169–75. [15] Palmu AA, Jokinen J, Borys D, Nieminen H, Ruokokoski E, Siira L, et al. Effectiveness of the ten-valent pneumococcal Haemophilus influenza protein D conjugate vaccine (PHiD-CV10) against invasive pneumococcal disease: a cluster randomised trial. Lancet 2013;381:214–22. [16] Pelton SI, Weycker D, Klein JO, Strutton D, Ciuryla V, Oster G. 7-Valent pneumococcal conjugate vaccine and lower respiratory tract infections: effectiveness of a 2-dose versus 3-dose primary series. Vaccine 2010;28:1575–82. [17] Stoecker C, Hampton LM, Moore MR. 7-Valent pneumococcal conjugate vaccine and otitis media: effectiveness of a 2-dose versus 3-dose primary series. Vaccine 2012;30:6256–62. [18] Beutels P, Thiry N, Van Damme P. Convincing or confusing? Economic evaluations of childhood pneumococcal conjugate vaccination-a review (2002–2006). Vaccine 2007;25:1355–67. [19] Public Health Agency of Canada. Publicly funded immunization programs in Canada. Routine schedule for infants and children including special programs and catch-up programs (as of December 2012). Ottawa, ON: Public Health Agency of Canada; 2013. Available at: . [20] Weinberger DM, Malley R, Lipsitch M. Serotype replacement in disease after pneumococcal vaccination. Lancet 2011;378:1962–73. [21] Statistics Canada. The Daily Child care: an eight-year profile 1994–1995 to 2002–2003. Available at: . [22] Singleton RJ, Hennessy TW, Bulkow LR, Hammitt LL, Zulzt T, Hurlburt DA, et al. Invasive pneumococcal disease caused by nonvaccine serotypes among Alaska Native children with high levels of 7-valent pneumococcal conjugate vaccine coverage. J Am Med Assoc 2007;297:1784–92. [23] Prymula R, Schuerman L. 10-Valent pneumococcal nontypeable Haemophilus influenzae PD conjugate vaccine: SynflorixTM . Expert Rev Vaccines 2009;8:1479–500. [24] Kaplan SL, Barson WJ, Lin PL, Romero JR, Bradley JS, Tan TQ, et al. Early trends for invasive pneumococcal infections in children after the introduction of the 13-valent pneumococcal conjugate vaccine. Pediatr Infect Dis J 2013;32:203–7. [25] Deceuninck G, De Wals P, Boulianne N, De Serres G. Effectiveness of pneumococcal conjugate vaccine using a 2 + 1 infant schedule in Quebec, Canada. Pediatr Infect Dis J 2010;29:546–9. [26] Douville-Fradet M, Amini R, Boulianne N, Khuc NH, De Wals P, Fortin É, et al. Impact du programme d’immunisation par le vaccin pneumococcique conjugué heptavalent (VPC-7) au Québec [Impact of the immunization program using the heptavalent pneumococcal conjugate vaccine (VPC-7) in Quebec]. Quebec City, QC: Institut national de santé publique du Québec; 2011. Available at: .
