Immunogenicity and safety of an enterovirus 71 vaccine in healthy Chinese children and infants: a randomised, double-blind, placebo-controlled phase 2 clinical trial

Articles

Immunogenicity and safety of an enterovirus 71 vaccine in healthy Chinese children and infants: a randomised, double-blind, placebo-controlled phase 2 clinical trial Feng-Cai Zhu, Zheng-Lun Liang, Xiu-Ling Li, Heng-Ming Ge, Fan-Yue Meng, Qun-Ying Mao, Yun-Tao Zhang, Yue-Mei Hu, Zhen-Yu Zhang, Jing-Xin Li, Fan Gao, Qing-Hua Chen, Qi-Yan Zhu, Kai Chu, Xing Wu, Xin Yao, Hui-Jie Guo, Xiao-Qin Chen, Pei Liu, Yu-Ying Dong, Feng-Xiang Li, Xin-Liang Shen, Jun-Zhi Wang

Summary Background Enterovirus 71 (EV71) outbreaks are a socioeconomic burden, especially in the western Pacific region. Results of phase 1 clinical trials suggest an EV71 vaccine has a clinically acceptable safety profile and immunogenicity. We aimed to assess the best possible dose and formulation, immunogenicity, and safety profile of this EV71 vaccine in healthy Chinese children. Methods This randomised, double-blind, placebo-controlled, phase 2 trial was undertaken at one site in Donghai County, Jiangsu Province, China. Eligible participants were healthy boys or girls aged 6–36 months. Participants were randomly assigned (1:1:1:1:1) to receive either 160 U, 320 U, or 640 U alum-adjuvant EV71 vaccine, 640 U adjuvantfree EV71 vaccine, or a placebo (containing alum adjuvant only), according to a blocked randomisation list generated by SAS 9.1. Participants and investigators were masked to the assignment. The primary endpoint was anti-EV71 neutralising antibody geometric mean titres (GMTs) at day 56, analysed according to protocol. The study is registered with ClinicalTrials.gov, number NCT01399853. Findings We randomly assigned 1200 participants, 240 (120 aged 6–11 months [infants] and 120 aged 12–36 months [children]) of whom were assigned to each dose. 1106 participants completed the study and were included in the according-to-protocol analysis. The main reasons for dropout were withdrawal of consent and refusal to donate a blood sample. Infants who received the 640 U adjuvant vaccine had the highest GMTs on day 56 (742·2 [95% CI 577·3–954·3]), followed by those who received the 320 U formulation (497·9 [383·1–647·0]). For children, those who received the 320 U formulation had the highest GMTs on day 56 (1383·2 [1037·3–1844·5]). Participants who received the vaccine had significantly higher GMTs than did who received placebo (p<0·0001). For the subgroup of participants who were seronegative at baseline, both infants and children who received the 640 U adjuvant vaccine had the highest GMTs on day 56 (522·8 [403·9–676·6] in infants and 708·4 [524·1–957·6] in children), followed by those who received the 320 U adjuvant vaccine (358·2 [280·5–457·5] in infants and 498·0 [383·4–646·9] in children). 549 (45·8%) of 1200 participants (95 CI 42·9–48·6%) reported at least one injection-site or systemic adverse reaction, but the incidence of adverse reactions did not differ significantly between groups (p=0·36). The 640 U alum-adjuvant vaccine group had a significantly higher incidence of induration than did the 640 U adjuvant-free group (p=0·001). Interpretation Taking immunogenicity, safety, and production capacity into account, the 320 U alum-adjuvant formulation of the EV71 vaccine is probably the best possible formulation for phase 3 trials. Funding The National Science and Technology Major Project (2011ZX10004-902) of the Chinese Ministry of Science and Technology, China’s 12–5 National Major Infectious Disease Program (2012ZX10002-001), and Beijing Vigoo Biological.

Introduction Enterovirus 71 (EV71)—a member of the Picornaviridae family of viruses—was first isolated in California, USA, in 1969.1,2 Since then, EV71 has been associated with more than 14 outbreaks throughout the world,3 including some serious epidemics,4–10 in regions such as Australia, Bulgaria (44 deaths in 1975),4 mainland China (>27 deaths in 2007, 126 deaths in 2008, and >255 deaths in 2009),11,12 Hong Kong, Hungary (47 deaths in 1978),4 India, Japan, Korea, Malaysia (41 deaths in 1997),10 Taiwan (78 deaths in 1998),6,8 USA, and Vietnam. EV71 has become the major cause of enterovirus-associated infectious disease now that poliomyelitis has been eradicated by vaccination in

many regions of the world. EV71 infection causes a range of effects, from asymptomatic infection, to mild hand, foot, and mouth disease, severe complications with the CNS, and cardiopulmonary failure.13,14 Mortality rates are as high as 82–94% in severe cases.15 EV71 outbreaks have caused many casualties and are a large socioeconomic burden in endemic countries and regions,16 especially in the western Pacific region. The incidence of hand, foot, and mouth disease, particularly that caused by EV71 infection, seems to be increasing across this region—eg, a two-fold increase in the number of hand, foot, and mouth disease cases compared with the same period in 2006 was reported in Singapore in 2008.17 Development

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

Published Online January 24, 2013 http://dx.doi.org/10.1016/ S0140-6736(12)61764-4 Jiangsu Provincial Center for Disease Control and Prevention, Nanjing, Jiangsu Province, China (F-C Zhu MSc, F-Y Meng MSc, Y-M Hu BSc, J-X Li MSc, K Chu MSc); National Institute for Food and Drug Control, Beijing, China (Z-L Liang PhD, Q-Y Mao PhD, F Gao MSc, X Wu MSc, X Yao PhD, F-X Li PhD, J-Z Wang PhD); Beijing Vigoo Biological, Beijing, China (X-L Li MSc, Y-T Zhang PhD, Q-H Chen MSc, H-J Guo MSc, X-L Shen MSc); Donghai Country Center for Disease Control and Prevention, Liangyungang, Jiangsu Province, China (H-M Ge BSc, Z-Y Zhang BSc, Q-Y Zhu BSc, X-Q Chen BSc); and Department of Public Health, Southeast University, Nanjing, Jiangsu Province, China (Prof P Liu PhD, Y-Y Dong MSc) Correspondence to: Mr Xin-Liang Shen, Beijing Vigoo Biological, No 4, Sanjianfangnanli, ChaoYang District, Beijing 100024, China [email protected] or Dr Jun-Zhi Wang, National Institute for Food and Drug Control, No 2, Tiantanxili, Beijing 100050, China [email protected]

1

Articles

of vaccines against EV71 is a dynamic topic with paramount public health importance. Results of phase 1 clinical trials with an EV71 vaccine,18–20 completed in China in 2011, suggested that two doses of inactivated EV71 vaccine given on day 0 and day 28 had a clinically acceptable safety profile and good immunogenicity in healthy Chinese infants and children,18–20 justifying a phase 2 study with this vaccine. We aimed to identify the best possible dose and formulation of the EV71 vaccine, provide evidence for the immunogenicity of the vaccine, and assess the safety profile (including comparison between alumadjuvant and adjuvant-free vaccines) and antibody persistence after vaccination. Because studies have suggested there could be crossimmunity between EV71 and poliovirus,21,22 we also planned to explore crossimmunity between these viruses.

