Incidence of severe Plasmodium falciparum malaria as a primary endpoint for vaccine efficacy trials in Bandiagara, Mali

Center for Vaccine Development, University of Maryland School of Medicine, 685 West Baltimore Street, HSF 480, Baltimore, MD, USA.
Vaccine (Impact Factor: 3.62). 08/2004; 22(23-24):3169-74. DOI: 10.1016/j.vaccine.2004.01.054
Source: PubMed


Potential endpoints for blood stage malaria vaccine efficacy trials include uncomplicated malaria disease, which is hard to differentiate from other febrile illnesses, and mortality, which requires prohibitively large sample sizes. Strictly defined severe malaria predicts malaria-associated mortality where case fatality rates are known. To assess the suitability of severe malaria as a trial endpoint, we conducted a census in 1999 and measured the incidence of severe malaria from 1999 to 2001 in Bandiagara, Mali. The annual incidence of severe malaria in children <6 years of age was 2.3% (n = 2,284) yielding an estimated sample size of 4,580 for a vaccine trial designed to detect 50% efficacy with 80% power at P = 0.05 with 5% loss to follow-up. A trial using severe malaria as an endpoint in this setting would thus require expanding the study population or the length of the trial. This approach may be useful in assessing the suitability of potential sites for malaria vaccine trials.


Available from: Yacouba Cissoko, Sep 29, 2014
Vaccine 22 (2004) 3169–3174
Incidence of severe Plasmodium falciparum malaria as a primary
endpoint for vaccine efficacy trials in Bandiagara, Mali
Kirsten E. Lyke
, Alassane Dicko
, Abdoulaye Kone
, Drissa Coulibaly
, Ando Guindo
Yacouba Cissoko
, Karim Traoré
, Christopher V. Plowe
, Ogobara K. Doumbo
Center for Vaccine Development, University of Maryland School of Medicine, 685 West Baltimore Street, HSF 480, Baltimore, MD, USA
Malaria Research and Training Center, Department of Epidemiology of Parasitic Diseases, Faculty of Medicine, Pharmacy and Dentistry,
Bandiagara Malaria Project, University of Bamako, Bamako, Mali
Received 24 June 2003; received in revised form 19 December 2003; accepted 15 January 2004
Available online 27 February 2004
Potential endpoints for blood stage malaria vaccine efficacy trials include uncomplicated malaria disease, which is hard to differen-
tiate from other febrile illnesses, and mortality, which requires prohibitively large sample sizes. Strictly defined severe malaria predicts
malaria-associated mortality where case fatalityratesareknown. Toassessthesuitabilityofseveremalariaasatrialendpoint,weconducted
a census in 1999 and measured the incidence of severe malaria from 1999 to 2001 in Bandiagara, Mali. The annual incidence of severe
malaria in children <6 years of age was 2.3% (n = 2, 284) yielding an estimated sample size of 4,580 for a vaccine trial designed to detect
50% efficacy with 80% power at P = 0.05 with 5% loss to follow-up. A trial using severe malaria as an endpoint in this setting would
thus require expanding the study population or the length of the trial. This approach may be useful in assessing the suitability of potential
sites for malaria vaccine trials.
© 2004 Elsevier Ltd. All rights reserved.
Keywords: Severe malaria; Vaccine endpoints; Plasmodium falciparum
1. Introduction
Plasmodium falciparum malaria is a multistage parasite
invoking stage-specific host immune responses. To date, the
approach to malaria vaccine development has primarily con-
centrated on eliciting these stage-specific immune responses.
Malaria vaccine candidates can be divided into those that
stimulate a pre-erythrocytic (sporozoite or liver) immune re-
sponse, those that block transmission of malaria by promot-
ing an antibody response against antigens expressed in the
sexual stage within the midgut of the mosquito, those that
target the asexual blood stages of the parasite, and vaccines
reducing the toxic effect of the infection. Natural immunity
against malaria appears to be acquired over time, requires
multiple infective bites, and is lost in the absence of contin-
ued exposure. The goal of blood stage malaria vaccines is
to protect against disease and death, which can be achieved
through acquisition of such immunity.
Corresponding author. Fax: +1-410-706-6205.
E-mail address: (C.V. Plowe).