Contents lists available at ScienceDirect
Vaccine journal homepage: www.elsevier.com/locate/vaccine
Impact of 2 + 1 pneumococcal conjugate vaccine program in the province of Quebec, Canada Philippe De Wals a,b,c,∗ , Brigitte Lefebvre d , France Markowski e , Geneviève Deceuninck c , Fannie Defay c , Monique Douville-Fradet b , Monique Landry e a
Department of Social and Preventive Medicine, Laval University, Quebec City, Canada Quebec National Public Health Institute (Institut national de Santé publique du Québec), Quebec City, Canada c Quebec University Hospital Research Centre, Quebec City, Canada d Quebec Public Health Laboratory (Laboratoire de santé publique du Québec), Quebec National Public Health Institute (Institut national de Santé publique du Québec), Montreal, Canada e Quebec Ministry of Health and Social Services (Ministère de la Santé et des Services sociaux du Québec), Montreal, Canada b
a r t i c l e
i n f o
Article history: Received 27 March 2013 Received in revised form 17 October 2013 Accepted 6 November 2013 Available online 31 January 2014 Keywords: Pneumococcal conjugate vaccine Immunization program Effectiveness Herd immunity Replacement
a b s t r a c t Background: Quebec was the first jurisdiction in the world to recommend a 3-dose (2 + 1) pneumococcal conjugate vaccine (PCV) schedule. The program was implemented in December 2004 with a catch-up for children <5 years. PCV-7 was first used and replaced, respectively, by PCV-10 in 2009 and by PCV-13 in 2011. Methods: Cases of invasive pneumococcal disease (IPD) notified to public health authorities and isolates submitted to the provincial reference laboratory during the period 2000–2011 were analyzed. Results: IPD incidence in children <5 years was 67/100,000 in 2001–2004, and decreased to 32/100,000 in 2007–2009 following PCV-7 implementation (p < 0.01). A further decrease to 24/100,000 was observed in 2010–2011 following PCV-10 introduction (p < 0.01). PCV-7 serotypes represented 82% of the total IPD cases in 2000–2004 and elimination was achieved in 2011. Main emerging serotypes were 19A and 7F. Children exposed to the PCV-10 experienced lower IPD rates and all serotypes contributed to the decline, mainly 7F and 19A. In adults, a decrease of low magnitude was observed in 2005–2006 but rates in 2007–2009 were higher than in the prevaccination period. Conclusions: A 3-dose PCV schedule with high uptake is highly effective and should be recommended worldwide. Serotype replacement eroded benefits especially in adults. PCV-10 introduction had an effect and the impact of PCV-13 use remains to be evaluated. © 2013 Elsevier Ltd. All rights reserved.
1. Introduction Invasive pneumococcal disease (IPD) constitutes a public health problem worldwide [1]. In Canada, the 7-valent CRM197 pneumococcal conjugate vaccine (PCV-7) was licensed and available on the private market in May 2001 [2]. Publicly funded immunization programs are a provincial/territorial jurisdiction and there are variations in the timing of program implementation and recommended schedule [3]. In the province of Quebec, PCV-7 was offered free of charge to high-risk children and those living in two northern
∗ Corresponding author at: Département de médecine sociale et préventive, Université Laval, 1050 Avenue de la Médecine, Québec, QC, Canada G1V 0A6. Tel.: +1 418 666 7000. E-mail addresses: [email protected], [email protected] (P. De Wals). 0264-410X/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.vaccine.2013.11.028
regions in October 2002, and four (3 + 1) doses were recommended in accordance with the guidelines of the National Advisory Committee on Immunization [2]. Following a careful scientific evaluation, Quebec was the first jurisdiction in the world to recommend a 3-dose (2 + 1) schedule (2, 4, and 12 months) for the routine immunization of low-risk infants [4]. This publicly funded program was launched in December 2004, along with a catch-up program consisting of 2 doses for 12–23 month-old children and 1 dose for 2–5 year-old children. Vaccine uptake was low (<20%) up to the end of the year 2004 and increased very rapidly thereafter [5]. Since 2005, more than 97% of infants are vaccinated by age 24 months and 93% are receiving the recommended number of doses [6]. The licensing of new generation vaccines lead to the introduction of the 10-valent protein-D pneumococcal conjugate vaccine (PCV-10) during the summer of 2009, and the introduction of the 13-valent CRM197 vaccine (PCV-13) in January 2011, with no catch-up in both instances [7,8]. The objective of the study, based on notifiable disease registry