Methods Study design and participants

See Online for appendix

We recruited individuals for this randomised, doubleblind, placebo-controlled, single-site trial in Donghai County, Jiangsu Province, China, between August and September, 2011. Eligible participants were healthy boys or girls aged 6–36 months. Exclusion criteria were history of hand, foot, and mouth disease or previous vaccination with EV71 vaccine; acute febrile disease on the day of enrolment; acute infection in the previous week; receipt of any inactivated vaccines in the previous 2 weeks; receipt of any attenuated live vaccines or any other drugs under investigation in the previous month; receipt of blood products in the previous 3 months; treatment with cytotoxic or immunosuppressive drugs in the previous 6 months; history of haematological, neurological, bleeding, or autoimmune disease; immunodeficiency; malnutrition; congenital defects; and being unable to comply with the study schedule. Further details are outlined in the protocol (appendix pp 28–87). We obtained written informed consent from the guardians of all participants and approval from the institutional review board of the Jiangsu Provincial Center of Disease Control and Prevention (JPCDC) before initiation of the study. The trial was undertaken by the JPCDC in accordance with the Declaration of Helsinki and Good Clinical Practice.

Randomisation and masking We stratified eligible participants by age (6–11 months [infants] and 12–36 months [children]) and assigned each a sequential number according to their sequence of enrolment. Participants were then randomly allocated (1:1:1:1:1) to receive either placebo, EV71 vaccine of 160 U with adjuvant, 320 U with adjuvant, 640 U with adjuvant, or 640 U without adjuvant, according to a blocked (block size of 20) randomisation list generated by the statistical analysis program SAS (version 9.1). The four vaccine formulations and placebo had identical packaging and 2

were blindly labelled with the sequential numbers, which were the only identifiers. Individuals involved in randomisation and masking did not participate in any other process of the trial. All participants, their guardians, and investigators were masked to the treatment allocation.

Procedures Beijing Vigoo Biological (Beijing, China) developed the inactivated EV71 vaccine19,20 using EV71 strain FY7VP5/ AH/CHN/2008 (subgenotype C4) as the seed virus (GenBank accession number JX025561). The three alumadjuvant vaccine formulations contained 160 U, 320 U, and 640 U of EV71 antigen with 0·25 μg, 0·5 μg, and 1·0 μg of total protein, respectively, and the one adjuvantfree vaccine formulation contained 640 U of EV71 antigen with 1·0 μg of total protein. The placebo contained no EV71 antigen. EV71 antigen content was measured and calculated by the parallel-line method with China’s national reference standard for EV71 antigen (lot 20100701, 1600 U/mL).11 Each dose of alum-adjuvant vaccine or placebo contained 0·18 mg of alum. The immunisation schedule consisted of two doses given on day 0 and day 28. We injected intramuscularly into the deltoid region in children aged 12–36 months and into the anterolateral side of thigh in infants aged 6–11 months. We monitored participants for immediate adverse events for 30 min after each injection, and then instructed the participants’ guardians to record adverse events daily on diary cards for 28 days after each vaccination. We graded adverse events according to the scale issued by the Chinese State Food and Drug Administration.23 Participants visted the hospital eight times (on days 0, 3, 7, 28, 31, 35, and 56, and month 8), to receive the vaccine, give a blood sample, or undergo a safety assessment. An independent data and safety monitoring committee was set up to oversee the recording of serious adverse events (SAEs) and provide final decisions of causality between SAEs and vaccination before unblinding. The committee also supervised the updating of the safety data within 28 days after each injection. We made one protocol change—that of cancelling the 14 month follow-up—after the trial had begun, mostly because an EV71 epidemic occurred in Donghai County 10 months after vaccine administration, which would have interfered with the concentration of and potential decline in vaccine-elicited antibody. We collected blood samples from participants on day 0, day 28, day 56, and month 8 for EV71 neutralising antibody tests. The Chinese National Institute for Food and Drug Control measured neutralising antibody against EV71 using a modified cytopathogenic effect assay.11,24 The dilution ranged from 1:8 to 1:16384. The cutoff value for seropositivity was defined as 1:8. All serum samples were blindly assayed twice and the titration was double-checked. We did routine urine tests in all participants immediately before and 3 days after each injection, and

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

Articles

also did routine blood tests and blood biochemical tests in participants aged 12–36 months immediately before and 3 days after each injection. To assess antibody persistence, we analysed the data from participants from whom we obtained four consecutive blood samples at days 0, 28, and 56, and at month 8. To assess crossimmunity between EV71 and poliovirus (an exploratory analysis based on the protocol), we applied the standard poliovirus neutralisation assay recommended by WHO25 to detect antibody titres against polioviruses using serum samples on day 0, 28, and 56 from infants only. We assessed crossimmunity in only the younger age group because poliovirus vaccination is administered to infants aged 3, 4, and 5 months old, so we concluded that infants aged 6–11 months might be more affected by EV71 vaccinations if crossimmunity was present than would the older age group. We used Sabin strains 1, 2, and 3 to detect poliovirus antibody types 1, 2, and 3, respectively.

The primary endpoint for immunogenicity was antiEV71 neutralising antibody geometric mean titres (GMTs) at day 56 after first vaccination. Secondary endpoints for immunogenicity were seroconversion and geometric mean fold increase (GMFI).