Defining clinical endpoints for malaria vaccine effi-
cacy trials has proved challenging. In the case of a pre-
erythrocytic vaccine, designed to induce sterile immunity,
breakthrough infection is easily determined by first para-
sitemia or first symptomatic infection. Blood stage vaccines
are intended to block disease and not infection. Unfortu-
nately, defining clinical disease can often be difficult. There
are no protective antibody thresholds, nor in vitro immuno-
logic assays that can distinguish immune individuals from
non-immune individuals [1]. In the absence of an objective
immune assay, the endpoint of a blood stage vaccine is
often reduction of a carefully defined clinical illness in im-
munized population specific to the site at which the trial is
conducted. Accurate definitions of malaria disease are not
only age and site specific, they likely vary over the course
of a transmission season as immunity is acquired. Addition-
ally, malarial disease has a variety of clinical symptoms,
broad range of parasitemia, and may be confused with other
childhood illnesses that occur in the setting of baseline lev-
els of malaria infection. Without accurate interpretations of
malaria disease, accurate predictions of malaria-associated
mortality may not be achieved.
0264-410X/$ see front matter © 2004 Elsevier Ltd. All rights reserved.
Page 1
3170 K.E. Lyke et al./Vaccine 22 (2004) 3169–3174
Investigators have sought to find alternative methods of
estimating vaccine efficacy through surrogate endpoints.
The use of malaria attributable fraction of fever has been
advocated as a marker for clinical malaria. Logistic regres-
sion methods modeling fever as a continuous function of
parasitemia appear to offer precise estimates of disease,
although definitions vary by site and parasite transmission
intensity. Additionally, results are unreliable in children un-
der 12 months or greater than 9 years of age [2–4]. Some
advocate using all-cause or malaria-associated mortality
as an endpoint, but this would require prohibitively large
sample sizes. As an example, insecticide-impregnated bed-
net studies that showed decreases in all-cause mortality
had sample sizes in the tens and hundreds of thousands
involving large catchment areas and multiple villages [5–7].
For the purposes of conducting phase III studies, the
World Health Organization (WHO) recommends establish-
ing protection against a single primary efficacy endpoint [8].
Severe disease and malaria mortality represent the greatest
public health consequences of infection. Incidence of severe
malaria as an endpoint offers the advantages of being a good
indicator of malaria-associated mortality where case fatal-
ity rates are known with and requiring fewer subjects than a
mortality endpoint. Thus, in a setting with a 10% case fatal-
ity rate for severe malaria, 10-fold fewer subjects would be
required for a study with severe malaria as the primary end-
point than a study assessing mortality due to malaria. While
imperfect, the criteria for severe malaria as defined by the
WHO are a standardized definition that can be applied uni-
formly [9]. Although this definition has its limitations, few
would disagree that in contrast to children meeting case def-
initions of uncomplicated malaria, most children who meet
criteria for severe malaria are suffering from malaria disease
and have a worse prognosis.
2. Methods and materials
2.1. Study site and subjects
As part of the development of a site for testing vaccines in
West Africa, a complete population census was conducted in
1999 in Bandiagara, Mali. Bandiagara is a rural town in east
central Mali, West Africa and has intense seasonal transmis-
sion (July–December) of P. falciparum malaria. During the
peak of the transmission season in October, the number of
infected bites per person per month ranged from 20 to 60 in
different regions of Bandiagara town. The village of Bandi-
agara is divided into eight quartiers or zones. The dominant
ethnic group is Dogon (80%) with the remainder of the pop-
ulation being 10% Peuhl, 3% Bambara, and 7% other eth-
nic groups. Demographic information on 12,583 individuals
(47% aged <12 years) was collected.
A clinic operating 24 h a day, 7 days a week was estab-
lished, providing the only source of medical care for severe
malaria available locally. As part of an ongoing effort to
characterize the immune response to malaria, index cases
of severe malaria in children aged 3 months to 14 years
were enrolled, but for the purposes of this analysis, only
children aged <6 years (those most susceptible to severe
malaria) were included. Study protocols were reviewed and
approved by the local Malian Institutional Review Board
(IRB) as well as the University of Maryland IRB. Village as-
sent was obtained from village chiefs, government officials
and traditional healers prior to study initiation. Individual
informed consent was obtained from the legal guardian of
each child prior to enrollment though care for severe malaria
was offered regardless of study participation. Local tradi-
tional healers, who are generally first-line providers for most
medical problems, agreed to refer cases of suspected severe
malaria to the study clinic.