1502
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
and laboratory surveillance data, was to assess the impact of the PCV program using different vaccines in succession on the incidence and serotype distribution of IPD in children and adults. This study adds to the existing literature regarding the impact of PCV7 programs using different schedules in different epidemiological contexts [9]. 2. Methods 2.1. Notifiable disease data IPD is a notifiable disease in Quebec since 2000 and is defined as a clinical infection associated with the identification of Streptococcus pneumoniae (Sp) by culture or nucleic acid amplification test (NAAT) in a normally sterile body fluid or site. IPD cases are reported to regional public health authorities by physicians (rarely) and laboratories (systematically). Following notification, a short enquiry is conducted by the public health services to validate the diagnosis. Data are entered into a web-based provincial registry file. 2.2. Laboratory surveillance data In 1996, the provincial reference laboratory initiated a laboratory surveillance program. Sentinel laboratories were invited to submit all their IPD isolates for bacteriologic confirmation and strain characterization. Sentinel laboratories include all major tertiary care university hospitals and a sample of regional hospitals located throughout the province. The size of the population covered by sentinel laboratories (n = 22) is not known precisely but is believed to have remained fairly constant over the period 2000–2011. In 2005, all microbiology laboratories in the province were requested to submit invasive Sp isolates from children <5 years of age to the reference laboratory. Serotype identification was performed using the traditional capsular swelling method (Quellung reaction) and, for selected serogroups, by a monoclonal antibody technique. Polymerase chain reaction (PCR) amplification of ply and Spn9802 genes was also used for some culture-negative specimens and no serotype identification was available for these cases. 2.3. Statistical methods Yearly incidence rates were calculated using population estimates derived from census data in denominators. The 2006 age-distribution was selected as a reference for standardization of rates (2006: population = 7631,552; livebirths = 81,962). Ageadjusted rate ratios (RR) and their statistical significance (p < 0.05; two-sided test) were computed by a Poisson model (SAS 9.2. software; SAS Institute, Cary, NC). No adjustment of p values was made for multiple statistical testing. Age and season-adjusted comparisons of cumulative incidence rates were performed between cohorts of births exposed to PCV-7 and/or PCV-10 for the infant primary immunization series and the booster toddler dose. The main vaccine use in each monthly cohort was estimated from the population-based immunization registry in the Quebec City area. Details on the methodology are provided in a previous publication [10]. 3. Results 3.1. Notifiable disease data During the period 2000–2011, a total of 10,437 IPD cases were reported in the Quebec population representing 94,938,086 person-years of observation (average rate = 11.0/100,000). The
Fig. 1. Age-specific rates of invasive pneumococcal disease as reported in the Registry of notifiable diseases in the province of Quebec, 2000–2011.
diagnostic laboratory method was recorded in 10,175 cases (97.4% of total) and the information was missing for the remaining cases. Sp was identified by culture in 10,162 cases and by NAAT in 13 cases. Age-adjusted incidence rates by calendar year are shown in Fig. 1 and statistical analyses are provided in Table 1. In children less than 5 years of age, IPD rate was 67.0/100,000 in 2001–2004, a period during which PCV-7 use was low. A marked decrease was observed following implementation of the PCV-7 program in December 2004, and the IPD rate was 31.9/100,000 in 2007–2009 (RR = 0.46; p < 0.01). A further decrease in rate was seen following PCV-10 introduction in 2009. This was only observed in children less than 3 years of age, those exposed to this new vaccine (see Fig. 1). Among persons ≥5-year old, a reduction of small magnitude in IPD rate was observed in 2005–2006 following PCV-7 introduction, but the incidence increased thereafter. 