Statistical analysis Sample size was calculated according to GMTs after vaccination, for all eligible participants and for the subgroup of participants who were seronegative at baseline, on the basis of results from a previous phase 1 study.20 Power Analysis and Sample Size (PASS) software (version 8.0) was used for the multiple comparisons of the treatment groups to estimate sample size. Assuming that 15% of participants would drop out and 60% would be seronegative at baseline, a sample size of 240 per treatment group (ie, 120 per age group within treatment group) was chosen, which would provide at least 80% power to detect a difference in log-10-transformed postvaccination GMTs of 2·3 in participants who were

1807 assessed for eligibility 607 excluded for not meeting inclusion criteria 1200 enrolled and stratified 600 aged 6–11 months 600 aged 12–36 months

240 assigned to and received first dose of 160 U alumadjuvant EV71 vaccine 120 aged 6–11 months 120 aged 12–36 months

240 assigned to and received first dose of 320 U alumadjuvant EV71 vaccine 120 aged 6–11 months 120 aged 12–36 months

240 assigned to and received first dose of 640 U alumadjuvant EV71 vaccine 120 aged 6–11 months 120 aged 12–36 months

240 assigned to and received first dose of 640 U adjuvant-free EV71 vaccine 120 aged 6–11 months 120 aged 12–36 months

240 assigned to and received first dose of placebo 120 aged 6–11 months 120 aged 12–36 months

6 discontinued 4 aged 6–11 months 2 aged 12–36 months

10 discontinued 6 aged 6–11 months 4 aged 12–36 months

15 discontinued 11 aged 6–11 months 4 aged 12–36 months

7 discontinued 4 aged 6–11 months 3 aged 12–36 months

12 discontinued 11 aged 6–11 months 1 aged 12–36 months

234 received second dose of 160 U with adjuvant 116 aged 6–11 months 118 aged 12–36 months

230 received second dose of 320 U with adjuvant 114 aged 6–11 months 116 aged 12–36 months

225 received second dose of 640 U with adjuvant 109 aged 6–11 months 116 aged 12–36 months

233 received second dose of 640 U without adjuvant 116 aged 6–11 months 117 aged 12–36 months

228 received second dose of placebo 109 aged 6–11 months 119 aged 12–36 months

12 discontinued 9 aged 6–11 months 3 aged 12–36 months

7 discontinued 3 aged 6–11 months 4 aged 12–36 months

8 discontinued 4 aged 6–11 months 4 aged 12–36 months

6 discontinued 5 aged 6–11 months 1 aged 12–36 months

11 discontinued 7 aged 6–11 months 4 aged 12–36 months

222 were included in immunogenicity analysis 107 aged 6–11 months 115 aged 12–36 months

223 were included in immunogenicity analysis 111 aged 6–11 months 112 aged 12–36 months

217 were included in immunogenicity analysis 105 aged 6–11 months 112 aged 12–36 months

227 were included in immunogenicity analysis 111 aged 6–11 months 116 aged 12–36 months

217 were included in immunogenicity analysis 102 aged 6–11 months 115 aged 12–36 months

Figure: Trial profile The reasons why 607 participants were excluded and did not undergo randomisation are listed in appendix p 1. The reasons why 94 of the 1200 participants were not included in the according-to-protocol cohort for immunogenicity analysis are listed in the appendix p 2.

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

3

Articles

seronegative at baseline, and 95% power to detect a difference of 2·0 or more in all eligible participants. Before the calculation of GMTs and GMFIs, antibody titres were log-transformed. Antibody titres below 1:8 were assigned values of 1:4 for calculation. In participants with a prevaccination titre of less than 1:8 160 U with adjuvant

320 U with adjuvant

640 U with adjuvant

640 U without adjuvant

Placebo

6–11 months (n=120) Age (months)

9·0 (1·7)

9·1 (1·7)

9·1 (1·7)

9·2 (1·8)

9·0 (1·8)

Boys

71 (59·2%)

72 (60·0%)

68 (56·7%)

72 (60·0%)

70 (58·3%)

Girls

49 (40·8%)

48 (40·0%)

52 (43·3%)

48 (40·0%)

50 (41·7%)

12–36 months (n=120) Age (months) 24·6 (6·8)

25·6 (6·9)

25·3 (7·4)

25·2 (6·5)

24·9 (7·1)

Boys

61 (50·8%)

69 (57·5%)

64 (53·3%)

67 (55·8%)

65 (54·2%)

Girls

59 (49·2%)

51 (42·5%)

56 (46·7%)

53 (44·2%)

55 (45·8%)

(ie, seronegative at baseline), seroconversion was defined as a postvaccination titre of 1:32 or greater, and in participants with a prevaccination titre of 1:8 or greater (ie, seropositive at baseline), it was defined as at least a four-fold increase after vaccination. Hypothesis testing was two-sided with an α value of 0·05. The χ² test or Fisher’s exact test was used to analyse categorical data, and multiple χ² comparisons were done on the basis of a Bonferoni-adjusted α value of 0·005 when relevant. Student’s t test or ANOVA was used to analyse continuous data and the StudentNewman-Keuls test for multiple comparisons was used if a significant difference between treatment groups was noted. Statistical analyses were done on an according-toprotocol basis by an independent statistician (PL) using SAS (version 9.1). The study is registered with ClinicalTrials.gov, number NCT01399853.

Data are mean (SD), and number (%).

Role of the funding source

Table 1: Baseline demographic characteristics of participants

One of the sponsors of the study (Beijing Vigoo Biological) was involved in trial design, but none of the

160 U with adjuvant

320 U with adjuvant

640 U with adjuvant

640 U without adjuvant

Placebo

111

102

p value

6–11 months n

107

111

105

Day 0 GMT

9·0 (6·4–12·7)

7·9 (5·8–10·8)

11·6 (7·9–17·2)

9·9 (7·0–14·0)

10·0 (7·0–14·3)

··

Day 28 GMT

21·2 (14·8–30·4)c

28·5 (20·4–39·7)bc

67·3 (47·5–95·4)a

43·3 (30·0–62·3)ab

Seroconversion

18 (16·8%, 10·3–25·3)

28 (25·2%, 17·5–34·4)

48 (45·7%, 36·0–55·7)

38 (34·2%, 25·5–43·8)

GMFI

2·4 (1·9–2·9)c

3·6 (2·9–4·5)bc

5·8 (4·7–7·2)a

10·7 (7·3–15·5)d

<0·0001

5 (4·9%, 1·6–11·1)