At enrollment, clinical information was taken and en-
tered onto standardized forms. Prior to weight-based quinine
and pyrimethamine–sulfadoxine (Fansidar
) therapy, bed-
side physical exams were performed and diagnostic periph-
eral blood smears, hemoglobin, and glucose obtained. Cases
were classified as severe malaria based on modified criteria
put forth by the World Health Organization [9].(Table 1).
2.2. Statistical analysis
Pooled analyses between clinical groups were made us-
ing two-sided Student’s t-test for continuous variables with
equal variance and
test for categorical variables using
SPSS 10.0 (Chicago, IL 2000) and EpiInfo 2000 (Centers
for Disease Control, Atlanta, GA 2000) with Fisher’s Ex-
act testing for cell quantity <5. A level of statistical signif-
icance (two-sided) was set at P<0.05. The incidence of
severe malaria was used to compute the sample size for a
phase III vaccine trial to detect an efficacy of 50% based on
the methods of Fleiss [10].
Table 1
Defining features of severe malaria base on modified World Health Or-
ganization criteria
Coma (Blantyre coma scale (BCS) <2)
Seizure (one or more witnessed by the investigators)
Obtundation (depressed consciousness with BCS >2)
Parasitemia >500,000/mm
Lethargy or prostration (clinical judgment or child >7 months unable to
sit unassisted)
Severe anemia (hemoglobin <5g/dl)
Respiratory distress (intercostal muscle retraction, deep breathing,
Hypoglycemia (glucose <40mg/dl)
Renal insufficiency as indicated by lack of urination for >1 day
Gross hematuria
Inability to eat or drink*
State of shock (systolic blood pressure <50mm Hg, rapid pulse, cold
Repeated vomiting*
The presence of any one of the signs or symptoms except for two criteria
marked by asterix, in the presence of parasitemia, was sufficient for the
diagnosis of severe malaria. The symptoms of intractable vomiting and/or
inability to eat or drink were included in the interest of safety.
Page 2
K.E. Lyke et al./Vaccine 22 (2004) 3169–3174 3171
Table 2
Demographic information on severe malaria in the greater Bandiagara, Mali West Africa region
Characteristic Bandiagara Outside Bandiagara
Survived Died Survived Died
Total enrolled 99 5 (4.8%) 35 8 (18.6%)
Gender (female) 49 (49%) 4 (80%) 17 (49%) 3 (38%)
Age (months) 38.7
20.7 27.6
Hemoglobin (g/dl) 8.8
5.4 6.6
Geometric mean parasite density (mm
) 231,072 20,522
91,772 50,178
Glucose (mg/dl) 119 84
120 109
Denotes statistical significance (P<0.05) between children that survived or died in region depicted.
Denotes statistical significance (P<0.05) between children from Bandiagara versus outside Bandiagara.
3. Results
3.1. Clinical characteristics
One hundred and four cases of severe malaria in chil-
dren aged 6 years (54 female, 50 male) from Bandia-
gara town were enrolled between October 6, 1999–October
7, 2001 with 30 additional enrollments from outlying vil-
lages. (See demographic information in Table 2) 52 cases
were enrolled from October 6, 1999–October 7, 2000 and
52 from October 7, 2000–October 7, 2001. The mean age at
enrollment for all cases of severe malaria (including those
from outside Bandiagara) was 35.8 months. Younger chil-
dren were at higher risk of death, with a mean age of 20.6
months in children who died from severe malaria (n =
13, overall case fatality rate 9.7%, P = 0.01). Among
children from Bandiagara town, only five deaths occurred,
yielding a case fatality rate of 4.8%. Enrollment criteria
for severe malaria included cerebral malaria (coma, ob-
tundation, seizure) in 49.3%, severe anemia (hemoglobin
5g/dl) 20.8%, hyperparasitemia (500,000 asexual para-
sites per cubic millimeter) 54.4%, prostration 17.2%, respi-
ratory distress 5.2%, and hypoglycemia in 0.7%. More than
one clinical criterion for severe malaria could be present in
a single individual. Residence outside of Bandiagara was
a strong predictor of death in the event of severe malaria
(OR = 4.83 [95% C.I.: 1.23–17.3] P = 0.01), likely due to
longer periods between onset of symptoms and initiation of
3.2. Calculated sample size requirement
Review of census data revealed 2,284 children resid-
ing in Bandiagara that were aged 6 years. We used
the average number of cases of severe malaria in a year
and census data, to calculate an incidence rate of se-
vere malaria of 2.3% (52/2,284) in children aged 6
years from Bandiagara town. Based on this incidence, a
placebo-controlled trial designed to detect a vaccine efficacy
of 50% with 80% power, two-tailed significance of 0.05
and 5% loss to follow-up would require a sample size of
4. Discussion
As potential malaria vaccinecandidates enter the pipeline,
strategies for assessing efficacy of these constructs and in
particular, blood stage vaccines, are needed. The WHO has
recommended the use of one of five endpoints (parasitemia,
uncomplicated disease, severe disease, malaria-associated
mortality and all-cause mortality) for use in malaria vaccine
field trials [8]. While severe manifestations of malaria dis-
ease are of greater public health importance and are more
precisely defined, they represent a smaller percentage of
overall cases, increasing sample size requirements. Mild
clinical disease is often advocated as an acceptable choice of
endpoint though this surrogate measure does not necessar-
ily reflect incidence of severe disease or death [8,11]. Using
severe malaria as an endpoint, we calculated that in a small
town in Mali with high, seasonal endemicity and stable dis-
ease incidence, a vaccine trial would require a sample size
of 4,580.
The use of severe malaria as a clinical endpoint in phase
III malaria vaccines offers advantages and disadvantages.
The sample size calculated utilizing the incidence of severe
malaria in Bandiagara is larger than that of previous blood
stage malaria vaccine efficacy trials [12–18]. As stated, the
total population of children, in Bandiagara, aged 6 years
is 2,284. Enrollment to meet the calculated sample size of
4,580 would require increased labor, catchment areas en-
compassing outside villages, improvedinfrastructure and en-
hanced costs. Paradoxically, the use of thisoutcome measure
may actually require less intense diagnostic monitoring than
documenting first clinical malaria infections or parasitemia
as endpoints. Clinics providing the sole source of care for
severe malaria in an isolated community may rely on pas-
sive case detection rather than active follow-up. In suggest-
ing passive case detection, we do not advocate an approach
in which children are allowed to deteriorate to a condition of
severe disease for the purposes of having a good study end-
point. On the contrary, education of local health care work-
ers, traditional healers, community leaders and population
at large to recognize signs of malaria disease minimizes de-
lays in treatment seeking and cases lost to follow-up. One
must note that climactic variation, political instability, and
Page 3
3172 K.E. Lyke et al./Vaccine 22 (2004) 3169–3174
Table 3
Summary analysis of published phase II/III field Plasmodium falciparum malaria vaccine trials and endpoints used in determining efficacy
Construct Site Year Number enrolled (age) Endpoint Definition Efficacy
(%) Author (Reference)
Columbia 1991 399 (Males, 18–21years) Clinical infection Smear positive and febrile 58.9–2 doses
82.8–3 doses
Amador [12]
Ecuador 1993 537 (>1 year) Blood parasitemia Smear positive 66.8 Semp
ertegui [13]
Columbia 1993 1548 (>1 year) Clinical infection Smear positive and symptoms 38.8 Valero [14]
Venezuela 1994 2360 (>11 years) Clinical infection Smear positive and symptoms 55.0 Noya [15]
Tanzania 1994 586 (1–5 years) Clinical infection Smear positive (>20,000/l),
temp >37.5
31.0 Alonso [16]
Gambia 1995 547 (6–11 months) Clinical infection Smear positive (>6000/l),
temp > 37.5
8.0-first episode
3.0-all episodes
D’Alessandro [17]
Thailand 1996 1221 (2–15 years) First clinical infection Smear positive, hx of
symptoms or temp > 38
9.0 Nosten [18]
Columbia 1996 1257 (1–86 years) Blood parasitemia Smear positive 35.2 Valero [21]
Brazil 1998 572 (7–60 years) Blood parasitemia Smear positive (Pf or Pv)
14.1-first episode-1.6-all
Urdaneta [22]
Tanzania 1999 1207 infants (at 1, 2,7 months) First clinical infection Smear positive, temp >37.5
C 2.0 Acosta [20]
Gambia 2001 306 (Males, 18–45 years) Blood parasitemia Smear positive 34 - 3 doses 47 -4 doses Bojang [19]
Combination B
Papua New Guinea 2002 120 (5-9 years) Reduction in parasite density parasites/l per 200 WBC 62.0 Genton [23]
Efficacy defined as reduction in defined endpoint during study period but does not include data on length of protection.
SPf66 construct is a synthetic peptide composed of amino acid sequences from three asexual stage merozoite proteins linked by NANP repeat sequences of the Pf circumsporozoite protein with
alum hydroxide adjuvant.
RTS, S construct composed of carboxyl-terminal half of Pf circumsporozoite protein (strain NF54 clone 3D7) fused to hepatitis B surface antigen in novel AS02 adjuvant.
Combination B construct composed of Pf ring-infected erythrocyte surface antigen (RESA) and merozoite surface proteins 1 and 2 (3D7 allele) in Montanide ISA720 oil-based adjuvant.
Pf: Plasmodium falciparum; Pv: Plasmodium vivax. Results only reported for P. falciparum.
Page 4
K.E. Lyke et al./Vaccine 22 (2004) 3169–3174 3173
improvements or decrements in the public health infrastruc-
ture may alter malaria transmission, affecting sample size
calculations based on disease prevalence. Nevertheless, over
6 years in Bandiagara, we have seen stable numbers of se-
vere malaria cases with reduced yearly fatality rates of 4–9%
following a marked initial reduction at study initiation, il-
lustrating the potential utility this model might have in field
To date, most malaria vaccine efficacy trials have tested
candidates that included pre-erythrocytic antigens [12–19].
(Table 3) Unlike these constructs, asexual blood stage
vaccines aim not to prevent infection but to reduce clin-
ical disease and death in residents of malaria-endemic
areas. The importance of developing endpoints that al-
low for generalization of conclusions is critical. This
point is well illustrated by the experience of field trials of
the SPf66 vaccine which is comprised of short synthetic
peptides derived from blood stage antigens and linked
by tetra-peptide repeat sequences from circumsporozoite
protein. Although initially appearing promising in South
American field trials (areas of low endemicity) [12–15],
later and more rigorously designed trials in Africa and
Asia (areas of high endemicity) met with disappointing
results [16–18,20]. As noted in Table 3, vaccine efficacy
ranged from 9 to 74.3% efficacy. Endpoints were de-
fined as parasitemia or symptomatic infection, although
the definition of clinical infection varied as did the age
range of the target enrollment population. Regional and
seasonal variability in malaria intensity and transmission,
parasite and host genetic diversity, differences in age dis-
tribution, and confounding variables may have contributed
to difficulties interpreting results. While it was ultimately
concluded that the Spf66 vaccine construct was not effi-
cacious, one wonders if standardized endpoints and trial
designs had been used from the start, conclusions might
have been made earlier with considerable effort and money
We conclude that severe malaria offers advantages over
most other possible endpoints for asexual blood stage
malaria vaccine efficacy studies. Incidence data gathered
in Bandiagara, Mali demonstrates stable disease in this
setting, permitting calculation of a sample sizes for fu-
ture trials here. This approach may be suitable at other
sites with relatively stable patterns of malaria transmis-
sion and epidemiology. We propose that measuring severe
malaria incidence and case fatality rates in demographi-
cally well-defined populations provides a suitable method
for identifying appropriate populations and calculating
sample sizes for blood stage malaria vaccine efficacy
This study was supported by contract no. N01-AI-85346
from the National Institutes of Health (NIH).
[1] Alonso PL, Tanner M, Smith T, Hayes RJ, Schellenberg JA, Lopez
MC, et al. A trial of the synthetic malaria vaccine SPf66 in Tanzania:
rationale and design. Vaccine 1994;12(2):181–6.
[2] Smith T, Schellenberg JA, Hayes R. Attributable fraction estimates
and case definitions for malaria in endemic areas. Stat Med
[3] Smith T, Hurt N, Teuscher T, Tanner M. Is fever a good sign for
clinical malaria in surveys of endemic communities? Am J Trop
Med Hyg 1995;52(4):306–10.
[4] McGuinness D, Koram K, Bennett S, Wagner G, Nkrumah F, Riley
E. Clinical case definitions for malaria: clinical malaria-associated
with very low parasite densities in African infants. Trans R Soc Trop
Med Hyg 1998;92(5):527–31.
[5] Abdulla S, Schellenberg JA, Nathan R, Mukasa O, Marchant T, Smith
T, et al. Impact on malaria morbidity of a programme supplying
insecticide treated nets in children aged under 2 years in Tanzania:
community cross-sectional study. BMJ 2001;322(7281):270–3.
[6] D’Alessandro U, Olaleye BO, McGuire W, Langercrock P, Bennett S,
Aikins MK, et al. Mortality and morbidity from malaria in Gambian
children after introduction of an impregnated bednet programme.
Lancet 1995;345(8948):479–83.
[7] Alonso PL, Lindsay SW, Armstrong JR, Conteh M, Hill AG, David
PH, et al. The effect of insecticide-treated bed nets on mortality of
Gambian children. Lancet 1991;337(8756):1499–502.
[8] World Health Organization. Guidelines for the evaluation of
Plasmodium falciparum vaccines in populations exposed to natural
infection, Geneva. 1997, p. 1 Document TDR/MAL/VAC/97.
[9] Severe falciparum malaria. World Health Organization, Commu-
nicable Diseases Cluster. Trans. R. Soc. Trop. Med. Hyg. 2000;94
Suppl 1:S1–90.
[10] Fleiss JL. John Willey and Sons, editor. Statistical methods for rate
and proportion. 2nd ed. John Willey and Sons, 1981. p. 44.
[11] Smith PHR, editor. Wiley and Sons, Malaria: Waiting for the Vaccine,
1991. p. 199.
[12] Amador R, Moreno A, Valero V, Murillo L, Mora AL, Rojas M,
et al. The first field trials of the chemically synthesized malaria
vaccine SPf66: safety, immunogenicitym, and protectivity. Vaccine
[13] Sempertegui F, Estrella B, Moscoso J, Piedrahita C, Hernandez D,
Gaybor J, et al. Safety, immunogenicity and protective effect of the
SPf66 malaria synthetic vaccine against Plasmodium falsciparum
infection in a randomized double-blind placebo-controlled field trial
in an endemic area of Ecuador. Vaccine 1994;12(4):337–42.
[14] Valero MV, Amador LR, Galindo C, Figueroa J, Bello MS,
Murillo LA, et al. Vaccination with SPf66, a chemically synthesised
vaccine, against Plasmodium falciparum malaria in Colombia. Lancet
[15] Noya O, Gabaldon BY, Alarcon de Noya B, Borges R, Zerpa N,
Urbaez JD, et al. A population-based clinical trial with the SPf66
synthetic Plasmodium falciparum malaria vaccine in Venezuela. J
Infect Dis 1994;170(2):396–402.
[16] Alonso PL, Smith T, Schellenberg JR, Masanja H, Mwankusye S,
Urassa H, et al. Randomized trial of efficacy of SPf66 vaccine against
Plasmodium falciparum malaria in children in southern Tanzania.
Lancet 1994;344(8931):1175–81.
[17] D’Alessandro U, Leach A, Drakely CJ, Bennett S, Olaleye BO,
Fegan GW, et al. Efficacy trial of malaria vaccine SPf66 in Gambian
infants. Lancet 1995;346(8973):462–7.
[18] Nosten F, Luxemburger C, Kyle DE, Ballou WR, Wittes J, Wah
E, et al. Randomized double-blind placebo-controlled trial of SPf66
malaria vaccine in children in northwestern Thailand. Shoklo SPf66
Malaria Vaccine Trial Group. Lancet 1996;348(9029):701–7.
[19] Bojang KA, Milligan PJ, Pinder M, Vigneron L, Alloueche A, Kester
K, et al. Efficacy of RTS,S/AS02 malaria vaccine against Plasmodium
falciparum infection in semi-immune adult men in The Gambia: a
randomized trial. Lancet 2001;358(9297):1927–34.
Page 5
3174 K.E. Lyke et al./Vaccine 22 (2004) 3169–3174
[20] Acosta CJ, Galindo CM, Schellenberg D, Aponte JJ, Kahigwa E,
Urassa H, et al. Evaluation of the SPf66 vaccine for malaria control
when delivered through the EPI scheme in Tanzania. Trop Med Int
Health 1999;4(5):368–76.
[21] Valero MV, Amador R, Aponte JJ, Narvaez A, Galindo C, Silva
Y, et al. Evaluation of SPf66 malaria vaccine during a 22-month
follow-up field trial in the Pacific coast of Colombia. Vaccine
[22] Urdaneta M, Prata A, Struchiner CJ, Tosta CE, Tauil P, Boulos M.
Evaluation of SPf66 malaria vaccine efficacy in Brazil. Am J Trop
Med Hyg 1998;58(3):378–85.
[23] Genton B, Betuela I, Felger I, Al-Yaman F, Anders RF, Saul
A, et al. A recombinant blood stage malaria vaccine reduces
Plasmodium falciparum density and exerts selective pressure on
parasite populations in a phase 1-2b trial in Papua New Guinea. J
Infect Dis 2002;185(6):820–7.