3.2. Laboratory surveillance data From 2000 to 2011, a total of 1.135 IPD cases in children <5 years were reported by sentinel laboratories (Fig. 2). The average yearly number of cases was 141 during the period 2000–2004, and started to decrease in 2005 to reach a minimum of 38 cases in 2006. It increased thereafter to reach 80 cases in 2009, with a second decrease in 2010–2011. During the 2001–2004 prevaccination period, PCV-7 serotypes represented 82% of the total cases (586/714). No such case was reported in 2011. The frequency of 6A cases also decreased: five cases per year in average during the 2001–2004 period and one case per year in average during the 2007–2011 period. The main emerging clones following PCV7 introduction belonged to serotypes 19A and 7F. However, the increasing trend of these two serotypes was reversed following PCV-10 introduction in 2009. As shown in Fig. 3, the pattern was different for persons ≥5 years of age (total of 3642 isolates). The decrease in the frequency of PCV7 serotypes was more gradual than in young children, the emerging serotypes were more diverse, and the magnitude of replacement was much higher. A more comprehensive picture is available for the period 2005–2011, when all laboratories submitted their isolates from children <5 years (Table 2). In 2005, 59 IPD cases in children <5 years of age were reported by the sentinel laboratories and 114 cases were reported by all laboratories (coverage of sentinel laboratories = 51.7%). The lowest IPD rate was observed in
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
1503
Table 1 Invasive pneumococcal disease rates (per 100,000 person-years) computed from the registry of notifiable diseases during four periods: (a) prior to PCV-7 program implementation (2001–2004), (b) during PCV-7 program implementation (2005–2006), (c) during PCV-7 program continuation (2007–2009), and (d) during PCV-10 program implementation (2010–2011). PCV-13 was introduced in 2011. Relative rates (RR) and statistical comparisons are (a) vs (b), (a) vs (c), and (c) vs (d). Period
0 to <5 years Rate RR p Value
2001–2004
2005–2006
2007–2009
2010–2011
(a)
(b)
(c)
(d)
67.0
31.9 0.46 <0.01
23.8 0.75 <0.01
–
26.4 0.39 <0.01
0 to <1 year Rate RR p Value
98.1 – –
37.4 0.38 <0.01
47.9 0.49 <0.01
27.3 0.57 <0.01
1 to <3 years Rate RR p Value
101.5 – –
34.6 0.34 <0.01
41.2 0.41 <0.01
30.9 0.75 0.015
3 to <5 years Rate RR p Value
19.4 – –
12.7 0.65 0.02
13.1 0.67 0.01
14.7 1.12 0.54
3.3
3.2 1.01 0.95
2.6 0.83 0.14
3.3 1.23 0.16
6.2
6.1 0.98 0.69
7.1 1.12 0.01
7.7 1.09 0.09
19.5
17.6 0.91 0.10
20.8 1.08 0.10
21.6 1.04 0.43
45.0
44.1 0.97 0.76
51.1 1.13 0.08
52.8 1.03 0.71
5 to <20 years Rate RR p Value 20 to <60 years Rate RR p Value 60 to <80 years Rate RR p Value 80+ years Rate RR p Value
Fig. 2. Serotypes in cases of invasive pneumococcal disease among children less than 5 years of age as reported by sentinel laboratories to the Quebec Public Health Laboratory, 2005–2011.
1504
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
Fig. 3. Serotypes in cases of invasive pneumococcal disease among persons ≥5 years of age as reported by sentinel laboratories to the Quebec Public Health Laboratory, 2005–2011.
2006 (20.2/100,000) and the rate increased steadily thereafter until 2009 (32.9/100,000). Following PCV-10 introduction in 2009, the frequency of 7F isolates decreased markedly in children 6–23 months of age exposed to this vaccine. An early indication of the impact of PCV-13 introduction in January 2011 is the low frequency of 19A isolates in children 12–23 months of age, a proportion of them having been primed with PCV-10 and boosted with PCV-13.
The impact of PCV-10 introduction in the summer of 2009 can be explored in an ecological comparison between a cohort of children in which PCV-7 was used both for the primary immunization series and the booster dose and a subsequent cohort of children born during the same months and in which PCV-10 was used (Table 3). PCV-10 use was associated with a significantly lower IPD incidence (p = 0.02), which was attributable to a lower frequency of all categories of serotypes.