<0·0001

4·4 (3·4–5·7)ab

1·1 (0·9–1·2)d

<0·0001

210·3 (161·5–273·9)c

11·8 (7·9–17·6)d

Day 56 GMT

357·0 (270·1–471·9)b

Seroconversion

93 (86·9%, 79·0–92·7)

GMFI

39·6 (28·1–55·9)a

497·9 (383·1–647·0)ab

742·2 (577·3–954·3)a

105 (94·6%, 88·6–98·0)

100 (95·2%, 89·2–98·4)

62·8 (48·0–82·4)a

63·9 (46·0–88·9)a

<0·0001

93 (83·8%, 75·6–90·1)

5 (4·9%, 1·6–11·1)

<0·0001

21·2 (15·9–28·3)b

1·2 (1·0–1·5)c

<0·0001

12–36 months n

115*

112

112

116

115

Day 0 GMT

20·1 (13·3–30·4)

26·7 (17·6–40·6)

GMT

61·8 (41·0–93·1)b

139·0 (91·4–211·2)a

Seroconversion

33 (28·7%, 20·7–37·9)

15·0 (10·3–21·8)

23·9(15·5–36·8)

21·4 (14·3–32·1)

··

Day 28

GMFI

3·1 (2·5–3·7)b

99·6 (66·8–148·7)ab

53 (47·3%, 37·8–57·0) 5·2 (4·1–6·6)a

63 (56·3%, 46·6–65·6) 6·6 (5·4–8·2)a

113·4 (74·6–172·3)ab 45 (38·8%, 29·9–48·3) 4·8 (3·7–6·1)a

28·6 (18·6–44·2)c

<0·0001

8 (7·0%, 3·1–13·3)

<0·0001

1·3 (1·1–1·7)c

<0·0001

Day 56 GMT

731·4 (536·2–997·8)b

Seroconversion

103 (89·6%, 82·5–94·5)

GMFI

36·4 (26·3–50·2)b

1383·2 (1037·3–1844·5)a

1330·2 (1000·9–1767·7)a

107 (95·5%, 89·9–98·5)

109 (97·3%, 92·4–99·4)

51·7 (38·6–69·4)a

88·7 (65·7–119·7)a

509·0 (372·8–694·8)b 98 (84·5%, 76·6–90·5) 21·3 (16·1–28·4)c

37·0 (22·9–59·7)c

<0·0001

16 (13·9%, 8·2–21·6)

<0·0001

1·7 (1·3–2·3)d

<0·0001

Data are GMT (95% CI), number of participants (%, 95% CI), or GMFI (95% CI). n=total participants included in immunogenic according-to-protocol analysis. The p values are the result of comparison between the five treatment groups. Multicomparisons were applied based on an adjusted α=0·005 for seroconversion rate. In the following sentences, > implies that the preceding group(s) had a significantly greater seroconversion rate than the following group(s). 6–11 months on day 28: 320 U and 640 U adjuvant and 640 U adjuvant-free > placebo; 640 U adjuvant and 640 U adjuvant-free > 160 U adjuvant; 640U adjuvant > 320 U adjuvant. 6–11 months on day 56: all vaccine formulations > placebo. 12–36 months on day 28: all vaccine formulations > placebo; 320 U and 640 U adjuvant > 160 U adjuvant. 12–36 months on day 56: all vaccine formulations > placebo; 640 U adjuvant > 640 U adjuvant-free. The Student-Newman-Keuls test was used for multicomparison of GMTs and GMFIs. a, b, c, and d (where a>b>c>d) mark statistically significant differences between each treatment group. Measurements on day 0 were taken immediately before vaccination. GMT=geometric mean titre. GMFI=geometric mean fold increase. *One participant in this group had a prevaccination antibody titre of 1:8192 and a postvaccination titre of 1:16 384 (upper end of assay) so was classed as not having had seroconversion after vaccination because he had only had a two-fold increase; however, it is possible that his post-vaccination titre was higher than 1:16 384.

Table 2: GMT, seroconversion rate, and GMFI of neutralising antibody against EV71

4

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

Articles

sponsors were involved in data collection, statistical analysis, interpretation, or writing of the report. F-CZ, X-LS, and J-ZW had full access to all the data in the study and had final responsibility for the decision to submit for publication.

Results We enrolled 1200 eligible participants, all of whom received the first dose of EV71 vaccine or placebo and completed the 28 day follow-up (figure). 1150 of these participants received a second dose on day 28, and 1148 completed the 56 day follow-up to assess safety. We obtained serum samples from 1200 participants at day 0, 1147 at day 28, 1106 at day 56, and 966 at month 8. We obtained three consecutive serum samples at days 0, 28, and 56 for 536 (89·3%) of 600 infants and 570 (95·0%) of 600 children, who we included in the according-to-protocol immunogenic analysis. Reasons why some participants were not included in accordingto-protocol immunogenic analysis are listed in the appendix (p 2). Table 1 shows baseline demographic characteristics. 94 participants had failed to comply with the protocol for the primary objective by day 56. We noted no significant difference in terms of age and sex between the 64 infants who dropped out and the 536 who completed the study

160 U with adjuvant

(age, p=0·07; sex, p=0·66), or between the the 30 children who dropped out and the 570 children who completed the study (age, p=0·71; sex, p=0·91). The number of participants who were seropositive at baseline varied from 40 (33·3%) of 120 children in the 640 U alum-adjuvant vaccine group (95% CI 25·0–42·5) to 54 (45·0%) of 120 children in the 320 U alum-adjuvant vaccine group (35·9–54·4) and from 22 (18·3%) of 120 infants in the 320 U alum-adjuvant vaccine group (11·9–26·4) to 32 (26·7%) of 120 infants in the placebo group (19·0–35·5). GMTs at baseline were low and similar between treatment groups in each age group (table 2). Participants who received the EV71 vaccine had significantly higher GMTs, seroconversion rates, and GMFIs after vaccination than did those in the placebo group (p<0·0001; table 2). The 640 U adjuvant-free formulation elicited the lowest GMT at day 56 of the four vaccine formulations. Participants receiving alumadjuvant formulations had significantly higher GMTs at day 56 than did those who received the adjuvant-free vaccine (table 2). By comparison, the placebo could hardly induce an antibody response. Results for the subgroup of participants who were seronegative at baseline are shown in table 3. The first dose of EV71 vaccine elicited moderate immune responses in this subgroup as shown by GMTs, seroconversion

