Immunogenicity of Fractional-Dose Vaccine During A Yellow Fever Outbreak - Preliminary Report

The n e w e ng l a n d j o u r na l of m e dic i n e
Original Article
Immunogenicity of Fractional-Dose Vaccine

during a Yellow Fever Outbreak
— Preliminary Report
Steve Ahuka‑Mundeke, M.D., Ph.D., Rebecca M. Casey, M.B., B.S., M.P.H.,
Jennifer B. Harris, Ph.D., M.P.H., Meredith G. Dixon, M.D.,
Pierre M. Nsele, M.D., Gabriel M. Kizito, M.D., Grace Umutesi, M.P.H.,
Janeen Laven, B.S., Gilson Paluku, M.D., M.P.H., Abdou S. Gueye, M.D., Ph.D.,
Terri B. Hyde, M.D., M.P.H., Guylain K.M. Sheria, M.D., Ph.D.,
Jean‑Jacques Muyembe‑Tanfum, M.D., Ph.D., and J. Erin Staples, M.D., Ph.D.
A BS T R AC T
BACKGROUND
In 2016, the response to a yellow fever outbreak in Angola and the Democratic From the Institut National de Recherche
Republic of Congo led to a global shortage of yellow fever vaccine. As a result, a Biomédicale (S.A.-M., P.M.N., G.M.K.,
J.-J.M.-T.), Division of Global Health Pro-
fractional dose of the 17DD yellow fever vaccine (containing one fifth [0.1 ml] of tection, Centers for Disease Control and
the standard dose) was offered to 7.6 million children 2 years of age or older and Prevention (CDC) (A.S.G.), and Pro-
nonpregnant adults in a preemptive campaign in Kinshasa. The goal of this study gramme Elargi de Vaccination, Ministre
de la Santé (G.K.M.S.) — all in Kinshasa,
was to assess the immune response to the fractional dose in a large-scale cam- Democratic Republic of Congo; Global
paign. Immunization Division (R.M.C., J.B.H.,
M.G.D., G.U., G.P., T.B.H.) and the Epi-
METHODS demic Intelligence Service (R.M.C.), CDC,
Atlanta; and Division of Vector-Borne
We recruited participants in four age strata at six vaccination sites. We assessed Diseases, CDC, Fort Collins, CO (J.L.,
neutralizing antibody titers against yellow fever virus in blood samples obtained J.E.S.). Address reprint requests to Dr.
before vaccination and 28 to 35 days after vaccination, using a plaque reduction Casey at the Epidemic Intelligence Ser-
vice, Global Immunization Division,
neutralization test with a 50% cutoff (PRNT50). Participants with a PRNT50 titer of 1600 Clifton Rd., Atlanta, GA 30329, or at
10 or higher at baseline were considered to be seropositive. Those with a baseline rcasey@cdc.gov.
titer of less than 10 who became seropositive at follow-up were classified as having This article was published on February 14,
undergone seroconversion. Participants who were seropositive at baseline and who 2018, at NEJM.org.
had an increase in the titer by a factor of 4 or more at follow-up were classified as DOI: 10.1056/NEJMoa1710430
having an immune response. Copyright © 2018 Massachusetts Medical Society.
RESULTS
Among 716 participants who completed follow-up, 705 (98%; 95% confidence in-
terval [CI], 97 to 99) were seropositive after vaccination. Among 493 participants
who were seronegative at baseline, 482 (98%; 95% CI, 96 to 99) underwent sero-
conversion. Among 223 participants who were seropositive at baseline, 148 (66%;
95% CI, 60 to 72) had an immune response. Lower baseline titers were associated
with a higher probability of having an immune response (P<0.001).
CONCLUSIONS
A fractional dose of the 17DD yellow fever vaccine was effective at inducing sero-
conversion in most of the participants who were seronegative at baseline. These
findings support the use of fractional-dose vaccination for outbreak control.
(Funded by the U.S. Agency for International Development and the Centers for
Disease Control and Prevention.)
n engl j med nejm.org 1
The New England Journal of Medicine
Downloaded from nejm.org on March 5, 2018. For personal use only. No other uses without permission.
Copyright © 2018 Massachusetts Medical Society. All rights reserved.
Y
ellow fever is a mosquito-borne campaign using a fractional dose of 17DD vac-
viral disease endemic to tropical and sub- cine (Bio-Manguinhos) at one fifth (0.1 ml) of the
tropical regions in Africa and the Americas. standard dose in all nonpregnant adults and in
Infection with yellow fever virus can result in children 2 years of age or older in August 2016.
subclinical to severe illness, characterized by We evaluated the immunologic response to this
fever, jaundice, and hemorrhage. There were fractional-dose vaccine delivered in a large-scale
an estimated 51,000 to 380,000 severe cases of vaccination campaign.
yellow fever and 19,000 to 180,000 deaths in
Africa in 2013.1 Treatment is managed to address Me thods
patients’ symptoms. However, the administration
of a highly effective vaccine is the primary meth- Study Participants and Design
od for prevention and control. All currently used From August 17 through August 26, 2016, the
yellow fever vaccines are live attenuated viral campaign was conducted at 2404 vaccination
vaccines derived from the 17D strain.2,3 Nearly sites in Kinshasa.13 We selected 6 vaccination sites
all studies have shown that one dose induces across the three geographic sectors of Kinshasa
seroconversion in more than 98% of recipients, on the basis of economically diverse catchment
and protection is believed to be lifelong.2,4,5 populations and logistic feasibility. Participants
In December 2015, a large yellow fever out- who presented for vaccination at one of these
break began in Angola and spread to the neigh- sites were approached for potential inclusion in a
boring Democratic Republic of Congo (DRC). convenience sample for the study, with an equal
The outbreak resulted in 962 confirmed cases and number of participants from four age strata: 2 to
more than 7000 suspected cases across the two 5 years, 6 to 12 years, 13 to 49 years, and 50 years
countries.6 Each year, 6 million doses of yellow or older. The cutoffs for these age strata were
fever vaccine are maintained by the World Health selected on the basis of data regarding immuno-
Organization (WHO) and partners in a global logic response to the yellow fever vaccine and
stockpile that can be used for outbreak response other vaccines.2,14
at the request of countries with inadequate vaccine All the participants who received fractional-
supply.7 However, the outbreaks in Angola and dose vaccination during the campaign were eli-
DRC used approximately 30 million doses and gible for inclusion unless they reported having
depleted the stockpile multiple times during 2016.6 immunosuppression, egg allergies, a history of
Faced with substantial global supply issues, problems with venipuncture, plans to relocate
the WHO reviewed available evidence on dose- from Kinshasa, or previous yellow fever vaccina-
sparing strategies for yellow fever vaccination, tion within the preceding 2 months. Children
including four studies involving three cohorts of under the age of 2 years and pregnant women
175 to 749 healthy adult participants.8-12 Two of were ineligible because they received full-dose
three cohorts were limited to male participants. vaccine according to the campaign operating pro-
All the studies showed a robust immune response cedures. The criteria for vaccine administration
to fractional doses of yellow fever vaccine as were determined by the public health authorities
small as one fifth to one tenth of the standard in the DRC. All the participants provided limited
dose. On the basis of this evidence, the WHO medical information and written informed con-
concluded that a fractional dose of the yellow sent to obtain blood samples. Parents or legal
fever vaccine could be used in adults and in chil- guardians provided written permission for par-
dren 2 years of age or older in response to an ticipants who were 17 years of age or younger.
emergency situation when the current vaccine Adolescents between the ages of 13 and 17 years
supply was insufficient.12 also provided written assent.
To prevent the spread of yellow fever in Kin-
shasa, the government planned a preemptive Study Oversight
campaign targeting approximately 7.6 million The study was sponsored by the U.S. Agency for
persons during a 10-day period.13 However, there International Development and the Centers for
was insufficient vaccine supply available to meet Disease Control and Prevention (CDC). The pro-
campaign needs. Thus, under guidance from the tocol (available with the full text of this article
WHO, the government of DRC implemented the at NEJM.org) was approved by the medical ethics
2 n engl j med nejm.org

Fr actional-Dose Vaccine in a Yellow Fever Outbreak
committee at the University of Kinshasa School ple was obtained. Female participants of repro-
of Public Health. In accordance with the human- ductive age were also asked about the date of the
subjects review procedures of the CDC, it was last menstrual period.
determined that the CDC was not formally en- Blood samples that were obtained before
gaged in human-subjects research. The study was vaccination and after vaccination were kept in
designed and supervised by the authors, who temperature-controlled coolers during the day
vouch for the accuracy and completeness of the and transported each afternoon to the Institut
data and analyses and the adherence of the study National de Recherche Biomédicale, where they
to the protocol. were centrifuged and serum was aliquoted into
cryovials and stored at −20°C. Serum samples
Baseline Visit and Vaccination were then shipped to the CDC Arbovirus Dis-
From each participant, we collected data regard- eases Laboratory in Fort Collins, Colorado, where
ing basic demographic characteristics, history of paired baseline and follow-up samples were test-
yellow fever vaccination, and recent symptoms ed for the presence of neutralizing antibodies
compatible with yellow fever disease (specifically, against yellow fever virus with the use of the
fever with jaundice). A phlebotomist obtained a plaque reduction neutralization test with a cutoff
baseline blood sample before vaccination. Cam- of 50% (PRNT50) and a cutoff of 90% (PRNT90),
paign staff members then administered a subcu- as described previously.16 Here, we report PRNT50
taneous dose of 17DD yellow fever vaccine at one titers, since this cutoff is routinely used in vac-
fifth of the standard dose (0.1 ml) to 764 par- cination trials of flavivirus vaccines and is recom-
ticipants from one of six lots: 253 participants mended by the WHO for establishing sufficient
received vaccine from lot 164VFC002Z, 104 from virus-neutralizing antibody in the serum in vac-
lot 164VFC003Z, 138 from lot 164VFC004Z, 127 cine immunogenicity studies conducted by vaccine
from lot 164VFC005Z, 38 from lot 164VFC007Z, manufacturers.17-19
and 94 from lot 164VFC008Z; no lot number was Participants with a PRNT50 titer of 10 or higher
recorded for doses administered to 10 partici- in their sample at baseline were considered to be
pants. seropositive. Those who had a baseline PRNT50
The 17DD vaccine was recommended by the titer of less than 10 and who became seroposi-
WHO for use in the campaign on the basis of tive at follow-up were classified as having under-
availability, clinical trial data indicating sero gone seroconversion. Participants who were sero-
response to fractional doses, and 5 years of positive at baseline and had an increase in titer by
batch-release data. According to these data, one a factor of 4 or more at follow-up were classified
fifth of a dose of the average batch potency had as having an immune response to vaccination.
8709 IU per dose, and one fifth of a dose of
minimum batch potency had 2692 IU per dose. Statistical Analysis
Both of these doses were above the minimum We determined that a sample of 760 participants
vaccine potency (1000 IU per dose) set by the would allow for an estimation of the immune
WHO.8-10,15 The vaccine was packaged in standard response in four age groups, based on an esti-
10-dose vials, which resulted in the use of each mated rate of immune response of 92% among
vial for approximately 50 fractional doses.13 Vac- the participants, with a 95% Wald confidence
cination was observed by study staff members to interval of ±5% and an attrition rate of 40%. All
ensure receipt of the dose. Adverse events after the participants who had baseline and follow-up
immunization were monitored as part of the samples were included in the analyses.
campaign procedures rather than as part of this Estimated proportions and 95% confidence
investigation. intervals were calculated with the use of the
Wilson method. We compared the proportion of
Follow-up and Testing Procedures participants who had undergone seroconversion
At 28 to 35 days after vaccination (follow-up in groups according to age and sex using the
period), participants who returned to the health chi-square test and Fisher’s exact test. Among
center were asked about yellow fever symptoms the participants who were seropositive at base-
and receipt of medications or medical treatment line, we assessed the association between the
during the interval between visits. A blood sam- baseline titer and immune response using the
Cochran–Armitage test for trend. Differences in R e sult s

immune response according to age group and
sex were adjusted for baseline titer subgroups Participants
with the use of the Cochran–Mantel–Haenszel Of the 863 persons who were screened, 790 met
test. We used the Kruskal–Wallis test and Wil- eligibility criteria, 764 were enrolled, and 716
coxon rank-sum test to compare geometric mean (94%) completed the follow-up visit (Fig. 1).
titers. Bonferroni corrections were used with Overall, 89 to 98% in each of the four age
pairwise comparisons. groups completed follow-up. Of the participants
with complete follow-up data, 50% were female,
and 79 (11%) reported having received previous
yellow fever vaccination; of these participants, all
863 Persons were assessed for eligibility but 5 were children 12 years of age or younger
(Table 1). A history of previous vaccination was
based primarily on oral report.
73 Were ineligible
Vaccine Response in the Overall Population

Of the 716 participants, 705 (98%; 95% confi-
790 Were eligible for inclusion
dence interval [CI], 97 to 99) were seropositive
after vaccination (Table 2). The proportion who
26 Declined to participate
were seropositive varied slightly according to age
(from 97% among those between the ages of 13
and 49 years to 100% among those between the
764 Participants were enrolled ages of 6 and 12 years) and according to sex
(97% among women and 99% among men), al-
though the differences among the groups were
40 Were excluded not significant. None of the participants who
20 Were lost to follow-up
12 Withdrew completed follow-up reported symptoms com-
4 Did not have an ade- patible with yellow fever after vaccination.
quate baseline blood
specimen
3 Had a complication Vaccine Response among Participants Who
during the first blood
draw
Were Seronegative at Baseline
1 Died before follow-up A total of 493 participants (69%) were seronega-
appointment
8 Were ineligible and inap- tive for neutralizing antibodies against yellow
propriately enrolled fever at baseline (Table 3). Among these partici-
3 Were previously vac-
cinated during a recent pants, seroconversion was reported in 482 (98%;
campaign 95% CI, 96 to 99). Among age groups, the lowest
5 Were pregnant or <2 yr
of age seroconversion rate was observed among chil-
dren between the ages of 2 and 5 years, although
the between-group differences in seroconversion
716 Completed follow-up at 28 to 35 days
were not significant (P = 0.06). At follow-up, par-
ticipants between the ages of 13 and 49 years
Figure 1. Enrollment and Follow-up. had a significantly elevated geometric mean titer
Of the 40 eligible participants who were excluded after of 2255 (95% CI, 1604 to 3171) as compared
the baseline data collection, 21 (52%) were female. Of with participants in all other age groups; children
the excluded participants, 19 (48%) were between the between the ages of 2 and 5 years had the lowest
ages of 2 and 5 years, 4 (10%) were between the ages geometric mean titer at 487 (95% CI, 293 to 810).
of 6 and 12 years, 8 (20%) were between the ages of
13 and 49 years, and 9 (22%) were 50 years of age or
The seroconversion rate among male participants
older. Investigation by the Ministry of Health determined (99%; 95% CI, 97 to 100) was significantly higher
that the single death after enrollment was related to a than that among female participants (96%; 95%
cardiac event and not to vaccination. CI, 93 to 98) (P = 0.03). However, the geometric
mean titers for male participants and female

Table 1. Characteristics of the 716 Study Participants at Baseline, According to Age Group.*
Characteristic Age Group

2–5 Yr 6–12 Yr 13–49 Yr ≥50 Yr All Ages
(N = 162) (N = 189) (N = 189) (N = 176) (N = 716)
number of participants (percent)
Female sex 89 (55) 87 (46) 102 (54) 80 (45) 358 (50)
Report of previous yellow fever vaccination 41 (25) 33 (17) 1 (1) 4 (2) 79 (11)
Oral report only 37 (23) 29 (15) 0 3 (2) 69 (10)
Vaccination card confirmed 0 3 (2) 1 (1) 0 4 (1)
Source not recorded 4 (2) 1 (1) 0 1 (1) 6 (1)
Report of symptoms compatible with yellow fever 2 (1) 1 (1) 1 (1) 3 (2) 7 (1)
in previous month†
Seropositivity at baseline‡ 85 (52) 75 (40) 27 (14) 36 (20) 223 (31)
* Percentages in subcategories may not add up to the overall percentage because of rounding.
† Symptoms were fever and jaundice.
‡ Seropositivity for neutralizing antibodies against yellow fever virus was defined as a titer on a plaque reduction neutralization test with a
cutoff of 50% (PRNT50) of 10 or higher.
participants did not differ significantly (P = 0.61). were at least 50 years of age had an immune
Among the 5 female participants of reproductive response.
age who did not undergo seroconversion, none
were pregnant on the basis of reports of men- Discussion
struation in the interval between vaccination
and the follow-up visit. In our study conducted during a mass vaccina-
tion campaign in Kinshasa, we found that de-
Vaccine Response among Participants Who tectable antibodies against the yellow fever virus
Were Seropositive at Baseline developed in 98% of the participants (≥2 years
At baseline, 223 participants (31%) were sero- of age) who were seronegative at baseline and
positive for yellow fever (Table 3). In this sub-
group, an immune response (titer of ≥4 times Table 2. Seropositivity at Follow-up for All 716 Participants, According to Age
the baseline value) was elicited in 148 (66%; 95% Group and Sex.*
CI, 60 to 72). There was an inverse correlation
between a participant’s baseline titer and the Variable Seropositivity at Follow-up P Value†
likelihood of having an immune response no./total no. % (95% CI)
(P<0.001). All the participants with a titer of 10 or All participants 705/716 98 (97–99)
20 had an immune response, as compared with Age group 0.08
none of the 11 participants who had a titer of 2–5 yr 158/162 98 (94–99)
2560 or higher at baseline (Fig. 2A). An anam-
6–12 yr 189/189 100 (98–100)
nestic response was more notable among those
13–49 yr 184/189 97 (94–99)
with lower baseline titers (Fig. 2B).
Among the participants who were seroposi- ≥50 yr 174/176 99 (96–100)
tive at baseline, there was no significant differ- Sex 0.06
ence in the proportion of male participants ver- Male 356/358 99 (98–100)
sus female participants who had an immune Female 349/358 97 (95–99)
response (P = 0.85). However, there was a sig-
nificant difference among age groups, even after * Seropositivity was defined as a result on PRNT50 testing of 10 or higher.
adjustment for the baseline titer (P<0.001); only † Texact
he P value is for the overall comparison among the subgroups by Fisher’s
test.
33% (95% CI, 20 to 50) of the participants who
Table 3. Seroconversion or Immune Response and Geometric Mean Titer at Follow-up, According to Serostatus at Baseline.*
Seroconversion or Immune Response Geometric Mean Titer

Subgroup at Follow-up P Value (95% CI) P Value
At Baseline At Follow-up
no./total no. % (95% CI)

Seronegative at baseline
All participants 482/493 98 (96–99) NA NA 1340 (1117–1607) NA
Age group 0.06† <0.001‡
2–5 yr 73/77 95 (87–98) NA 487 (293–810)
6–12 yr 114/114 100 (97–100) NA 1234 (911–1673)
13–49 yr 157/162 97 (93–99) NA 2255 (1604–3171)§
≥50 yr 138/140 99 (95–100) NA 1368 (999–1872)¶
Sex 0.03† 0.61‡
Male 251/253 99 (97–100) NA 1469 (1154–1870)
Female 231/240 96 (93–98) NA 1215 (924–1600)
Seropositive at baseline
All participants 148/223 66 (60–72) NA 87 (69–110) 1292 (1039–1607) NA
Age group <0.001‖ 0.04**
2–5 yr 57/85 67 (57–76) 90 (66–123) 1114 (787–1576)
6–12 yr 62/75 83 (73–90) 50 (35–72) 1366 (962–1938)
13–49 yr 17/27 63 (44–78) 160 (65–392) 2252 (1255–4039)
≥50 yr 12/36 33 (20–50)†† 160 (79–324) 1076 (533–2175)
Sex 0.85‖ 0.55**
Male 73/105 70 (60–78) 71 (51–98) 1255 (930–1694)
Female 75/118 64 (55–72) 105 (76–146) 1326 (964–1823)
* NA denotes not applicable.

† The P value in this category is for the difference in rates of seroconversion, according to age group or sex, as calculated by Fisher’s exact test.
‡ The P value in this category is for global testing of differences in the geometric mean titer at follow-up, according to age group (by the
Kruskal–Wallis test) or sex (by the Wilcoxon rank-sum test).
§ Participants in this category had significantly higher titers than those in all the other age groups, as calculated by the Wilcoxon rank-sum
test using the Bonferroni correction with an alpha level of 0.05 (0.008 after adjustment for the number of pairwise comparisons).
¶ Participants in this category had significantly higher titers than those who were between the ages of 2 and 5 years, as calculated by the
Wilcoxon rank-sum test using the Bonferroni correction with an alpha level of 0.05 (0.008 after adjustment for the number of pairwise
comparisons).
‖ The P value in this category is for the difference in immune response, according to age group or sex, as calculated by the Cochran–Mantel–
Haenszel test after adjustment for the baseline titer.
** The P value in this category is for the global test for differences in the increase in titer from baseline to follow-up, according to age group
(by the Kruskal–Wallis test) or sex (by the Wilcoxon rank-sum test).
†† Participants in this category were significantly less likely to have an immune response than those in all other age groups, as calculated
by the chi-square test using the Bonferroni correction with an alpha level of 0.05 (0.008 after adjustment for the number of pairwise com-
parisons).
who received one fifth of the standard dose of 2017 in Brazil, where more than 26 million vac-
the 17DD vaccine. This rate of seroconversion cine doses of yellow fever vaccine were distrib-
suggests that the use of fractional-dose vaccina- uted to control an outbreak during the begin-
tion is a viable approach for providing immunity ning of the year.20
and thus containing yellow fever outbreaks. This The proportion of participants who under-
finding is important, given the ongoing risk of went seroconversion was similar to that seen
outbreaks of yellow fever globally, as shown in among full-dose vaccine recipients, in whom

more than 98% undergo seroconversion.2 Our

A Immune Response at Follow-up
results are also similar to those seen in other 100
studies of fractional-dose vaccination against 90
yellow fever in which participants have received
Participants with Immune

80
as little as one fifth to one tenth of the standard 70
Response (%)
dose.8-11 The previously published studies were 60
performed in healthy, mostly male adults partici- 50
pating in well-controlled clinical trials. In con- 40
trast, our cohort included children and adults of 30
both sexes, and the vaccine was administered in 20
a mass campaign setting. Given the campaign 10
setting, it is notable that our results were similar 0
10– 20 40– 80 160– 320 620– 1280 ≥2560
to those in the controlled studies. (N= 70) (N= 65) (N= 42) (N= 35) (N= 11)
In 2003 in the DRC, yellow fever vaccine was Geometric Mean Titer at Baseline
introduced into the childhood vaccination pro-
gram and is administered to children at the age B Geometric Mean Titer at Follow-up
of 9 months.21 The routine use of yellow fever 8000
vaccine probably accounts for the higher rate of 7000
baseline seropositivity that we found among
Geometric Mean Titer
6000
children 12 years of age or younger than we
5000
found among the other age groups. Overall sero-
4000
positivity at follow-up and rates of seroconver-
sion did not vary significantly across age groups, 3000
even though there were differences in follow-up 2000

geometric mean titers. In addition, participants 1000
who were 50 years of age or older were less likely 0
than younger participants to have an immune 10– 20 40– 80 160– 320 620– 1280 ≥2560
(N= 70) (N= 65) (N= 42) (N= 35) (N= 11)
response if they had neutralizing antibodies at
baseline. Several trials of yellow fever vaccines Geometric Mean Titer at Baseline
have suggested a lower immunologic response to
Figure 2. Immune Response and Geometric Mean Titer at Follow-up
vaccination among children than among adults.5 among Participants Who Were Seropositive at Baseline.
In addition, data suggest that older adults may Panel A shows the proportion of participants who had an immune response
not have as vigorous a response to yellow fever after fractional-dose vaccination against yellow fever, according to the geo-
vaccine or other vaccines as do younger adults.2,14,22 metric mean titer of neutralizing antibodies at baseline. Panel B shows the
Although we found a slightly lower rate of geometric mean titer at follow-up according to the titer at baseline. In both
seroconversion among female participants than panels, I bars indicate 95% confidence intervals.
among male participants, the geometric mean
titers after vaccination were similar in both sexes.
The slightly higher immunologic response to the fractional doses of yellow fever vaccine. As part
vaccine among male participants has also been of this evaluation, participants have been fol-
reported in trials in which the full-dose yellow lowed up at 1 year after vaccination. Testing is
fever vaccine was used,23-25 which suggests that currently ongoing.
the observed difference is not unique to the frac- Our study has several limitations. First, we
tional dose. did not include a control group of participants
It is unknown whether the kinetics of anti- who received a full dose of yellow fever vaccine
body persistence will differ between fractional because of technical and ethical issues, so we
and full-dose administration of yellow fever vac- could not directly compare the immune response
cine. The geometric mean titers that were ob- after the fractional dose with that after a full
served in our study suggest that immunity will dose. Also, the use of PRNT50 titers may have
probably persist for years and possibly be life- caused incorrect classification of participants
long. Martins et al.8 found detectable antibodies with low titers as being seropositive for neutral-
at 10 months after the administration of various izing antibodies against the yellow fever virus
when in fact the titer was due to cross-reactive yellow fever vaccine was appropriate for a re-
antibodies. However, since serum samples were sponse to a yellow fever outbreak among children
obtained before vaccination and after vaccina- 2 years of age or older and among nonpregnant
tion from each participant, the specific response adults. Additional studies are needed to deter-
to the vaccine could still be assessed. We did not mine whether fractional-dose vaccination pro-
calibrate the yellow fever antibody titers using vides adequate seroprotection in children under
an international reference preparation, which the age of 2 years, in pregnant women, and in
makes it difficult to compare our titers with persons who are infected with the human im-
those obtained in the other recent studies of munodeficiency virus.12 In addition, future studies
fractional-dose vaccination.17 International stan- need to verify that similar results are obtained
dardization of testing results for yellow fever has with other 17D-derived yellow fever vaccines
only recently been recommended, so very few (17D-204 and 17D-213) and in populations with
data have been generated. The use of PRNT titers differing exposures to flaviviruses. The use of
was preferred for this study to allow for com- fractional doses of yellow fever vaccine could
parison with much of the published data. In ad- reduce the amount of vaccine needed for reactive
dition, we did not collect safety data during this campaigns and provide flexibility in manage-
evaluation.26 However, the adverse-event monitor- ment of the global stockpile of yellow fever vac-
ing systems that were in place for the campaign cine when outbreaks occur.
did not identify any acute signals of concern
The views expressed in this article are those of the authors
associated with fractional-dose vaccination. En- and do not necessarily represent the official position of the
hanced surveillance detected 0.5 serious adverse Centers for Disease Control and Prevention.
events per 100,000 doses after the campaign,13 a Supported by the U.S. Agency for International Development
and the Centers for Disease Control and Prevention.
rate that is similar to that after full-dose cam- Disclosure forms provided by the authors are available with
paigns conducted in West Africa.27 Finally, we the full text of this article at NEJM.org.
could not formally assess the effectiveness of the We thank all the study participants and members of the cam-
paign and study staff; staff members at the Institut National de
fractional dose with regard to preventing viral Recherche Biomédicale, the World Health Organization (WHO),
transmission during the outbreak, since the out- the Centers for Disease Control and Prevention (CDC), the U.S.
break was waning at the time of the campaign. Agency for International Development, and Management Sci-
ences for Health; Joachim Hombach and Sergio Yactayo of the
However, no new confirmed cases of yellow fever WHO and Alan Barrett of the University of Texas Medical Branch
were detected in Kinshasa after the campaign for their technical support; Jason Velez and Barbara W. Johnson
despite ongoing surveillance. for laboratory assistance at the CDC Arbovirus Diseases Labora-
tory; and the Ministry of Health of the Democratic Republic of
In conclusion, we found that the immuno- Congo and the CDC Global Disease Detection Operations Center
logic response to a fractional dose of the 17DD for their collaboration in implementing this study.
References
1. Garske T, Van Kerkhove MD, Yactayo 2015. MMWR Morb Mortal Wkly Rep 2015; valho R. Studies on yellow fever vaccine.
S, et al. Yellow fever in Africa: estimating 64:647-50. III. Dose response in volunteers. J Biol
the burden of disease and impact of mass 6. Situation report:yellow fever situa- Stand 1988;16:77-82.
vaccination from outbreak and serologi- tional report, 28 October 2016. Geneva: 11. Roukens AH, Gelinck LB, Visser LG.
cal data. PLoS Med 2014;11(5):e1001638. World Health Organization (http://apps Intradermal vaccination to protect against
2. Monath TP, Gershman M, Staples JE, .who.int/i ris/bitstream/10665/250661/1/ yellow fever and influenza. Curr Top Micro-
Barrett ADT. Yellow fever. In:Plotkin SA, yellowfeversitrep28Oct16-eng.pdf?ua=1). biol Immunol 2012;351:159-79.
Orenstein WA, Offit PA, eds. Vaccines. 7. Update on progress controlling yel- 12. Fractional dose yellow fever vaccine
6th ed. Philadelphia:Elsevier Saunders, low fever in Africa, 2004-2008. Wkly Epi- as a dose-sparing option for outbreak re-
2013:870-968. demiol Rec 2008;83:450-8. sponse:WHO Secretariat information
3. Beck AS, Barrett AD. Current status 8. Martins RM, Maia MdeL, Farias RH, paper. Geneva:World Health Organiza-
and future prospects of yellow fever vac- et al. 17DD yellow fever vaccine: a double tion, July 20, 2016 (http://apps.who.int/iris/
cines. Expert Rev Vaccines 2015;14:1479- blind, randomized clinical trial of immu- bitstream/10665/246236/1/WHO-YF-SAGE
92. nogenicity and safety on a dose-response -16.1-eng.pdf?ua=1).
4. Vaccines and vaccination against yel- study. Hum Vaccin Immunother 2013;9: 13. Yellow fever mass vaccination cam-
low fever: WHO position paper — June 879-88. paign using fractional dose in Kinshasa,
2013. Wkly Epidemiol Rec 2013;88:269- 9. Campi-Azevedo AC, de Almeida Este- DRC. Geneva:World Health Organization,
83. vam P, Coelho-Dos-Reis JG, et al. Sub September 26, 2016 (http://www .who
5. Staples JE, Bocchini JA Jr, Rubin L, doses of 17DD yellow fever vaccine elicit .int/i mmunization/sage/meetings/2016/
Fischer M. Yellow fever vaccine booster equivalent virological/immunological ki- october/4_Yellow_fever_mass_vaccination
doses: recommendations of the Advisory netics timeline. BMC Infect Dis 2014;14:391. _campaign_using_fractional_dose_in_
Committee on Immunization Practices, 10. Lopes Ode S, Guimarães SS, de Car Kinshasa.pdf?ua=1).

14. Grohskopf LA, Sokolow LZ, Broder sultation on immunological endpoints for Sonnenberg K. Evaluation of an indirect
KR, et al. Prevention and control of sea- evaluation of new Japanese encephalitis immunofluorescence assay for detection
sonal influenza with vaccines. MMWR vaccines, WHO, Geneva, 2-3 September, of immunoglobulin M (IgM) and IgG anti-
Recomm Rep 2016;65:1-54. 2004. Vaccine 2005;23:5205-11. bodies against yellow fever virus. Clin
15. Fractional dosing of yellow fever vac- 19. Roehrig JT, Hombach J, Barrett AD. Vaccine Immunol 2008;15:177-81.
cines:review of vaccine potency and sta- Guidelines for plaque-reduction neutral- 24. Camacho LA, da Silva Freire M, da Luz
bility information, available evidence, and ization testing of human antibodies to Fernandes Leal M, et al. Immunogenicity
evidence gaps. Presented at the Strategic Dengue viruses. Viral Immunol 2008;21: of WHO-17D and Brazilian 17DD yellow
Advisory Group of Experts (SAGE) meeting, 123-32. fever vaccines: a randomized trial. Rev
Geneva, October 18–20, 2016 (http://www 20. Centro de Operações de Emergências Saude Publica 2004;38:671-8.
.who.int/i mmunization/sage/meetings/ em Saúde Pública sobre Febre Amarela. 25. Monath TP, Nichols R, Archambault
2016/october/presentations_background Informe – nº 43. 2017 (http://portalarquivos WT, et al. Comparative safety and immu-
_docs/en/index1.html). .saude.gov.br/i mages/pdf/2017/junho/02/ nogenicity of two yellow fever 17D vac-
16. Beaty BJ, Calisher CH, Shope RE. Arbo- COES-FEBRE-AMARELA---INFORME-43 cines (ARILVAX and YF-VAX) in a phase
viruses. In:Lennette EH, Lennette DA, ---Atualiza----o-em-31maio2017.pdf). III multicenter, double-blind clinical trial.
Lennette ET, eds. Diagnostic procedures 21. Immunization, vaccines and biologi- Am J Trop Med Hyg 2002;66:533-41.
for viral, rickettsial and chlamydial Infec- cals:data, statistics and graphics. Geneva: 26. Yellow fever vaccine: WHO position
tions. 7th ed. Washington, DC:American World Health Organization, 2017 (http:// on the use of fractional doses — June
Public Health Association, 1995:189-212. www.who.int/i mmunization/monitoring 2017. Wkly Epidemiol Rec 2017;92:345-
17. Recommendations to assure the qual- _surveillance/data/en/). 50.
ity, safety and efficacy of live attenuated 22. Roukens AH, Soonawala D, Joosten 27. Breugelmans JG, Lewis RF, Agbenu E,
yellow fever vaccines. Geneva:World Health SA, et al. Elderly subjects have a delayed et al. Adverse events following yellow fever
Organization, 2010 (http://www.who.int/ antibody response and prolonged virae- preventive vaccination campaigns in eight
biologicals/Y F_Recommendations_post mia following yellow fever vaccination: African countries from 2007 to 2010. Vac-
_ECBS_FINAL_rev_12_Nov_2010.pdf). a prospective controlled cohort study. PLoS cine 2013;31:1819-29.
18. Hombach J, Solomon T, Kurane I, Ja- One 2011;6(12):e27753. Copyright © 2018 Massachusetts Medical Society.
cobson J, Wood D. Report on a WHO con- 23. Niedrig M, Kürsteiner O, Herzog C,

Immunogenicity of Fractional-Dose Vaccine During A Yellow Fever Outbreak - Preliminary Report

Uploaded by

Copyright:

Available Formats

Immunogenicity of Fractional-Dose Vaccine During A Yellow Fever Outbreak - Preliminary Report

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Immunogenicity of Fractional-Dose Vaccine During A Yellow Fever Outbreak - Preliminary Report

Uploaded by

Copyright:

Available Formats

The n e w e ng l a n d j o u r na l of m e dic i n e