Table 2 Cases (and rate per 100,000 person-years) of invasive pneumococcal disease in children less than 5 years of age reported to the Quebec Public Health Laboratory, 2005–2011. Age-group
Serotype
2008
2009
2010
0 to <6 months
PCV-7 1 7F 19A Others Total Rate
2005 7 0 2 0 3 12 32.1
2006 4 0 3 2 3 12 30.9
2007 1 0 3 1 7 12 29.00
1 0 1 3 5 10 23.4
2 0 2 5 3 12 27.1
1 0 3 6 2 12 27.2
2011 0 0 2 4 5 11 25.0
6 months to <1 year
PCV-7 1 7F 19A Others Total Rate
8 0 0 3 3 14 37.5
3 0 2 3 7 15 38.6
5 0 1 7 14 27 65.2
1 0 2 13 10 26 60.9
1 0 5 11 7 24 54.2
0 0 0 2 0 2 4.5
0 0 0 3 6 9 20.4
1 to <2 years
PCV-7 1 7F 19A Others Total Rate
27 0 0 5 13 45 60.2
7 0 1 6 13 27 36.1
1 0 4 15 19 39 49.7
1 0 2 22 19 44 52.6
0 1 9 27 19 56 64.9
2 0 1 26 20 49 54.8
0 0 2 9 22 32 36.0
2 to <5 years
PCV-7 1 7F 19A Others Total Rate
25 0 1 4 13 43 19.2
6 1 3 8 4 22 9.8
5 3 2 5 16 31 13.7
1 1 2 20 17 41 17.7
3 0 4 18 20 45 18.6
1 3 2 22 10 38 15.0
0 1 3 16 15 35 13.2
All <5 years
Total Rate
114 30.6
76 20.2
109 28.2
121 30.2
137 33.00
101 23.4
87 19.6
23 (30.8)a 11 (14.8) 10 (13.4) 1 (1.3) 1 (1.3) Rate ratio = 0.55; p = 0.02. Other comparisons are not statistically significant. a
June–October 2009
1505
4. Discussion
PCV-10 + PCV-10 38,829 (74,557)
40 (56.3) 20 (28.1) 15 (21.1) 4 (5.6) 1 (1.4) PCV7 + PCV7
6–30 months (December 2007–December 2009) 6–30 months (December 2009–December 2011) June–October 2007
37,086 (71,081)
PCV-7 types
Main vaccine primary series + booster Persons at risk (person-years) Age observed (time period) Birth cohort
Table 3 Cases and rate of invasive pneumococcal disease in cohorts of children exposed to different pneumococcal conjugate vaccines.
IPD cases (rate per 100,000 person-years)
PCV-10 types
19A
Other types
All types
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
In Quebec, the introduction of a 2 + 1 dose PCV program with high uptake was followed by the complete elimination of vaccineserotypes and a marked reduction in all-type IPD incidence in children. This is congruent with results of studies in European countries where a 2 + 1 schedule is also recommended [11,12]. Elimination or quasi-elimination of vaccine-serotypes in the target population was observed in the US and in the Canadian province of Alberta where 3 + 1 doses PCV programs were implemented [13,14]. Other studies using different methodologies did not show substantial differences in the efficacy of 2 + 1 and 3 + 1 regimens to prevent IPD, lower respiratory tract infections, and otitis media in children [15–17]. The number of doses is a major driver of the acceptability and cost-effectiveness of immunization programs [18]. In view of this study, along with the accruing evidence on direct effectiveness of 2 + 1 schedule [9], PCV schedules relying on three doses instead of four should definitively be recommended worldwide. In recent years, all other Canadian provinces have moved from a 3 + 1 toward a 2 + 1 PCV schedule [19]. The overall impact of PCV programs is influenced by indirect effects including herd protection and serotype replacement. In Quebec, serotype replacement completely eroded the benefits of herd protection in adults. Different levels of replacement have been observed in different countries [9]. In the US, the magnitude of serotype replacement in children and adults was much lower than in Quebec [13]. It may be that the speed at which PCV-7 was introduced with active catch-up in Quebec could have played a role in creating an ecological vacuum that was rapidly occupied by non-vaccine clones with invasiveness potential [20]. Another contributing factor could have been the high proportion of children attending day-care services in Quebec (61% in 2000–2001), thanks to a very generous publicly funded program [21]. Replacement was also rapid and of high magnitude in the Native communities of Alaska [22]. It may be that the frequency and intimacy of contacts among children could influence the speed and magnitude of serotype replacement. This is the first experience analyzing the transition from PCV7 to PCV-10 for the routine immunization of children. Children exposed to the new vaccine experienced lower IPD rates and all serotypes contributed to the decline. Preliminary results published in a previous publication are thus confirmed [10]. Protection against 7F and some level of cross-protection