320 U with adjuvant

640 U with adjuvant

640 U without adjuvant

Placebo

p value

6–11 months n

85

93

80

85

78

GMT

10·5 (8·3-13·2)c

16·1 (12·9–20·1)b

30·4 (24·0-38·5)a

21·2 (16·2-27·8)b

4·4 (3·9–4·8)d

Seroconversion

13 (15·3%, 8·4–24·7)

22 (23·7%, 15·5–33·6)

38 (47·5%, 36·2–59·0)

28 (32·9%, 23·1–44·0)

1 (1·3%, 0·0–6·9)

<0·0001

1·1 (1·0–1·2)d

<0·0001

4·8 (3·8–6·0)d

<0·0001

Day 28

GMFI

2·6 (2·1–3·3)c

4·0 (3·2–5·0)b

7·6 (6·0–9·6)a

5·3 (4·0–7·0)b

<0·0001

Day 56 GMT

280·9 (207·4–380·6)b

358·2 (280·5–457·5)b

522·8 (403·9–676·6)a 79 (98·8%, 93·2–100·0)

141·0 (111·2–178·8)c

Seroconversion

80 (94·1%, 86·8–98·1)

91 (97·9%, 92·5–99·7)

GMFI

70·2 (51·8–95·2)b

89·6 (70·1–114·4)b

71

59

74

GMT

14·5 (11·5–18·4)b

24·3 (17·6–33·6)a

28·0 (21·8–35·8)a

22·5 (17·3–29·4)a

6·4 (4·4–9·3)c

Seroconversion

15 (21·1%, 12·3–32·4)

24 (40·7%, 28·1–54·3)

34 (46·0%, 34·3–57·9)

26 (37·7%, 26·3–50·2)

6 (8·6%, 3·2–17·7)

<0·0001

1·6 (1·1–2·3)c

<0·0001

7·3 (4·8–11·1)d

<0·0001

130·7 (101·0–169·2)a

79 (92·9%, 85·3–97·4)

2 (2·6%, 0·3–9·0)

<0·0001

35·3 (27·8–44·7)c

1·2 (1·0–1·5)d

<0·0001

12–36 months n

69

70

Day 28

GMFI

3·6 (2·9–4·6)b

6·1 (4·4–8·4)a

7·0 (5·5–8·9)a

5·6 (4·3-7·4)a

<0·0001

Day 56 GMT

372·7 (277·9–499·9)b

Seroconversion

71 (100·0%, 94·9–100·0)

GMFI

93·2 (69·5–125·0)b

498·0 (383·4–646·9)ab 59 (100·0%, 93·9–100·0) 124·5 (95·9–161·7)ab

708·4 (524·1–957·6)a 73 (98·7%, 92·7–100·0) 177·1 (131·0–239·4)a

171·9 (133·9–220·7)c 67 (97·1%, 89·9–99·7)

7 (10·0%, 4·1–19·5)

<0·0001

43·0 (33·5–55·2)c

1·8 (1·2–2·8)d

<0·0001

Data are GMT (95% CI), number of participants (%, 95% CI), or GMFI (95% CI). n=total participants with baseline antibody titre of <1:8 included in immunogenic according-to-protocol analysis. The p values are the result of comparison between the five treatment groups. Multicomparisons were applied based on an adjusted α=0·005 for seroconversion rate. In the following sentences, > implies that the preceding group(s) had a significantly greater seroconversion rate than the following group(s). 6–11 months on day 28: all vaccine formulations > placebo; 640 U adjuvant > 160 U and 320 U adjuvant. 6–11 months on day 56: all vaccine formulations > placebo. 12–36 months on day 28: 320 U and 640 U adjuvant and 640 U adjuvant-free > placebo; 640 U adjuvant > 160 U adjuvant. 12–36 months on day 56: all vaccine formulations > placebo. The Student-Newman-Keuls test was used for multicomparison of GMTs and GMFIs. a, b, c, and d (where a>b>c>d) mark statistically significant differences between each treatment group. Measurements on day 0 were taken immediately before vaccination. GMT=geometric mean titre. GMFI=geometric mean fold increase.

Table 3: GMT, seroconversion rate, and GMFI of neutralising antibody against EV71 in the subgroup of participants with baseline antibody titre of <1:8

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

5

Articles

160U with adjuvant

320U with adjuvant

640U with adjuvant

640U without adjuvant

Placebo

n (%)

n (%)

n (%)

n (%)

n (%)

95% CI

95% CI

95% CI

95% CI

p value 95% CI

6–11 months Grade 1

47 (39·2%) 30·4–48·5

40 (33·3%)

25·0–42·5

39 (32·5%)

Grade 2

21 (17·5%)

11·2–25·5

28 (23·3%)

16·1–31·9

25 (20·8%) 14·0–29·2

Grade 3

1 (0·8%)

0·0–4·6

1 (0·8%)

0·0–4·6

69 (57·5%)

48·2–66·5

Any

69 (57·5%) 48·2–66·5

0 64 (53·3%)

24·2–41·7 ·· 44·0–62·5

56 (46·7%) 37·5–56·0

46 (38·3%) 29·6–47·7

0·16

23 (19·2%)

12·6–27·4

26 (21·7%)

14·7–30·1

0·83

1 (0·8%)

0·0–4·6

1 (0·8%)

0·0–4·6

1·00

80 (66·7%) 57·5–75·0

73 (60·8%) 51·5–69·6 0·30

12–36 months Grade 1

23 (19·2%) 12·6–27·4

25 (20·8%) 14·0–29·2

31 (25·8%) 18·3–34·6

31 (25·8%) 18·3–34·6

28 (23·3%) 16·1–31·9

0·66

Grade 2

10 (8·3%)

12 (10·0%)

12 (10·0%)

12 (10·0%)

10 (8·3%)

0·98

Grade 3 Any

0

4·1–14·8 ··

33 (27·5%) 19·8–36·4

0

5·3–16·8 ··

37 (30·8%) 22·7–39·9

0 43 (35·8%)

5·3–16·8 ··

0

5·3–16·8 ··

27·3-45·1

43 (35·8%) 27·3–45·1

107 (44·6%) 38·2–51·1

123 (51·3%) 44·7–57·7

0 38 (31·7%)

4·1–14·8 ··

··

23·5–40·8 0·60

Both groups Any

102 (42·5%) 36·2–49·0

106 (44·2%) 37·8–50·7

111 (46·3%) 39·8–52·8 0·36

Data are number of participants (% of participants), 95% CI, or p value. Grade 1 events were mild (ie, no interference with activity). Grade 2 events were moderate (ie, some interference with activity). Grade 3 events were severe (ie, prevented activity).

Table 4: Incidence of adverse reactions within 28 days after vaccinations, by treatment and age groups

rates, and GMFIs, which were substantially boosted after the second dose. Participants who received the 640 U adjuvant formulation had the highest GMTs on day 56, followed by those who received the 320 U adjuvant formulation, and then those who received the 160 U adjuvant formulation (table 3). We also recorded a similar dose-dependent antibody response with respect to GMFIs, which suggests that the higher dose was more immunogenic. However, all four formulations elicited significantly higher immune response in terms of seroconversion than did placebo (tables 2, 3). To assess antibody persistence, we analysed data from 461 (76·8%) infants and 505 (84·2%) children from whom we obtained a fourth blood sample at month 8. The results showed that the GMTs had decreased to 86·3 (160 U adjuvant vaccine), 96·3 (320 U adjuvant vaccine), 177·4 (640 U adjuvant vaccine), and 58·8 (640 U adjuvantfree vaccine). However, the proportion of infants or children with an EV71 antibody titre of greater than 1:8 at month 8 remained almost the same as that on day 56 (appendix p 4). 549 (45·8%) of 1200 participants (95% CI 42·9–48·6) reported 904 injection-site or systemic adverse reactions by day 56. Incidences of reactions were not significantly different between groups (table 4). Adverse reactions were more frequent in infants than in children (p<0·0001). During 56 day follow-up, 80 (6·7%) of 1200 participants (95% CI 5·3–8·2) reported injection-site adverse reactions, and 515 (42·9%, 40·1–45·8) reported systemic adverse reactions. Incidence of systemic adverse reactions did not differ between treatment groups within each age group. The most common systemic adverse reaction across treatment groups was fever, with an incidence in infants ranging from 31 (25·8%, 95% CI 18·3–34·6) in the 640 U alum-adjuvant vaccine group to 47 (39·2%, 30·4–48·5) in 6

the 320 U alum-adjuvant vaccine group and an incidence in children ranging from 20 (16·7%, 10·5–24·6) in the 320 U alum-adjuvant vaccine group to 31 (25·8%, 18·3–34·6) in the 640 U alum-adjuvant vaccine group. The second most common adverse reaction was diarrhoea, with an incidence in infants ranging from 25 (20·8%, 14·0–29·2) in the 640 U alum-adjuvant vaccine group to 33 (27·5%, 19·8–36·4) in both the 320 U alum-adjuvant and 640 U adjuvant-free vaccine groups and an incidence in children ranging from 8 (6·7%, 2·9–12·7) in both the 160 U and 320 U alum-adjuvant vaccine groups to 14 (11·7%, 6·5–18·8) in the placebo group (appendix pp 5–8). Ten (4·4%) of 225 participants (95% CI 2·2–8·0%) who received two doses of the 640 U alum-adjuvant vaccine had injection-site adverse reactions after the second shot compared with one (0·4%) of 233 participants (0·0–2·4%) who received two doses of the 640 U adjuvant-free vaccine (appendix pp 9–10). Significantly more participants receiving alum-adjuvant vaccines or placebo had induration around the injection site than did those receiving the 640 U adjuvant-free vaccine (p=0·0191 for all participants, p=0·0410 for infants, p=0·44 for children; appendix pp 5–8). In particular, the 640 U alum-adjuvant vaccine was associated with a significantly higher incidence of induration than was the 640 U adjuvant-free formulation (p=0·001). Four infants reported grade 3 adverse reactions after vaccinations (appendix pp 11–20). Five SAEs were reported during the 56 day follow-up (appendix p 21) and resolved after hospital admission during this period. None was considered to be causally related to vaccination by the data and safety monitoring committee. One was EV71-associated hand, foot, and mouth disease, diagnosed in a participant who received placebo. Two participants withdrew from the second vaccination because of SAEs. No SAEs were fatal and we

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

Articles

detected no clinically relevant abnormality in routine blood tests, blood biochemical tests, and urinalysis after vaccinations (appendix pp 22–26). Of the 94 participants who were not included in the according-to-protocol cohort for immunogenicity analysis, 53 (56·4%) had adverse reactions, and of the 1106 who completed the study, 496 (44·9%) had adverse reactions (p=0·0311). However, we noted no significant difference in the dropout rates between treatment groups (appendix p 3). Consecutive serum samples from 480 infants on day 0, 28, and 56 were tested for antipoliovirus types 1, 2, and 3 antibody (appendix p 27). All these infants had previously received oral poliovirus vaccine, in line with the Chinese paediatric routine immunisation programme. The prevaccination antipoliovirus titres did not differ significantly between treatment groups (antipoliovirus type 1, p=0·1478; antipoliovirus type 2, p=0·0527; and antipoliovirus type 3, p=0·5686). Compared with concentrations measured on day 0, we noted a small but non-significant decrease in antipoliovirus antibody concentration on day 28 and day 56 in all treatment groups. GMFIs of type 1, 2, and 3 antipoliovirus antibody did not differ significantly between groups after the first dose (antipoliovirus type 1, p=0·9824; antipoliovirus type 2, p=0·1743; and antipoliovirus type 3, p=0·7016) or the second dose (p=0·8549, p=0·2875, p=0·5737, respectively).

Discussion All four vaccine formulations induced significantly greater neutralising antibody responses by day 28 (after the second dose was administered) than did placebo. We chose GMT of EV71 neutralising antibody as the primary endpoint of immunogenicity because neutralising antibody is protective against EV71 virus and therefore acts as a marker to suggest that the vaccine has induced a response.12 The results show a dose-dependent relation for the three alumadjuvant formulations in infants and children who were seronegative at baseline, with the 640 U adjuvant formulation showing higher immunogenicity than the 320 U and 160 U adjuvant formulations. The 640U alumadjuvant formulation elicited a stronger immune response than the 640 U adjuvant-free counterpart. We noted a similar trends in immune response between infants and children. However, the vaccine induced a higher immune response in children than in infants, irrespective of baseline serological status, possibly because children have a more mature immune system than do infants. Antibody titres had significantly declined by month 8 compared with that at day 56, suggesting a need for a booster shot. A similar decline in antibody titre was reported in participants after receiving the inactivated poliovirus vaccine.26–28 Generally, the vaccine had a good safety profile in all participants. Most reactions were mild or moderate and severe adverse reactions were uncommon. Incidence of adverse reactions within 28 days of both vaccinations were in line with previous phase 1 trials of this vaccine,19,20

and trials of other vaccines in children and infants (panel).29 Participants with systemic adverse reactions were evenly distributed between the vaccine and placebo groups. However, the addition of alum caused more injection-site adverse reactions than did the adjuvantfree formulation. Several factors contribute to the choice of best possible formulation, including immunogenicity, safety, and production capacity. According to the results, the 640 U adjuvant formulation seems to be more immunogenic than the 320 U formulation. However, GMTs and seroconversion rates did not differ significantly between these two formulations in infants and children on day 56, and both formulations elicited a robust immune response. The 640 U adjuvant formulation was associated with a higher incidence of injection-site reactions (induration) than were the other formulations. Although this difference in incidence was not statistically significant, this result might affect its acceptability. Considering the potentially huge demand for an EV71 vaccine in China, the 320 U adjuvant formulation will be able to supply a larger dose volume than the 640 U adjuvant formulation; therefore, we recommend the 320 U adjuvant formulation for phase 3 trials. Since children should receive routine oral poliovirus vaccine before they reach 6 months of age, crossimmunity—if any exists—could interfere with the response to EV71 vaccine given after 6 months of age. Clarification of whether crossimmunity took place would help to establish the age at which to start vaccination. We could not find evidence of crossimmunity, which is inconsistent with findings in animals and a study in Panel: Research in context Systematic review We searched PubMed with the terms “EV71 vaccine” or “enterovirus 71 vaccine” and did not apply publication date or language restrictions. We did our latest search on June 5, 2012. We found three phase 1 trials: one was an open-label trial19 in participants aged 5–22 years and the other two were randomised, parallel-group, controlled trials18,20 in children aged 6–60 months. The inactivated enterovirus 71 (EV71) vaccine was well tolerated and immunogenic as shown by neutralising antibody titres in adults, children, and infants,19,20 warranting further study. Interpretation This phase 2 trial not only provides more solid evidence for the immunogenicity and safety profile of the vaccines, including a comparison between alum-adjuvant and adjuvant-free vaccines, but also assessed vaccine-induced neutralising antibody over 8 months and crossimmunity against poliovirus types 1, 2, and 3. The alum-adjuvanted 320 U formulation is the best possible formulation for the phase 3 trial after taking immunogenicity, safety, and production capacity into account.

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

7

Articles

For more on the phase 3 trial see http://clinicaltrials.gov/ show/NCT01508247

Taiwan.16,30 Therefore, the interference with the immune response to EV71 vaccine due to poliovirus vaccinationinduced crossimmunity might be less of a problem than was previously thought. We did not assess crossimmunity between EV71 genotype C4 and other genotypes (ie, A, B4, B5, C2, and C5), mainly because genotype C4 is the only prevalent genotype to be found in mainland China since the first EV71-associated case was discovered in this country. Genotype C4 is also circulating in Japan, South Korea, Taiwan, Thailand, and Vietman. During recruitment, we excluded participants with a definite history of hand, foot, and mouth disease; however, we might have enrolled participants with subclinical EV71 infection or an EV71-associated infection not associated with hand, foot, and mouth disease. We did not do serological screening before enrolment to exclude seropositive participants because no rapid method to test for EV71 neutralising antibody exists (although we did take samples to determine seropositivity at baseline immediately before injection). Additionally, based on our experience of routine immunisation practice, pre-existing antibody tests are neither possible nor necessary before vaccination. To eliminate the effect of pre-existing antibody, therefore, we analysed a subset of participants in the according-to-protocol cohort who were seronegative at baseline, and came to the same conclusions as the original according-to-protocol analysis (table 3). We did not undertake active surveillance of hand, foot, and mouth disease or EV71-associated infections between day 56 and month 8. However, there was a small-scale EV71 epidemic during that time according to the National Infectious Disease Information Management System at the study site. This outbreak might explain why the GMT and seropositivity rate in the placebo group slightly increased at month 8 compared with at day 56, especially in infants. We recognise that this small-scale epidemic might have led to some underestimation of the decrease in antibody titres over time. Another limitation of this study was that it was a single-centre trial, which might limit the extent to which the results can be generalised. However, the aim of this phase 2 trial was to establish the optimum vaccine formulation that could be used for the phase 3 trial. As a result of our findings, we chose the adjuvant 320 U formulation as the candidate formulation for our phase 3 trial, a multicentre, randomised, double-blind, placebo-controlled trial (in progress) to assess the efficacy, immunogenicity, and safety of the 320 U adjuvant EV71 vaccine. Vaccine efficacy is the primary outcome. About 10 000 participants aged 6–35 months have been recruited, vaccinated, and followed up to assess efficacy and safety. Contributors F-CZ, J-ZW, and X-LS designed the trial and the study protocol, and contributed to the review and revision of the report. F-CZ was principal investigator of this trial. Z-LL and X-LL contributed to the study protocol and led the laboratory analyses. H-MG, Y-MH, Z-YZ, Q-YZ, KC, and

8

X-QC participated in the site work, including the recruitment, data collection, and data interpretation. F-YM, Y-TZ, and J-XL obtained the data, drafted the manuscript, supervised the study, and revised the report. Q-YM, FG, Q-HC, XW, XY, H-JG and F-XL contributed to the laboratory analyses, data interpretation, and literature search. PL contributed to the data analysis. J-XL also was involved in the quality control of the statistical analysis and creation of the tables and figure. Y-YD contributed to the data collection, literature search, and creation of the tables and figure. All authors reviewed and approved the final version of the report. Conflicts of interest X-LL, Y-TZ, H-JG, Q-HC, and X-LS are employees of Beijing Vigoo Biological. The other authors declare that they have no conflicts of interest. Acknowledgments We thank all the investigators from Donghai County Center for Disease Control and Prevention, who contributed to the site work of the trial. A special thanks goes to Yuan-Zheng Qiu for his help with the manuscript revisions. References 1 Qiu J. Enterovirus 71 infection: a new threat to global public health? Lancet Neurol 2008; 7: 868–69. 2 AbuBakar S, Chan YF, Lam SK. Outbreaks of enterovirus 71 infection. N Engl J Med 2000; 342: 355–56. 3 Weng KF, Chen LL, Huang PN, Shih SR. Neural pathogenesis of enterovirus 71 infection. Microbes Infect 2010; 12: 505–10. 4 Bible JM, Pantelidis P, Chan PK, Tong CY. Genetic evolution of enterovirus 71: epidemiological and pathological implications. Rev Med Virol 2007; 17: 371–79. 5 Cordey S, Petty TJ, Schibler M, et al. Identification of site-specific adaptations conferring increased neural cell tropism during human enterovirus 71 infection. PLoS Pathog 2012; 8: e1002826. 6 Ho M, Chen ER, Hsu KH, et al, for the Taiwan Enterovirus Epidemic Working Group. An epidemic of enterovirus 71 infection in Taiwan. N Engl J Med 1999; 341: 929–35. 7 Mizuta K, Abiko C, Murata T, et al. Frequent importation of enterovirus 71 from surrounding countries into the local community of Yamagata, Japan, between 1998 and 2003. J Clin Microbiol 2005; 43: 6171–75. 8 Ma E, Lam T, Chan KC, Wong C, Chuang SK. Changing epidemiology of hand, foot, and mouth disease in Hong Kong, 2001–2009. Jpn J Infect Dis 2010; 63: 422–26. 9 Sarma N, Sarkar A, Mukherjee A, Ghosh A, Dhar S, Malakar R. Epidemic of hand, foot and mouth disease in West Bengal, India in August, 2007: a multicentric study. Indian J Dermatol 2009; 54: 26–30. 10 Chua KB, Kasri AR. Hand foot and mouth disease due to enterovirus 71 in Malaysia. Virol Sin 2011; 26: 221–28. 11 Liang Z, Mao Q, Gao Q, et al. Establishing China’s national standards of antigen content and neutralizing antibody responses for evaluation of enterovirus 71 (EV71) vaccines. Vaccine 2011; 29: 9668–74. 12 Mao Q, Li N, Yu X, et al. Antigenicity, animal protective effect and genetic characteristics of candidate vaccine strains of enterovirus 71. Arch Virol 2012; 157: 37–41. 13 Solomon T, Lewthwaite P, Perera D, Cardosa MJ, McMinn P, Ooi MH. Virology, epidemiology, pathogenesis, and control of enterovirus 71. Lancet Infect Dis 2010; 10: 778–90. 14 Ooi MH, Wong SC, Lewthwaite P, Cardosa MJ, Solomon T. Clinical features, diagnosis, and management of enterovirus 71. Lancet Neurol 2010; 9: 1097–105. 15 Koroleva GA, Lukashev AN, Khudiakova LV, Mustafina AN, Lashkevich VA. Encephalomyelitis caused by enterovirus type 71 in children. Vopr Virusol 2010; 55: 4–10 (in Russian). 16 Huang WC, Huang LM, Kao CL, et al. Seroprevalence of enterovirus 71 and no evidence of crossprotection of enterovirus 71 antibody against the other enteroviruses in kindergarten children in Taipei city. J Microbiol Immunol Infect 2012; 45: 96–101. 17 WHO. A guide to clinical management and public health response for hand, foot and mouth disease (HFMD). Geneva, Switzerland: World Health Organization, 2011. http://www.wpro.who.int/ publications/docs/GuidancefortheclinicalmanagementofHFMD.pdf (accessed Sept 28, 2012).

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

Articles

18

19

20

21

22

23

24

Li YP, Liang ZL, Gao Q, et al. Safety and immunogenicity of a novel human enterovirus 71 (EV71) vaccine: a randomized, placebo-controlled, double-blind, phase I clinical trial. Vaccine 2012; 30: 3295–303. Meng FY, Li JX, Li XL, et al. Tolerability and immunogenicity of an inactivated enterovirus 71 vaccine in Chinese healthy adults and children: an open label, phase 1 clinical trial. Hum Vaccin Immunother 2012; 8: 668–74. Zhu FC, Wang JZ, Li XL, et al. Reactogenicity and immunogenicity of an enterovirus 71 vaccine in Chinese healthy children and infants. Pediatr Infect Dis J 2012; 31: 1158–65. Deng C, Yang C, Wan J, Zhu L, Leng Q. Irregular poliovirus vaccination correlates to pulmonary edema of hand, foot, and mouth disease. Clin Vaccine Immunol 2011; 18: 1589–90. Leng Q, Yang C, Zhu K, Deng C, Zhu L, Wan J. Pre-existing heterologous immunity to poliovirus vaccination may mitigate severity of hand, food and mouth disease caused by EV71. BMC Proc 2011; 5 (suppl 1): 27. State Food and Drug Administration. The standard guidelines for adverse reactions grading of vaccine clinical trials, Oct 14, 2005 (in Chinese). http://www.sda.gov.cn/WS01/CL0844/9350_5.html (accessed Oct 5, 2012). Mao Q-Y, Liao X-Y, Yu X, et al. Dynamic change of mother-source neutralizing antibodies against enterovirus 71 and coxsackievirus A16 in infants. Chin Med J (Engl) 2010; 123: 1679–84.

25

26

27 28

29

30

WHO. Polio laboratory manual, 4th edn, 2004. Geneva, Switzerland: World Health Organization, 2004. www.who.int/vaccines/en/poliolab/ WHO-Polio-Manual-9.pdf (accessed Oct 5, 2012). Asturias EJ, Dueger EL, Omer SB, et al. Randomized trial of inactivated and live polio vaccine schedules in Guatemalan infants. J Infect Dis 2007; 196: 692–98. Bonnet M-C, Dutta A. World wide experience with inactivated poliovirus vaccine. Vaccine 2008; 26: 4978–83. WHO Weekly Epidemiological Record. Polio vaccines and polio immunization in the pre-eradication era: WHO position paper. Geneva, Switzerland: World Health Organization, 2010. http://www. who.int/wer/2010/wer8523/en/ (accessed Oct 5, 2012). Gossger N, Snape MD, Yu LM, et al. Immunogenicity and tolerability of recombinant serogroup B meningococcal vaccine administered with or without routine infant vaccinations according to different immunization schedules: a randomized controlled trial. JAMA 2012; 307: 573–82. Chen CW, Lee YP, Wang YF, Yu CK. Formaldehyde-inactivated human enterovirus 71 vaccine is compatible for co-immunization with a commercial pentavalent vaccine. Vaccine 2011; 29: 2772–76.

www.thelancet.com Published online January 24, 2013 http://dx.doi.org/10.1016/S0140-6736(12)61764-4

9