Mutual Interference on the Immune Response to Yellow Fever Vaccine and A

Home , Cholera vaccine, Measles vaccine, MMR vaccine, Mumps vaccine, Rubella vaccine

Vaccine 29 (2011) 6327–6334

Contents lists available at ScienceDirect

Vaccine

jou rnal homepage: www.elsevier.com/locate/vaccine

Mutual interference on the immune response to yellow fever vaccine and a

combined vaccine against measles, mumps and rubella

a a,∗ b

Juliana Romualdo Nascimento Silva , Luiz Antonio B. Camacho , Marilda M. Siqueira ,

c d e f

Marcos de Silva Freire , Yvone P. Castro , Maria de Lourdes S. Maia , Anna Maya Y. Yamamura ,

g h 1

Reinaldo M. Martins , Maria de Luz F. Leal , Collaborative Group for the Study of Yellow Fever Vaccines

Departamento de Epidemiologia, Escola Nacional de Saúde Pública, FIOCRUZ, Rua Leopoldo Bulhões, 1480, sala 820, Manguinhos, Rio de Janeiro 21041-210, RJ, Brazil

Laboratório de Vírus Respiratórios e do Sarampo (LVRS), Instituto Oswaldo Cruz, FIOCRUZ, Pavilhão Helio e Peggy Pereira, sala B105, Rua Leopoldo Bulhões 1480, Manguinhos, Rio

de Janeiro 21041-210, RJ, Brazil

Vice-Diretoria de Desenvolvimento Tecnológico de Bio-Manguinhos, FIOCRUZ, Av. Brasil, 4365 – Manguinhos, Rio de Janeiro 21040-360, RJ, Brazil

Programa de Imunizac¸ ões da Secretaria Distrital de Saúde, SGAP, Lote 6, Bloco G, Parque de Apoio, SIA, Brasilia 71410-010, DF, Brazil

Assessoria Clínica, Bio-Manguinhos, FIOCRUZ, Av. Brasil, 4365 – Manguinhos, Rio de Janeiro 21040-360, RJ, Brazil

Laboratorio de Tecnologia Virológica, Bio-Manguinhos, FIOCRUZ, Av. Brasil, 4365 – Pavilhão Rocha Lima, Manguinhos, Rio de Janeiro 21040-360, RJ, Brazil

Consultor Cientíﬁco, Bio-Manguinhos, FIOCRUZ, Av. Brasil, 4365 – Manguinhos, Rio de Janeiro 21040-360, RJ, Brazil

Vice-Diretoria de Qualidade de Bio-Manguinhos, FIOCRUZ, Av. Brasil, 4365 – Manguinhos, Rio de Janeiro 21040-360, RJ, Brazil

a r t i c l e i n f o a b s t r a c t

Article history: A randomized trial was conducted to assess the immunogenicity and reactogenicity of yellow fever vac-

Received 21 November 2010

cines (YFV) given either simultaneously in separate injections, or 30 days or more after a combined

Received in revised form 5 May 2011

measles–mumps–rubella (MMR) vaccine. Volunteers were also randomized to YFV produced from 17DD

Accepted 9 May 2011

and WHO-17D-213 substrains. The study group comprised 1769 healthy 12-month-old children brought

Available online 2 June 2011

to health care centers in Brasilia for routine vaccination. The reactogenicity was of the type and frequency

expected for the vaccines and no severe adverse event was associated to either vaccine. Seroconversion

Keywords:

and seropositivity 30 days or more after vaccination against yellow fever was similar across groups

Yellow fever vaccine

deﬁned by YFV substrain. Subjects injected YFV and MMR simultaneously had lower seroconversion

Combined vaccine against measles

rates – 90% for rubella, 70% for yellow fever and 61% for mumps – compared with those vaccinated 30

Rubella and mumps

Immunogenicity days apart – 97% for rubella, 87% for yellow fever and 71% for mumps. Seroconversion rates for measles

Reactogenicity were higher than 98% in both comparison groups. Geometric mean titers for rubella and for yellow fever

Brazil were approximately three times higher among those who got the vaccines 30 days apart. For measles

and mumps antibodies GMTs were similar across groups. MMR’s interference in immune response of YFV

and YFV’s interference in immune response of rubella and mumps components of MMR had never been

reported before but are consistent with previous observations from other live vaccines. These results

may affect the recommendations regarding primary vaccination with yellow fever vaccine and MMR.

1. Introduction recommended between doses of different live attenuated vaccines

[2]. The Brazilian National Immunization Program (PNI) recom-

Simultaneous administration of vaccines in the same visit to mended against intervals shorter than 15 days between the yellow

a health service is recommended as a strategy to avoid the loss fever vaccine and other live attenuated vaccines for lack of infor-

of opportunities for vaccination [1]. A minimum of four weeks is mation regarding the interference between these antigens [3]. The

Advisory Committee on Immunization Practices (ACIP, Centers for

Disease Control and Prevention) recommends that injectable or

∗ nasally administered live vaccines be given on the same day or ≥4

Corresponding author. Tel.: +55 21 2598 2630; fax: +55 21 2270 6772.

weeks apart, to minimize the potential risk for interference [4]. The

E-mail addresses: [email protected] (J.R. Nascimento Silva),

luiz.camacho@ensp.ﬁocruz.br, [email protected] (L.A.B. Camacho), World Health Organization (WHO) strongly recommends the yel-

mmsiq@ioc.ﬁocruz.br (M.M. Siqueira), freire@bio.ﬁocruz.br (M.d.S. Freire),

low fever vaccine at nine months of age, at the same time of the

[email protected] (Y.P. Castro), lourdes.maia@bio.ﬁocruz.br (M.d.L.S. Maia),

measles vaccine in routine immunization in endemic areas [5].

anna@bio.ﬁocruz.br (A.M.Y. Yamamura), Rmenezes@bio.ﬁocruz.br (R.M. Martins),

The high immunogenicity of substrains 17DD yellow fever vac-

Malu@bio.ﬁocruz.br (M.d.L.F. Leal).

For full description, see Appendix A. cine was conﬁrmed in recent studies in adults and children over

6328 J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334

2 years of age [6,7]. A study with children of 9 months showed was produced from the seed lot 993FB013Z (4.70 log10 PFU/0.5-

no interference when measles and yellow fever vaccines were mL), whereas the 17D substrain vaccine (lot 04UVFAEX34 with

administered simultaneously [8]. In contrast, a multicenter study 4.91 log10 PFU/0.5-mL) was produced from the seed batch of the

in children aged 6–23 months showed a rate of seroconversion World Health Organization (WHO 17D-213/77). Children were

and geometric mean titers (GMTs) signiﬁcantly lower than those given the type of vaccine against yellow fever to which they were

of adults. The data suggested that simultaneous vaccination against randomly assigned. Study subjects were also randomized to receive

yellow fever and measles could interfere with the immune response YFV (either substrain) simultaneously or 30 days after MMR (Fig. 1),

against yellow fever (at that time a monovalent measles vaccine by means of computer generated random numbers, printed and

was administered at 9 months of age) [6]. In Brazil and other coun- placed in opaque envelopes, sealed and numbered. After signing the

tries the measles vaccine is currently used in combination with the consent form the envelopes were opened in the order of presenta-

vaccine against rubella and mumps. There are no published data on tion of the volunteers. Randomization used permutation blocks of

the interference of the yellow fever vaccine (YFV) and the rubella size 6, ratio of 1:1. The codes were opened after statistical analysis.

and mumps vaccines [9]. Each vial of vaccine was used in only one participant. The MMR vac-

A study was conducted to assess the differences in immuno- cine was administered according to routine immunization services,

genicity and reactogenicity (1) between yellow fever vaccines (YFV) without interference from the study.

produced with WHO 17D-213/77 and 17DD substrains, and (2)

between YFV (either substrain) injected simultaneously or after 30

2.3. Sample size

days administering a combined vaccine against measles, rubella

and mumps (MMR) [10]. The postulated effects of MMR on the

The number of participants was calculated using the following

response to YFV could not be distinguished for each one of MMR

parameters: beta = 0.2, alpha = 0.05 (two-tailed test), 90% serocon-

components, but the reciprocal was veriﬁed. For conciseness, this

version in one group (p1), and minimum difference between the

paper highlighted the results for yellow fever and rubella, as elim-

groups (p1 − p2) of 5 percentage points [11]. The sample size with

ination of rubella and congenital rubella syndrome may require

a 20% correction for loss of follow up was 1740 children, 870 in each

vaccination in the age range in which the yellow fever vaccine

comparison group.

is recommended in many countries. Moreover, the interaction of

measles vaccines and YFV had been reported in previous studies.

Results for measles and mumps are presented brieﬂy. 2.4. Data collection and outcomes of interest

2. Subjects and methods A questionnaire was administered before vaccination with

items on age, sex, birth weight and weight at vaccination, immu-

This was a randomized study whose methods were described nization history and history of allergies to food and drugs. We asked

previously [10] and will be presented brieﬂy below. Comparison of the children’s parents to record daily, in a diary, during the 10 days

YFV produced with WHO 17D-213/77 and 17DD substrains was after the vaccination, the adverse events expected for the yellow

double-blinded, whereas the comparison between YFV injected fever vaccine (fever, vomiting, pain and redness at the injection site

simultaneously or 30 days after MMR was unblinded. and irritability) and any health problems observed in that period.

The clinical events occurring after this period were recorded on a

2.1. Study population and eligibility criteria postvaccination questionnaire.

Samples of 4 mL of blood were collected on the day of MMR

Fieldwork was conducted from February to July 2006 in nineteen vaccination and 30 days after yellow fever vaccination to titrate

public health centers from Federal District, the only Brazilian State antibodies against yellow fever, rubella, measles and mumps. Thus,

where routine yellow fever vaccine and MMR vaccine were given subgroups deﬁned by the interval between the vaccines also dif-

simultaneously. Children aged 12–23 months who presented for fered in the interval between post-vaccination blood collection and

routine vaccination were invited to participate. The exclusion cri- MMR: 30 days in those who received the vaccines on the same day

teria for the study were based on contraindications for yellow fever and 60 days in those who received YFV 30 days after.

vaccination [3]: severe malnutrition, immunosuppression, admin- The titration of antibodies against yellow fever and the anti-

istration of immunoglobulin or other blood products within 60 bodies against measles was performed at Virologic Technology

days before or after vaccination, hypersensitivity to gelatin or egg Laboratory of Bio-Manguinhos (LATEV, FIOCRUZ, Rio de Janeiro)

◦

chicken and derivatives, fever of 37.5 C or more. Children were not with Plaque Reduction Neutralization Test (PRNT).

included if obstacles to return for vaccination against yellow fever PRNT was conducted in serial twofold dilutions starting at 1:5,

◦

␮

or post-vaccination blood collection were anticipated. in 50 L aliquots of heat inactivated (at 56 C for 30 min) serum,

in 96-well tissue culture plates. A positive monkey serum sample

2.2. Vaccines, randomization and blinding with yellow fever antibody content calibrated by a WHO Interna-

tional Reference Preparation, with 1115 mIU/mL was the standard

Regardless of their participation in the study, children received serum for each set of tests [12]. For measles the standard serum

the MMR vaccine available for routine immunization in health care contained 3000 mUI/mL [13]. The log10 dilution of the test sera and

units. At the time of this ﬁeld study, there were two MMR vaccines the standard serum, which reduced the plaque numbers by 50% rel-

available: MMRI , MSD (measles strain Moraten; mumps strain ative to the virus control, was determined by linear interpolation.

Jeryl Lynn; rubella strain Wistar 27/3) and vacina combinada con- To convert reciprocal dilutions into mIU/mL a unitage constant was

tra rubéola, sarampo e caxumba , Bio-Manguinhos/GSK (measles calculated for each assay run, dividing the antibody concentration

strain Schwarz; mumps strain RIT 4385; rubella strain Wistar RA in the standard serum by the reciprocal dilution of the standard

27/3). serum in that assay run. Results of test sera in reciprocal dilution

Study subjects received a 0.5 mL dose of yellow fever vaccine were multiplied by the unitage constant to transform into mIU/mL,

(YFV) from one of the two sub-strains, injected subcutaneously in which improves comparability of results across assay runs [14,15].

the deltoid region. YF vaccines were put in identical vials labeled The primary endpoints of the study were antibody titers to

with codes generated by a statistician and disclosed only to the yellow fever in mIU/mL and categories (seropositive: titer higher

staff who conducted the labeling. The 17DD substrain vaccine than 2.7 log10 mIU/mL or reciprocal dilution higher than 10).

J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334 6329

1943 children recruited,

received MMR

115 excluded because of problems in labeling of

ser um samples in one health care unit.

1828 children randomized:

906 assign ed to simultaneous YFV and MMR

922 assign ed to YFV 30 days or more after MMR

59 withdrew their consent to receive the YFV in the study

22 missed blood collectio n (withdrawal, change of address

etc.)

post-vacc ine serum test unavailable (insufficient or

inappro priate specimens) for:

YF: 24; rubella: 43; measles: 54; mumps: 56

1694 children with data on measles, mumps and

rubella antibodies

1723 childr en with data on yellow fever

Simult aneous YFV YFV 30 days or more Simultaneous YFV YFV 30 days or more

and MMR (N=839) after MMR (N=852) and MMR (N=864) after MMR (N=859)

17D-2 13: n= 426 17D-213: n= 411 17D-213: n= 444 17D-213: n= 411

17DD: n= 413 17DD: n= 441 17DD: n= 420 17DD: n= 448

Fig. 1. Flow diagram of a randomized trial of yellow fever vaccines (YFV) injected simultaneously or 30 days after a combined measles-mumps-rubella (MMR) vaccine.

This is a description of the assignment of volunteers to allocation groups deﬁned by vaccine type and by time interval between yellow fever vaccine and a combined

measles-mumps-rubella vaccine.

Seroconversion was deﬁned as quadrupling of pre-vaccination anti- fever was evaluated by comparing the proportions of seroconver-

bodies against yellow fever. sion for yellow fever in individuals in subgroups deﬁned by the

Serologic testing for rubella antibodies (ELISA, Enzygnost interval between vaccinations. Children without post-vaccination

Anti-Rubella-Virus/IgG, Dade Behring, Germany) and for mumps serological test, or who violated eligibility criteria were disregarded

antibodies (ELISA, Enzygnost Anti-Parotitis-Virus/IgG, Dade in “per-protocol analysis”. With this approach, analysis of immune

Behring, Germany) were performed at the Respiratory Virus Lab- response considered the vaccine actually administered, regardless

oratory of Instituto Oswaldo Cruz (FIOCRUZ, Rio de Janeiro), and of randomization group.

the results expressed in International Units per milliliter of serum The probability of seroconversion was adjusted for the covari-

(IU/mL). The primary endpoints for rubella were post-vaccination ates of interest (age, sex, pre-vaccination seropositivity, time

antibody titers in IU/mL and categories (non-reactive: <4.0 IU/mL; between pre- and post-vaccination blood collection, and comorbid-

inconclusive: 4.0–6.5 IU/mL; reactive: >6.5 IU/mL). For mumps, sera ity) in a logistic regression model. Similarly, the logarithm (base 10)

with antibody titers ≥231 U/mL were considered reactive, implying of post-vaccination antibody titers was modeled by multiple lin-

that borderline titers were considered seropositive. Both for rubella ear regression for the variable types of vaccines, sex, age, antibody

and for mumps, seroconversion was deﬁned as seropositivity in titers pre-vaccination, time between blood collection, pre- and

subjects who were non-reactive before vaccination. post-vaccination and comorbidity. The stepwise entry of variables

in the model and continuation in the ﬁnal model were determined

2.5. Statistical analysis by their relevance and statistical signiﬁcance (p < 0.20 and p < 0.05,

respectively).

The proportion of seroconversion, the geometric mean titer

(GMT) and proportion of adverse events after vaccination were 2.6. Ethical considerations

compared across groups deﬁned by types of yellow fever vac-

cine and interval between vaccinations. The statistical signiﬁcance This study was approved by the Ethics in Research Committees

of differences in proportions was analyzed by chi-square test, of the National School of Public Health-Fiocruz (document 236A/03

whereas for the differences in the means of antibody titer CEP-FIOCRUZ), and the Ministry of Health of the Federal District

logarithms the Student’s t test was used. Reverse cumulative dis- (Document SES-DF CEP-069/2005) authorized by ANVISA and reg-

tribution plots were constructed to display the complete range of istered in the International Standard Randomised Controlled Trial

serologic data. The level of signiﬁcance was 5%. Data were analyzed Number Register (ISRCTN 72367932).

using SPSS version 13.0 (SPSS, Inc., Chicago, IL).

The complete cohort (“intention-to-treat”) for analysis of 3. Results

adverse events included children with data on reactogenicity, even

those who failed to adhere to the study protocol. For the analysis of From a total of 1943 children, 115 in one health center

immunogenicity, the cohort consisted of all subjects randomized to were disregarded in the analysis because of inconsistencies in

YFV types, keeping subjects in the groups to which they were ran- identiﬁcation numbers of blood samples. All the remaining 1828

domly assigned. The interaction of the MMR vaccine and yellow children received the MMR (Bio-Manguinhos/GSK, 48.5%, Merck,

6330 J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334

Table 1

Baseline characteristics of study subjects according to type (virus substrain) of yellow fever vaccine (YFV) and to whether YFV was given simultaneously or 30 days after

combined vaccine against measles, rubella and mumps (MMR).

Characteristic Type of YFV Interval between vaccines

Simultaneous YFV and MMR YFV 30 days after MMR Total

Study subjects, n (%) Unvaccinated 23 (1.3) 36 (2.0) 59 (3.2)

17D-213 452 (24.7) 428 (23.4) 880 (48.1)

17DD 432 (23.6) 457 (25.0) 889 (48.6)

Total 906 (49.6) 922 (50.4) 1828 (100.0)

Male, n (%) Unvaccinated 14 (0.8) 15 (0.8) 29 (1.6)

17D-213 245 (13.3) 246 (13.5) 491 (26.9)

17DD 243 (13.3) 240 (13.1) 483 (26.4)

Total 502 (27.4) 501 (27.5) 1003 (54.9)

Age in months, mean (SD) Unvaccinated 12.3 (0.9) 12.1 (0.9) 12.2 (0.9)

17D-213 12.2 (0.9) 12.2 (0.8) 12.2 (0.9)

17DD 12.2 (1.1) 12.2 (0.9) 12.2 (1.0)

Total 12.2 (1.0) 12.2 (0.9) 12.2 (0.9)

Birth weight (g), mean (SD) Unvaccinated 3189 (548) 3269 (546) 3238 (543)

17D-213 3193 (471) 3176 (499) 3185 (484)

17DD 3193 (499) 3177 (492) 3185 (495)

Total 3193 (486) 3180 (497) 3186 (491)

Weight at vaccination (g), mean Unvaccinated 9926 (1135) 9496 (1307) 9651 (1252)

(SD) 17D-213 9540 (1203) 9715 (1234) 9627 (1221)

17DD 9632 (1217) 9580 (1247) 9605 (1232)

Total 9593 (1209) 9640 (1244) 9617 (1226)

Seropositive for yellow fever, Unvaccinated 0 (0.0) 5 (0.3) 5 (0.3)

before YFV, n (%) 17D-213 45 (2.5) 36 (2.0) 81 (4.5)

17DD 42 (2.3) 55 (3.0) 97 (5.3)

Total 87 (4.8) 96 (5.3) 183 (10.1)

Seropositive for rubella, before Unvaccinated – 1 (0.1) 1 (0.1)

MMR, n (%) 17D-213 10 (0.5) 12 (0.7) 22 (1.2)

17DD 9 (0.5) 6 (0.3) 15 (0.8)

Total 19 (1.0) 19 (1.1) 38 (2.1)

Seropositive for measles, before Unvaccinated – – –

MMR, n (%) 17D-213 5 (0.3) – 5 (0.3)

17DD 2 (0.1) 3 (0.2) 5 (0.3)

Total 7 (0.4) 3 (0.2) 10 (0.6)

Seropositive for mumps, before Unvaccinated – – –

MMR, n (%) 17D-213 2 (0.1) 1 (0.0) 3 (0.1)

17DD 3 (0.2) 5 (0.3) 8 (0.5)

Total 5 (0.3) 6 (0.3) 11 (0.6)

35.6%, not recorded, 15.9%) and 59 (3.2%) did not receive yellow deﬁned by the types of yellow fever vaccines showed no signiﬁcant

fever vaccine in the study. In the intention-to-treat analysis, we differences in immune response (p > 0.5, Table 2 and Fig. 2).

included the 1769 children who received yellow fever vaccine, and In the logistic model for seroconversion only the interval

were thus randomly assigned to one type of YFV. Among those, between vaccines showed a statistically signiﬁcant association

22 (1.2%) did not return for blood sampling after vaccination. Of (OR = 3.80, 95% CI: 2.39–6.05). In multivariate linear regression

those who returned, 43 (2.5%), 54 (3.1%), 56 (3.2%) and 24 (1.4%) did model the explanatory variables for the logarithm of antibodies

not have post-vaccination serological status for rubella, measles, against rubella were interval between administration of YFV and

mumps and yellow fever, respectively (Fig. 1). The total loss was MMR (p < 0.001), gender (p < 0.001), and logarithm of time between

13.5% and included subjects who did not return for vaccination or blood collection and MMR (p < 0.001).

blood collection, or whose specimens were lost or were insufﬁcient The rates of seroconversion for measles were 98.2% in the

to perform the serological tests. These losses were not selec- group with simultaneous YFV and MMR, and 99.2% among those

tive regarding study groups. Six children assigned to vaccination who received YFV 30 days or more after MMR (p = 0.090). GMTs

with an interval of 30 days received the vaccines simultaneously, were 3.44 IU/mL (95% CI: 3.20–3.70 IU/mL) and 3.19 IU/mL (95% CI:

whereas in 5 children the opposite occurred. The 59 volunteers 3.00–3.39 IU/mL), respectively. The seroconversion and GMTs were

lost between the two randomization procedures were similar to similar across groups who got different substrains of YFV: 98.9%

those volunteers randomized to the vaccine against yellow fever, seroconversion and GMT of 3.35 IU/mL (95% CI: 3.13–3.58 IU/mL)

according to gender, age weight, and the proportion seropositive in children in the 17D-213 group; 98.4% seroconversion and GMT

for rubella and yellow fever (Table 1). The base-line characteristics equal to 3.28 IU/mL (95% CI: 3.07–3.51 IU/mL) in the 17-DD group

were well-balanced across comparison groups (Table 1). The pro- (p = 0.521).

portion of children seropositive to yellow fever before vaccination The rates of seroconversion for mumps were 61.1% in the group

was substantially higher than for measles, mumps and rubella. with simultaneous YFV and MMR, and 70.8% among those who

received YFV 30 days or more after MMR (p < 0.001). GMTs were

3.1. Immune response to the rubella component of MMR 335.5 mIU/mL (95% CI: 314.4–358.0 mIU/mL) and 414.1 mIU/mL

(95% CI: 388.0–442.1 mIU/mL), respectively. The seroconversion

The proportion of seroconversion and magnitude of immune and GMT were similar across groups who got different sub-

response (GMT and distribution of postvaccination antibody titers) strains of YFV: 67.0% seroconversion and GMT of 384.7 mIU/mL

for rubella were substantially higher in the group in which YFV (95% CI: 359.9–411.2 mIU/mL) in children in the 17D-213 group;

and MMR were given 30 days apart, compared to those vaccinated 65.2% seroconversion and GMT equal to 362.6 mIU/mL (95% CI:

simultaneously (p < 0.001, Table 2 and Fig. 2). In contrast, the groups 340.0–386.7 mIU/mL) in the 17-DD group (p = 0.497).

J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334 6331

Table 2

Proportion of seroconversion and geometric mean titer (GMT) of antibodies against rubella after combined vaccine against measles, rubella and mumps (MMR) according

to type (virus substrain) of yellow fever vaccine (YFV) and to whether YFV was given simultaneously or 30 days after MMR – complete cohort (n = 1769) and subgroup of

protocol adherers (n = 1686).

Analysis Type of YFV Interval between vaccines

Simultaneous YFV and MMR YFV 30 days after MMR

Seroconversion (%) (95% CI) Complete 17D-213 86.3 (82.8–89.3) 94.9 (92.3–96.8)

Cohort 17DD 86.1 (82.5–89.2) 94.3 (91.7–96.3)

Total 86.2 (83.8–89.0) 94.4 (92.6–95.6)

Protocol 17D-213 90.7 (87.5–93.2) 98.0 (96.2–99.1)

Adherers 17DD 89.8 (86.4–92.5) 96.4 (94.2–97.9)

Total 90.2 (88.0–92.2) 97.2 (95.8–98.2)

GMT (IU/mL) (95% CI) Complete 17D-213 24.1 (22.0–26.4) 62.2 (57.2–67.5)

Cohort 17DD 25.6 (23.3–28.1) 58.2 (53.2–63.7)

Total 24.8 (23.2–26.5) 60.1 (56.5–63.9)

Protocol 17D-213 24.3 (22.1–26.6) 61.5 (56.6–66.9)

Adherers 17DD 25.5 (23.2–28.0) 58.4 (53.3–63.9)

Total 24.9 (23.3–26.6) 59.9 (56.3–63.7)

Fig. 2. Reverse cumulative distribution of the logarithm (base 10) of rubella and yellow fever antibody titers after vaccination, according to type (virus substrain) of yellow

fever vaccine (YFV) and time interval between the administration YFV and combined vaccine against measles-mumps-rubella (MMR) (complete cohort). These ﬁgures provide

a complete representation of the performance of the vaccines across comparison groups in the whole range of values of antibody titers.

6332 J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334

Reverse cumulative distribution curves for antibody titers after dent of the substrain of the vaccine against yellow fever and time

MMR, support the ﬁnding of similar immunogenicity across groups of blood collection for serology after vaccination.

deﬁned by YFV substrains, and groups in which YFV and MMR were The rate of seroconversion to rubella in the group vaccinated

given either simultaneously or 30 days apart (data not shown). For 30 days or more apart was consistent with that observed in other

mumps, the curves were also consistent with the small difference studies with MMR vaccines [17–19] but the lower magnitude of the

in the GMT shown above. response to the rubella and mumps components of MMR in children

For each of the three components, the proportions of serocon- vaccinated simultaneously against yellow fever is unprecedented

version, did not differ substantially in children who received MMR in the literature. Signiﬁcant reduction in the response to yellow

vaccine from different producers, whereas GMTs were slightly fever vaccine in children had been observed after administra-

higher among those who received the MSD vaccine (data not tion of combined vaccine against smallpox and measles [20], and

shown). simultaneous vaccination against cholera [21,22] and hepatitis B

[23]. Other studies have not found evidence of interference of

YFV simultaneous to or combined with vaccines against smallpox

3.2. Immune response to yellow fever vaccine

and diphtheria–tetanus–pertussis [24], measles [8,24–28], hep-

atitis A [29,30], hepatitis B [23,31,32], typhoid fever [33] and

The proportion of seroconversion and magnitude of immune

poliomyelitis [32]. Although immunologically plausible, the appar-

response (GMT and distribution of postvaccination antibody titers)

ent interference in the adaptive immunity to other vaccines is not

were greater in the group vaccinated with an interval of 30

well understood and may involve the very complex innate immu-

days compared to simultaneous vaccination (p < 0.001, Table 3

nity mechanisms that can limit the initial contact with a virus and,

and Fig. 2). In contrast, the groups deﬁned by the types of yel-

therefore, affect the magnitude of the adaptive immune response

low fever vaccines showed no signiﬁcant difference in immune

[34–37]. A major public health implication of this interference is the

response (p > 0.5, Table 2 and Fig. 2). The logistic model (data not

accumulation of a cohort of susceptible subjects long before the age

shown) showed a strong association of seroconversion (OR = 4.53,

recommended for revaccination. That could possibly affect disease

95% CI: 3.12–6.57) and post-vaccination seropositivity (OR = 7.60,

control, particularly of yellow fever, for which revaccination every

95% CI: 5.06–11.40) with the interval between administration of

10 years is recommended, disregarding the age at primovaccination

YFV and MMR, adjusted for the interval between blood collection

and the weaker immune response in infants [6,20,25].

and vaccination with MMR. In multivariate linear model (data not

Similar to the results of this study, a multicenter study with

shown) log10 post-vaccination antibody titers against yellow fever

the 17DD substrain vaccine showed 88% seroconversion in children

were strongly correlated to the interval between YFV and MMR

aged 12–23 months and 72% in infants aged 9–11 months [6]. At

(p < 0.001), adjusted for the time interval between blood collection

the time that study was conducted the monovalent measles vaccine

and MMR vaccine (p < 0.001). The interaction term for the type of

was administered at 9 months of age.

vaccine and the interval between YFV and MMR was not statistically

signiﬁcant. As the study outcomes relied on serological tests the implica-

tions of their accuracy results should be considered brieﬂy. PRNT is

the “gold standard” for detection of measles [38] and of yellow fever

3.3. Adverse events

antibodies [39], but for yellow fever it lacks the level of standard-

ization of measles PRNT [15] and commercial tests such as those

Adverse events were reported in 23% of the children and had

used for rubella and mumps antibodies. The proportion of reduc-

low or moderate severity: fever (14.2%), vomiting (1.9%), irritabil-

tion in number of plaques with neutralization, for instance, might

ity (3.3%), pain (2.8%) and redness (1.5%) at the injection site. The

affect the sensitivity of the PRNT and partly explain the variability

proportion of adverse events was higher in the group vaccinated

of GMT across studies. Nevertheless, seroconversion may be a more

simultaneously, but this difference was statistically signiﬁcant only

robust measure of response to immunization. The pre-vaccination

for fever (16.6% for simultaneous vaccination, 11.8% for vaccination

seropositivity for yellow fever could indicate false positive results

with 30-day interval, p = 0.01) and for any signs/symptoms (27.3%

of the PRNT, previous vaccination held outside the Federal District

for simultaneous vaccination and 18.8% for vaccination with 30-

where the vaccine is given at 9 months of age, and to maternal

day interval, p = 0.02). The differences in reactogenicity according

antibodies [40]. Randomization safeguarded against selection bias

to YFV types were small and not statistically signiﬁcant (p > 0.05).

but misclassiﬁcation of seropositivity might have weakened the

Local events (pain and redness on the injection site) occurred

differences between groups. However, with seroconversion as the

earlier (1–2 days) than the systemic events (fever, vomiting and

outcome we focused on the variation in the immune status after

irritability) (4–6 days). Adverse events in the group vaccinated

vaccination and made allowance for pre-vaccination seropositivity.

simultaneously with MMR and YFV did not differ in average time

Systemic adverse events following simultaneous administration

of onset of signs/symptoms (p > 0.09). The duration of signs and

of the vaccines had a similar pattern observed after the vaccine

symptoms was on average 2–3 days, with median of 1–2 days. The

against yellow fever was given separately. The time of onset of

difference between groups deﬁned by interval between vaccines

symptoms, as well as its duration is compatible with events related

was small and not statistically signiﬁcant (p > 0.10).

to vaccination, but there is no way to distinguish them from signs

and symptoms of coincidental clinical conditions. The frequency

4. Discussion of adverse events following immunization was consistent with

that reported in the literature for yellow fever [41], except for

The expanding arsenal of vaccines given in the ﬁrst two years fever, which was about twice as high. The higher frequency of

of life has been accompanied by extensive research on the possi- signs/symptoms was a plausible outcome of simultaneous vacci-

bilities and limitations of combined and simultaneous application nation with four viral agents.

of live attenuated vaccines [16]. This study demonstrated that con- Neutralizing antibodies against rubella are generally consid-

comitant administration (in separate syringes) of a yellow fever ered the protective immune response [42]. As titration of these

≥

vaccine and a combined vaccine against measles, rubella and antibodies is not routinely available, antibody levels ( 15 IU)

mumps induced lower seroconversion rates and GMT compared measured by other methods are considered to protect against

to the immune response to the same vaccines given 30 days apart. rubella infection [43]. The reduction of immunogenicity of rubella

The reduction in the magnitude of immune response was indepen- vaccine when given simultaneously to yellow fever vaccine can

J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334 6333

Table 3

Proportion of seroconversion and geometric mean titer (GMT) of antibodies against yellow fever after YFV according to type (virus substrain) of yellow fever vaccine (YFV)

and time interval between YFV and combined vaccine against measles, rubella and mumps (MMR) – complete cohort (n = 1769) and subgroup of protocol adherers (n = 1648).

Analysis Type of YFV Interval between vaccines

Simultaneous YFV 30 days after MMR

Seroconversion (%) (95% CI) Complete 17D-213 67.3 (62.7–71.6) 80.8 (76.8–84.5)

Cohort 17DD 65.7 (61.0–70.2) 84.7 (81.0–87.9)

Total 66.5 (63.3–69.6) 82.8 (80.2–85.2)

Protocol 17D-213 69.4 (64.9–73.7) 86.5 (82.8–89.7)

Adherers 17DD 70.0 (65.2–74.4) 88.8 (85.4–91.2)

Total 69.7 (66.4–72.8) 87.7 (85.3–89.8)

GMT (mIU/mL) (95% CI) Complete 17D-213 1019.4 (909.0–1143.2) 3115.8 (2770.8–3503.7)

Cohort 17DD 1105.9 (974.0–1255.6) 3652.5 (3226.5–4134.6)

Total 1060.2 (1015.1–1154.6) 3381.3 (3236.8–3683.7)

Protocol 17D-213 1023.4 (909.6–1151.5) 3106.4 (2760.9–3495.2)

Adherers 17DD 1109.5 (976.1–1261.2) 3664.4 (3233.8–4152.3)

Total 1064.6 (976.0–1161.2) 3385.2 (3105.2–3690.4)

lead to a greater proportion of vaccinees with lower titers and, Contributors: LABC, MSF, MLFL, MLSM participated in the con-

consequently, increased risk of reinfection by wild viruses [44]. ception and design of the study; LABC, YPC and MLSM participated

The reduction of 7 percentage points in seroconversion to rubella, in acquisition of data; LABC, JRNS, AMYY, MSF, MMS participated

when MMR and YFV were given simultaneously, is signiﬁcant from in the analysis and interpretation of data; JRNS and LABC pre-

immunological and public health standpoints. In a cohort of 500 pared the draft of the article. All authors revised the article and

girls vaccinated at age 12 observed for 16 years [45] seropositiv- gave ﬁnal approval of the version to be submitted. Conﬂict of inter-

ity decreased from 100% to 94% and the GMT declined from 1:110 est statement: L.A.B. Camacho and M.M. Siqueira are researchers in

to 1:18. In a context of low circulation of wild virus, it is possi- FIOCRUZ and collaborate in several research projects sponsored by

ble that children with lower titers after vaccination may become Bio-Manguinhos, the manufacturer of the yellow fever vaccines.

susceptible before revaccination. M.S. Freire, M.L.S. Maia, A.M.Y. Yamamura, R.M. Martins and M.L.F.

The seroconversion rate for mumps in this study is within the Leal are employees of Bio-Manguinhos. All authors have approved

range reported before for vaccines of Jeryl Lynn strain [46]. The poor the ﬁnal article. Funding: National Immunization Program, Min-

immune response to the mumps component of MMR of two major istry of Health; Fundac¸ ão Oswaldo Cruz-FIOCRUZ; CNPq (Brazilian

manufacturers, contrasted with optimal performance for measles National Research Council); Local and State Health Departments.

and rubella shown above. A thorough review of the laboratory

methods, and tests with the vaccine in a controlled setting did not

Appendix A. The members of the Collaborative Group for

disclose major problems. Nevertheless, MMR in routine immuniza-

Studies with Yellow Fever Vaccine are:

tion rather than in research settings could be more vulnerable to

cold chain breach and operational errors, and possibly explain vac-

Steering Committee: Luiz Antonio Bastos Camacho (Principal

cination failures. None of those factors seemed to account for the

Investigator), Marcos da Silva Freire, Maria da Luz Fernandes

differences in immunogenicity between randomized groups.

Leal, Maria de Lourdes Souza Maia and Reinaldo Menezes Mar-

tins (FIOCRUZ). Collaborating centers and investigators: Anna Maia

Yoshida Yamamura (Laboratory of Vırus Technology-FIOCRUZ),

5. Conclusions

Helena Keico Sato (Sao Paulo State Health Secretary), Guilherme

Cortes Fernandes (Santa Casa de Misericordia de Juiz de Fora, MG),

Although vaccination against measles, mumps and rubella and

Ivone Perez de Castro (Health Secretary of Brasilia, DF), Jandira

yellow fever in general do not coincide in the basic immunization

Campos Lemos (Health Secretary of Minas Gerais), and Eugenio

calendar, the simultaneous application to avoid loss of opportunity

Martins Barros (Health Secretary of Campo Grande, MS). Collabora-

may be needed in areas of difﬁcult access and when travel to areas

tors: Takumi Igushi (FIOCRUZ), Maristela Batista (Health Secretary

where yellow fever vaccine is required. The results of this study

of Juiz de Fora, MG), Maria da Conceicao Barros (Health Secre-

indicate the need to revise the guidelines for simultaneous vac-

tary of Campo Grande, MS) and Elisabete Paganini Sao Paulo (State

cination with the vaccines against yellow fever vaccine and MMR.

Health Secretary), Marileide Nascimento (FIOCRUZ) e Nilce da Silva

Postponing the yellow fever vaccine could be considered taking into

(FIOCRUZ). Data Monitoring Committee: Meri Baran (Health Secre-

account the epidemiological context. Revaccination against those

tary of Rio de Janeiro), Eduardo Pernambuco (University of Rio de

agents in shorter period than currently proposed could be recom-

Janeiro) and Antonio Jose Leal Costa (Federal University of Rio de

mended when the risk of disease and poor access did not allow an

Janeiro).

interval of more than 30 days between vaccinations.

These conclusions apply to primary vaccination in children less

than two years old. As primary vaccination against yellow fever References

in older children and adults, and a booster dose at any age induce

[1] Decker MD, Edwards KM. Combination vaccines. In: Plotkin SA, Orenstein WA,

stronger immune response, interference from other live virus vac-

editors. Vaccines. third ed. Philadelphia: W.B. Saunders Company; 1999.

cines should be less pronounced and possibly irrelevant. [2] Centers for Disease Control and Prevention. General recommendations on

immunization: recommendations of the Advisory Committee on Immuniza-

tions Practices (ACIP). MMWR 2002;51(RR-2):1–35.

[3] Ministério da Saúde. Manual de Procedimentos para Vacinac¸ ão. In: Programa de

Acknowledgements

Imunizac¸ ões, editor. Fundac¸ ão Nacional de Saúde. fourth ed. Brasília: Ministério

da Saúde; 2001.

[4] Centers for Disease Control and Prevention. Yellow fever vaccine: recommen-

We thank the parents and guardians of the infants for their coop-

dations of the Advisory Committee on Immunization Practices (ACIP). MMWR

eration. We are also grateful for the invaluable collaboration of 2010;59(RR-7).

many research assistants in health care centers and laboratories. [5] World Health Organization. Fact sheet No. 100; 2001.

6334 J.R. Nascimento Silva et al. / Vaccine 29 (2011) 6327–6334

[6] Grupo Colaborativo do Programa Nacional de Imunizac¸ ões para o Estudo da [26] Soula G, Sylla A, Pichard E, Kodio B, Bentejac MC, Teulières L, et al. Etude d’un

Soroconversão pela Vacina contra Febre Amarela. Estudo multicêntrico de soro- noveau vaccin combiné contre la ﬁevre jaune et la rougeol chez des enfants

conversão pela vacina contra febre amarela. In: CSCS, editor. VII Congresso agés de 6 a 24 mois au Mali. Bull Soc Pathol Exot 1991;84(5):885–97.

Brasileiro de Saúde Coletiva. 2003. [27] Mouchon D, Pignon D, Vicens R, Tu-Ha-Thanh, Tekaia F, Teulières L, et al. Étude

[7] Camacho LAB, Freire MS, Leal MLF, Aguiar SG, Nascimento JP, Iguchi T, Omoto de la Vaccination Combinée Rougeole-Fièvre Jaune Chez L’Énfant Africaine Agé

TM, Mann G, Freire MS, et al. Immunogenicity of WHO-17D and Brazilian 17DD de 6 a 10 Mois. Bull Soc Pathol Exot 1990;83:537–51.

yellow fever vaccines: a randomized trial. Rev Saúde Pública 2004;38(5):671–8. [28] Lhuillier M, Mazzariol MJ, Zadi S, Le Cam N, Bentejac MC, Adamowicz L, et al.

[8] Stefano I, Sato HK, Pannutti CS, Omoto TM, Mann G, Freire MS, et al. Recent Study of combined vaccination against yellow fever and measles in infants from

immunization against measles does not interfere with the sero-response of six to nine months. J Biol Stand 1989;17:9–15.

yellow fever vaccine. Vaccine 1999;17:1042–6. [29] Gil A, Gonzalez A, Dal-Re R, Calero JR. Interference assessment of yellow fever

[9] Monath TP, Cetron MS, Teuwen DE. Yellow fever. In: Plotkin SA, Orenstein vaccine with the immune response to a single-dose inactivated hepatitis A vac-

WA, Ofﬁt PA, editors. Vaccines. fourth ed. Philadelphia: W.B. Saunders Elsevier; cine (1440 EL.U.). A controlled study in adults. Vaccine 1996;14(11):1028–30.

2008. [30] Dumas R, Forrat R, Lang J, Farinelli T, Loutan L. Safety and immunogenicity

[10] Collaborative Group for Studies with Yellow Fever Vaccine.Randomized, of a new inactivated hepatitis A vaccine and concurrent administration with

double-blind, multicenter study of the immunogenicity and reactogenicity of a typhoid fever vaccine or a typhoid fever + yellow fever vaccine. Adv Ther

17DD and WHO 17D-213/77 yellow fever vaccines in children: implications for 1997;14:160–7.

the Brazilian National Immunization Program. Vaccine 2007;25:3118–23. [31] Coursaget P, Bringer L, Bourdil C, Yvonnet B, Chiron JP, Sarr G, et al. Simulta-

[11] Fleiss J, editor. Statistical methods for rates and proportions. second ed. New neous injection of hepatitis B and measles vaccines. Trans R Soc Trop Med Hyg

York: John Wiley & Sons; 1981. 1991;85(6):788.

[12] National Institute for Biological Standards and Control. WHO reference [32] Coursaget P, Fritzell B, Blondeau C, Saliou P, Diop-Mar I. Simultaneous injection

reagent. The 1st international reference preparation for anti-yellow fever of plasma-derived or recombinant hepatitis B vaccines with yellow fever and

serum monkey NIBSC code: YF. Available from: http://www.nibsc.ac.uk/ killed polio vaccines. Vaccine 1995;13(1):109–11.

documents/ifu/YF.pdf; 2008 [cited 15.03.2011]. [33] Ambrosch F, Fritzell B, Gregor J, Jonas S, Kollaritsch H, Teulieres L, et al. Com-

[13] WHO Expert Committee on Biological Standardization. Report of a collabora- bined vaccination against yellow fever and typhoid fever: a comparative trial.

tive study to assess the suitability of a replacement for the 2nd International Vaccine 1994;12(7):625–8.

Standard for anti-measles serum. Geneva: World Health Organization; Octo- [34] Reinhardt B, Jaspert R, Niedrig M, Kostner C, L’age-Stehr J. Development of

ber 23–27, 2006. Report No.: WHO/BS/06.2031. http://whqlibdoc.who.int/ viremia and humoral and cellular parameters of immune activation after vacci-

hq/2006/WHO BS 06.2031 eng.pdf. nation with yellow fever virus strain 17D: a model of human ﬂavivirus infection.

[14] Cohen BJ, Parry RP, Doblas D, Samuel D, Warrener L, Andrews N, et al. Measles J Med Virol 1998;56(2):159–67.

immunity testing: comparison of two measles IgG ELISAs with plaque reduction [35] Wheelock EF, Sibley WA. Circulating virus, interferon and antibody after

neutralisation assay. J Virol Methods 2006;131(2):209–12. vaccination with the 17-D strain of yellow fever virus. N Engl J Med

[15] Cohen BJ, Audet S, Andrews N, Beeler J, WHO working group on measles plaque 1965;273(4):194–8.

reduction neutralization test. Plaque reduction neutralization test for measles [36] Petralli JK, Merigan TC, Wilbur JR. Action of endogenous interferon against

antibodies: description of standardized laboratory method for use in immuno- vaccinia infection in children. Lancet 1965;2:401–5.

genicity studies of aerosol vaccination. Vaccine 2007;26:59–66. [37] Umino Y, Saito S, Fukuda A, Hishiyama M, Sugiura A. Improvement in

[16] Gizurarson S. Clinically relevant vaccine–vaccine interactions. A guide for prac- potency assay of measles–mumps–rubella trivalent vaccine: interference

titioners. BioDrugs 1998;9(June (6)):443–53. between components and measures for its elimination. J Virol Methods

[17] Rojo CC, Iglesias MR, Olvera J, Girón MA. Study of the immune response engen- 1990;27(2):159–68.

dered by different combined measles, mumps and rubella (MMR) vaccines in [38] Ratnam S, Gadag V, West T, Burris J, Oates E, Stead F, et al. Comparison of

an area of Andalusia (Spain). Vaccine 2003;22:280–6. commercial enzyme immunoassay kits with plaque reduction neutralization

[18] Klinge J, Korn LS, Heininger K, Stehr UK. Comparison of immunogenicity test for detection of measles antibody. J Clin Microbiol 1995;33(4):811–5.

and reactogenicity of a measles, mumps and rubella (MMR) vaccine in Ger- [39] Niedrig M, Lademann M, Emmerich P, Lafrenz M. Assessment of IgG antibodies

man children vaccinated at 9–11, 12–14 or 15–17 months of age. Vaccine against yellow fever virus after vaccination with 17D by different assays: neu-

2000;18:3134–40. tralization test, haemagglutination inhibition test, immunoﬂuorescence assay

[19] Tischer A, Gerike E. Immune response after primary and re-vaccination and ELISA. Trop Med Int Health 1999;4(12):867–71.

with different combined vaccines against measles, mumps, rubella. Vaccine [40] Suzano CE, Amaral E, Sato HK, Papaiordanou PM, Campinas Group on Yellow

2000;18:1382–92. Fever Immunization during Pregnancy. The effects of yellow fever immuniza-

[20] Meyer Jr HM, Hostetler Jr DD, Bernhein BC, Rogers NG, Lambin P, Chassary A, tion (17DD) inadvertently used in early pregnancy during a mass campaign in

et al. Response of Volta children to jet inoculation of combined live measles, Brazil. Vaccine 2006;24(9):1421–6.

smallpox and yellow fever vaccines. Bull World Health Organ 1964;30:783–94. [41] Camacho LAB, Freire MS, Leal MLF, Aguiar SG, Nascimento JP, Iguchi T, et al.

[21] Felsenfeld O, Wolf RH, Gyr K, Grant LS, Dutta NK. Simultaneous vaccination Reactogenicity of yellow fever vaccines in a randomized, placebo-controlled.

against cholera and yellow fever. Lancet 1973;I:457–8. Rev Saúde Pública 2005;39(3):413–20.

[22] Gateff C, Dodin A, Wiart J. A comparison of the serological effects of classical [42] Plotkin SA. Immunologic correlates of protection induced by vaccination. Pedi-

cholera vaccine and of puriﬁed fraction, with or without simultaneous yellow atr Infect Dis J 2001;20(1):63–75.

fever vaccine. Ann Microbiol (Paris) 1975;126(2):231–46. [43] Matter L, Kogelschatz K, Germann D. Serum levels of rubella virus antibodies

[23] Yvonnet B, Coursaget P, Deubel V, Diop-Mar I, Digoutte JP, Chiron JP, et al. indicating immunity: response to vaccination of subjects with low or unde-

Simultaneous administration of hepatitis B and yellow fever vaccines. J Med tectable antibody concentrations. J Infect Dis 1997;175(4):749–55.

Virol 1986;19:307–11. [44] Atkinson W, Kroger A, Pickering LK. General immunization practices. In: Plotkin

[24] Ruben FL, Smith EA, Foster SO. Simultaneous administration of smallpox, S, Orenstein W, Ofﬁt P, editors. Vaccines. ﬁfth ed. Philadelphia: Saunders Else-

measles, yellow fever, and diptheria–pertussis–tetanus antigens to Nigerian vier; 2008.

children. Bull World Health Organ 1973;48:175–81. [45] Christenson B, Bottiger M. Long-term follow-up study of rubella antibodies in

[25] Adu FD, Omotade OO, Oyedele OI, Ikusika O, Odemuyiwa SO, Onoja AL. Field naturally immune and vaccinated young adults. Vaccine 1994;12:41–5.

trial of combined yellow fever and measles vaccines among children in Nigeria. [46] Plotkin SA, Rubin SA. Mumps vaccine. In: Plotkin SA, Orenstein WA, Ofﬁt PA,

East Afr Med J 1996;73(9):579–82. editors. Vaccines. Philadelphia: W.B. Saunders Elsevier; 2008.