against 19A were expected in the light of results of immunogenicity studies with PCV-10 showing the induction of functional antibodies [23]. At this time, there is no good explanation for the low rate of other serotypes not included in the vaccine. In the US (excluding Alaska), the introduction of PCV13 following PCV-7 use was associated with a decline in overall IPD rates among children, a marked decline for the six additional serotypes and a low level of replacement [24]. It will be interesting to analyze the transition from PCV-10 to PCV-13 in Quebec. There are limitations in the surveillance system for IPD in Quebec. Mandatory notification of IPD was enforced in 2000, and improvement in reporting can be suspected during the first year. For this reason, the year 2000 was excluded from statistical analyses. Another weakness is the poor quality of information regarding the clinical presentation and the immunization status of patients. For these reasons, a case–control study was initiated in 2005, and preliminary results confirmed the high effectiveness of the 2 + 1 schedule [25]. The direct protection provided by PCV can be measured in a case–control study, although ecological analyses are needed to capture indirect effects, including herd protection and replacement. Sentinel laboratories provided an unrepresentative albeit constant sample of total IPD cases diagnosed in the province. The population covered by this network cannot be defined precisely and no incidence rate could be computed. The quality of
1506
P. De Wals et al. / Vaccine 32 (2014) 1501–1506
laboratory surveillance for IPD in children was improved in 2005, when all microbiology laboratories in the province were invited to participate instead of a limited number of sentinel laboratories. In a record linkage study performed in 2006, the sensitivity of the notifiable disease registry was 94% and was 89% for the reference laboratory, while the concordance between the two sources was 95% [26]. There are plans to extend laboratory surveillance for IPD in adults to all microbiology laboratories in the province but budgetary constraint is an issue. In conclusion, a 2 + 1 PCV schedule with high uptake is highly effective to prevent IPD and should be recommended worldwide. Unfortunately, in Quebec, serotype replacement eroded indirect benefits especially in adults. PCV-10 introduction had a marked effect and the impact of PCV-13 use remains to be evaluated. Funding The study was supported by a research grant from the Quebec Ministry of Health and Social Services (Ministère de la Santé et des Services sociaux du Québec). Acknowledgments The authors thank all health professionals who are contributing to the surveillance of invasive pneumococcal disease in the province of Quebec. Conflict of interest statement: Philippe De Wals received research grants, honoraria and reimbursements of travel expenses from vaccine manufacturers including Glaxo-Smith-Kline, Novartis, Sanofi Pasteur, Merck and Wyeth, as well as from governmental agencies including the Quebec Ministry of Health and Social Services, Health Canada, and the Public Health Agency of Canada. The other authors have no conflict of interests. Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j.vaccine. 2013.11.028. References [1] Butler JC. Epidemiology of pneumococcal disease. In: Tuomanen EI, Mitchell TJ, Morrison DA, Spratt BG, editors. The pneumococcus. Washington, DC: ASM Press; 2004. p. 136–47. [2] National Advisory Committee on Immunization (NACI). Statement on recommended use of pneumococcal conjugate vaccine. CCDR 2002;28(ACS-2):1–32. [3] De Wals P. Optimizing the acceptability, effectiveness and costs of immunization programs: the Quebec experience. Expert Rev Vaccines 2011;10:55–62. [4] Quebec Immunization Committee. Assessment of the appropriateness of an immunization program for pneumococcal infections in children using a reduced number of doses of conjugate vaccine. Montreal, QC: Institut national de santé publique du Québec; 2005. Available at:
[7] Comité sur l’immunisation du Québec (CIQ). Évaluation de la pertinence d’un nouveau vaccin antipneumococcique conjugué au Québec [Evaluation of the appropriateness of a new pneumococcal conjugate vaccine in Quebec]. Quebec City, QC: Institut national de santé publique du Québec; 2009. Available at: