Villar, Luis Á. MD*; Rivera-Medina, Doris M. MD†; Arredondo-García, José Luis MD‡; Boaz, Mark PhD§; Starr-Spires, Linda PhD§; Thakur, Manoj MS§; Zambrano, Betzana MD¶; Miranda, María C. MD‖; Rivas, Enrique MD**; Dayan, Gustavo H. MD§
The dengue virus is a member of the Flavivirus (FV) genus, which also includes the yellow fever (YF) virus.1 Dengue disease is caused by any of 4 dengue virus serotypes (serotypes 1–4) and is transmitted primarily by the Aedes aegypti mosquito.2 Most dengue virus infections are asymptomatic, but clinical manifestations may range from mild febrile illness to severe and potentially fatal disease.2
Dengue disease is endemic in the Caribbean, Central and South America and Mexico, where the number of cases has increased dramatically in recent decades.3 This increase has been attributed to various factors, including deterioration of the A. aegypti eradication program, population growth, increased urbanization and climatic changes.3–5 In 2011, nearly 1.1 million cases of dengue disease, including over 19,000 cases of severe dengue and more than 700 dengue-related deaths, were reported in Latin America.6
No specific treatment exists for dengue disease, and preventive measures, which include mosquito control and personal protection from bites, have limited effectiveness. A recombinant, live-attenuated, tetravalent dengue vaccine (CYD-TDV; Sanofi Pasteur, Lyon, France) is under development and contains 4 recombinant viruses (CYD 1–4), each of which has genes encoding the dengue premembrane and envelope proteins of 1 of the 4 dengue serotypes and genes encoding the nonstructural and capsid proteins of the attenuated YF-17D vaccine virus.7,8 The safety and immunogenicity of CYD-TDV have been established in various populations and age ranges.9–17 The primary objectives of this study were to evaluate the safety and immunogenicity of CYD-TDV in 9–16 year olds in dengue endemic countries in Latin America.
MATERIALS AND METHODS
Trial Design and Participants
This randomized, observer-blind (first and second injections), single-blind (third injection), controlled trial was conducted in Colombia, Honduras, Mexico and Puerto Rico (National Clinical Trials Identifier: NCT00993447). Healthy volunteers aged 9–16 years were randomized 2:1 to receive 3 subcutaneous injections of CYD-TDV or a control (placebo [0.9% NaCl] as the first and second subcutaneous injections and a tetanus/diphtheria/acellular pertussis vaccine as the third intramuscular injection). Injections were administered at 0, 6 and 12 months. Group allocation was performed using an interactive voice response system based on randomization lists generated using the block method with stratification by center.
The major exclusion criteria were any immunodeficiency or chronic illness or treatment that could interfere with the trial conduct or results; known systemic hypersensitivity to any component of the trial vaccines; receipt of any vaccine in the 4 weeks preceding the first trial dose or planned vaccination in the 4 weeks following the first trial dose. Pregnant or breast-feeding women were excluded.
Girls of child-bearing potential were required to abstain from sexual intercourse or use an effective method of contraception for at least 4 weeks before the first dose until at least 4 weeks after the last dose.
The study was conducted in accordance with the Declaration of Helsinki (Edinburgh, October 2000), Good Clinical Practice, International Conference on Harmonization guidelines and national and local ethical requirements. Written informed consent was obtained from participants and/or a parent or legal guardian.
CYD-TDV was supplied as a powder for suspension in solvent (0.4% NaCl containing 2.5% human serum albumin). The vaccine suspension contained approximately 5 log10 50% cell culture infectious dose of each of the 4 live-attenuated recombinant dengue viruses per 0.5 mL dose. The active control vaccine was a licensed tetanus/diphtheria/acellular pertussis booster vaccine (Adacel; Sanofi Pasteur, Swiftwater, PA) that was provided ready for use.
Safety and Reactogenicity
Adverse events (AEs) documented were AEs occurring within 30 minutes after each dose; solicited injection-site reactions (pain, erythema, swelling) up to 7 days after each dose; solicited systemic reactions (fever, headache, malaise, myalgia, asthenia) up to 14 days after each dose; unsolicited AEs up to 28 days after each dose; serious AEs (SAEs) occurring at any time until 6 months after the last dose. An independent data monitoring committee monitored safety during the study.
In the event of fever (temperature ≥38°C) for ≥48 hours within 28 days after any dose, CYD-TDV viremia was investigated by quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) using methods similar to those reported elsewhere.18
Serum levels of neutralizing antibodies against each of CYD-TDV’s 4 dengue serotype parental strains were determined before and 28 days after each dose using a 50% plaque reduction neutralization test19,20 (TM Timiryasova, MI Bonaparte, P Luo, R Zedar, BT Hu and SW Hildreth, article submitted). Serial 2-fold dilutions of heat-inactivated serum were mixed with a constant challenge dose of each dengue virus. The mixtures were inoculated into plate wells confluent with Vero cell monolayers. After adsorption, the cell monolayers were incubated for a few days. Plaques in each well, which indicated the presence of cells infected with dengue virus, were fixed, immunostained and counted. Titers were expressed as the highest reciprocal serum dilution at which the mean number of plaques was reduced by 50% compared with the count obtained from control wells. Geometric mean titers (GMTs) were calculated. A 50% plaque reduction neutralization test assay was used to assess YF seropositivity in the first blood sample. This assay was also performed in Vero cells and used the YF-17D strain with crystal violet staining for enumeration. Seropositivity was defined as antibody titers ≥10 1/dil.
Participants were classified as FV-seropositive or FV-seronegative at baseline. FV-seropositive individuals had dengue and/or YF antibody titer(s) ≥10 1/dil, whereas FV-seronegative individuals had dengue and YF antibody titers <10 1/dil.
Surveillance and Confirmation of Symptomatic Dengue
Active detection of dengue disease was based on planned contacts (≥1 per week) and/or follow-up of school absentees. Suspected cases (temperature ≥38°C for ≥48 hours and no evidence of local infection) were tested for dengue virus in an acute serum sample (taken within 7 days of fever onset). Confirmed dengue cases were those with wild-type (WT) dengue virus confirmed virologically by non-structural protein 5 serotype-specific WT quantitative RT-PCR (qRT-PCR) or by dengue nonstructural protein 1 antigen enzyme-linked immunosorbent assay (Platelia, Biorad Laboratories, Marnes-La-Coquette, France). For the qRT-PCR, a screening assay to detect any WT dengue virus targeting the 3′-UTR region was applied before the serotype-specific WT qRT-PCRs.21 Additional exploratory serotyping was performed using tests including a SYBR Green-based qRT-PCR, Simplexa dengue qRT-PCR (Focus Diagnostics, Inc. Cypress, CA) and sequencing assays directed at the PCR-targeted 3′UTR and non-structural protein 5 regions.
Probable cases were those that were positive for dengue immunoglobulin (Ig) M in acute or convalescent (taken 7–14 days after the acute sample) serum samples or those demonstrating a 4-fold increase in dengue IgG antibody titers (EL1500M and EL1500G, respectively, Focus Diagnostics Inc., Cypress, CA) between acute and convalescent samples. As these tests cannot distinguish between antibody responses to the vaccine or natural infection, cases that occurred within 28 days after an injection were not assessed for dengue based on serology. The independent data monitoring committee assessed virologically confirmed cases for severity.
The planned sample size was set arbitrarily at 600 volunteers. With a sample size of 400 participants in the CYD-TDV group, there was a 95% probability of observing an event that had a true incidence of 0.75% in this group. Analyses were descriptive with no hypothesis testing. For the main parameters, 95% confidence intervals of point estimates were calculated using normal approximation for quantitative data and exact binomial distribution (Clopper–Pearson method)22 for proportions. Reverse cumulative distribution curves at baseline and postdose 3 vaccination antibody titers were derived to illustrate immune responses to each serotype.
Between October 2009 and February 2010, 600 volunteers aged 9–16 years were enrolled, randomized and received the first injection (CYD-TDV group, n = 401; control group, n = 199). The numbers of participants enrolled in Colombia, Honduras, Mexico and Puerto Rico were 159, 158, 177 and 106, respectively. The male-to-female ratio was balanced in the CYD-TDV group (49.1% male). In the control group, there was a slightly higher proportion of females (54.3%) than males (45.7%). Mean age was similar for the CYD-TDV and control groups (12.6 [range 9.0–17.0] vs. 12.5 [range 9.0–16.8] years, respectively). The majority of participants in both groups were FV-seropositive at baseline (CYD-TDV 78.8%; control 80.4%). For dengue 75.1% and 77.9% of participants in the CYD-TDV and control groups, respectively, were seropositive for at least 1 serotype. For YF, 70.1% and 73.4%, respectively, were seropositive. The study was completed at day 28 after the third dose by 90% of participants in each group (Fig. 1).
Safety and Reactogenicity
Ten (2.5%) participants in the CYD-TDV group and 15 (7.5%) in the control group experienced SAEs (Fig. 2), none of which were considered vaccination related. All participants with SAEs recovered and continued in the study. SAEs in the CYD-TDV group were dengue disease (n = 3, see below), appendicitis (n = 3), pyelonephritis (n = 1), urinary tract infection (n = 1), accidental poisoning (n = 1) and epilepsy in a subject with personal and family history of seizures (n = 1). One participant in the CYD-TDV group was withdrawn from the study 2 days after the first dose following a diagnosis of hepatitis A that was not considered vaccination related. One participant in the control group experienced an unsolicited systemic AE (dizziness) within 30 minutes after the first dose of placebo. No notable differences in the types or rates of unsolicited AEs were observed between the groups (Fig. 2).
Most solicited reactions were mild or moderate in severity (Table 1), appeared between days 0 and 3 postdose and resolved within 3 days. Overall, comparable or slightly higher rates of solicited injection-site and systemic reactions were observed with CYD-TDV compared with placebo after the first and second doses (Table 1 and Fig. 2). After the third dose, rates of solicited reactions were lower with CYD-TDV than with tetanus/diphtheria/acellular pertussis vaccine. Fewer injection-site and systemic reactions were reported after the second and third doses of CYD-TDV compared with the first dose (Table 1 and Fig. 2).
The overall reactogenicity of CYD-TDV was not markedly different between the baseline FV-seronegative (n = 316) and FV-seropositive (n = 85) subsets. Systemic reaction reporting rates were 61.9%, 41.0% and 37.7% for FV-seronegative participants and 56.8%, 47.1% and 39.7% for FV-seropositive participants after the first, second and third doses, respectively. For injection site reactions, these reporting rates were 38.1%, 42.3% and 28.6% for FV-seronegative participants and 29.9%, 23.4% and 23.7%for FV-seropositive participants.
Fever lasting ≥48 hours within 28 days of injection was reported for 16 participants in the CYD-TDV group and 12 participants in the control group. No CYD-TDV viremia was detected.
GMTs of antibodies at baseline were similar for the CYD-TDV and control groups for all 4 serotypes (Table 2). GMTs increased after each dose of CYD-TDV for serotypes 1 and 3 and after first and second doses for serotype 2. For serotype 4, GMTs were highest after the first dose of CYD-TDV. GMTs in the CYD-TDV group were higher than at baseline at all time-points for all 4 serotypes. The GMT ratio from baseline to postdose 3 ranged from 3.2 for serotype 1 to 5.5 for serotype 4. GMTs differed depending on baseline FV immune status. GMTs increased mainly after the first dose among participants who were FV-seropositive at baseline; however, a more gradual increase after each dose was observed for serotypes 1–3 among those who were FV-seronegative. The GMTs postdose 3 for serotypes 1–4, respectively, were 580, 741, 827 and 341 for participants who were FV-seropositive at baseline and 34.6, 101, 174 and 119 for those who were FV-seronegative (Fig. 3). There were no appreciable changes in GMTs after any dose in the control group.
Among all subjects, seropositivity rates against dengue at baseline were 64.1%, 69.3%, 69.6% and 62.6% for serotypes 1–4, respectively (Fig. 4). Seropositivity rates increased after each CYD-TDV injection. After the third injection, serotype-specific seropositivity rates were 94.2% or higher, and 100%, 98.6% and 93.4% of participants were seropositive for at least 2, at least 3 and all 4 serotypes, respectively. There was no appreciable change in the seropositivity rate after any dose in the control group.
After the third dose of CYD-TDV, seropositivity rates were ≥95% for serotypes 2–4 irrespective of baseline FV immune status. For serotype 1, the seropositivity rate was 97.6% for those who were FV-seropositive at baseline and 81.8% for those who were FV-seronegative. After the third dose of CYD-TDV, similar seropositivity rates (≥95%) for at least 1, 2 or 3 serotypes were observed among participants who were FV-seropositive or FV-seronegative at baseline. However, a lower seropositivity rate for all 4 serotypes was observed in FV-seronegative participants (77.9%) compared with those who were FV-seropositive (97.6%).
Dengue Disease During the Study
Dengue was suspected for 43 of 401 participants in the CYD-TDV group and 29 of 199 in the control group. Virological analyses planned in the protocol confirmed dengue infection in 14 of these suspected cases (6/401, 1.5% in the CYD-TDV group, and 8/199, 4.0% in the control group) (Table 3). The dengue serotype was identified by qRT-PCR (ie, according to protocol) in 9 of 14 cases: 5 were serotype 1 and 4 were serotype 3.
Exploratory analyses identified the serotype in the remaining 5 of 14 virologically confirmed cases and in 5 additional cases that screened positive for dengue but were not considered virologically confirmed according to the criteria defined in the protocol. Of these cases serotyped by exploratory analyses, 1 was serotype 1, 7 were serotype 2 and 2 were serotype 4 (Table 3).
The distribution of serotypes by country and study group was as follows: 6 serotype 1 cases occurred in Colombia (CYD-TDV group, n = 2), Honduras (CYD-TDV, n = 1; control group, n = 1), Mexico (control group, n = 1) and Puerto Rico (CYD-TDV group, n = 1); 7 serotype 2 cases occurred in Honduras (CYD-TDV group, n = 2; control group, n = 5); 4 serotype 3 cases occurred in Colombia (CYD-TDV group, n = 2; control group, n = 2); and 2 serotype 4 cases occurred in Colombia (both in the control group).
Considering the 19 cases that were either virologically confirmed according to the protocol or serotyped in exploratory analysis, 8 occurred in the CYD-TDV group (8/401, 2%) and 11 occurred in the control group (11/199, 5.5%) (Table 4). Ten probable cases of dengue disease (10/401, 2.5%) were detected by serological criteria in the CYD-TDV group and 13 (13/199, 6.5%) in the control group. The independent data monitoring committee assessed only 2 cases in the control group as severe.
This was the first multicenter phase II study of the CYD-TDV candidate vaccine in Latin America and was conducted in a predominantly FV-seropositive population of children and adolescents from 4 countries. The study showed that 3 doses of vaccine had a favorable safety profile compared with a licensed control vaccine and elicited neutralizing antibody responses against all 4 dengue serotypes.
Previous phase I and II studies have established that a 3-dose regimen of CYD-TDV induces a balanced neutralizing antibody response with a favorable safety profile, in different populations and age ranges, in both FV endemic and nonendemic countries, including Peru and Mexico.9–17 The first studies to enroll children and adolescents used a 0–3.5–12 month vaccination schedule and were conducted in dengue endemic (Philippines) and nonendemic (Mexico City) areas.14,15 Based on these early studies, the 0–6–12 month vaccination schedule was selected. The GMTs observed in our study after 3 doses of CYD-TDV were higher than those reported for adolescents from studies in endemic areas of Asia (Philippines 131–28714; Singapore 28.5–79.217). Although the reasons for this are unclear, it may be related to higher baseline FV seropositivity rates and GMTs in our study than among adolescents in the Asian studies. Although neutralizing antibody responses to CYD-TDV were detected, there is no known correlate of protection for dengue, and it is not known to what extent antibodies detected by 50% plaque reduction neutralization test relate to protection against dengue infection or disease. Furthermore, the level of antibodies predictive of protection may not be the same for the 4 dengue serotypes.
Nearly 80% of our study population was FV-seropositive at baseline, suggesting natural exposure to either dengue, other flaviviruses or YF vaccination. However, as YF vaccination is only routinely given in Colombia, where the virus is endemic, the observed high YF seropositivity rates at baseline may be related in part to cross-reactivity between dengue and YF antibodies that has been described in the literature.23–25 Higher dengue antibody titers after CYD-TDV vaccination were observed in participants who were FV seropositive at baseline, suggesting that preexisting FV antibody responses have a beneficial impact on the vaccine-induced antibody response. Similar findings have been reported previously in a FV-seropositive population14 and participants who had been vaccinated against YF.11 Comparisons of results of the present study on the basis of FV immune status at baseline should be interpreted with caution because of the relatively low number of FV-seronegative participants.
Slight increases in GMTs and seropositivity rates for dengue virus serotypes in the control group during the course of the present study likely reflect natural exposure to dengue virus and, indeed, we detected a number of cases of symptomatic infection. The majority of cases with serotype data were from Colombia (n = 8, serotypes 1, 3 and 4) and Honduras (n = 9, serotypes 1 and 2) where dengue epidemics were reported during the study period.26 The serotypes identified with either the per-protocol or exploratory methods are consistent with the predominant circulating serotypes in the respective countries at that time. The lack of detection of the circulating serotype 2 Nicaraguan strain in the routine qRT-PCR assay appears to be due to differences in this isolate’s sequence compared with the probes used in the assay (data not shown). The diagnostic testing algorithm would therefore need to be adjusted in future studies to ensure detection of this isolate by qRT-PCR. Although efficacy was not one of the study objectives, the lower incidence of virologically confirmed dengue in the vaccine group than in the control group, including against serotype 2, may suggest vaccine protection against disease. These results seem different to the ones obtained in a phase IIb proof-of-concept efficacy trial conducted in Thailand that provided evidence of protective efficacy against dengue serotypes 1, 3 and 427 but not against serotype 2. Differences in the epidemiology and virus strain circulation in Asia compared with Latin America may be potentially associated to such different results. An ongoing phase III trial (NCT01374516) in 5 countries in Latin America will provide pivotal data on the efficacy of CYD-TDV in the region.
In conclusion, this study establishes the safety and immunogenicity of CYD-TDV in 9–16 year olds in areas where dengue disease is endemic in Latin America. The findings of this study are consistent with observations from other phase I and II trials with CYD-TDV and support the continued development of CYD-TDV.
The authors take full responsibility for the content of this contribution and thank Sarah Whitfield (supported by Sanofi Pasteur) for assisting with preparation of article drafts. The authors would also like to thank Grenville Marsh at Sanofi Pasteur for providing critical comments and suggestions on the drafts. The authors would like to thank all of the volunteers and their parents/guardians who participated in the trial and the study-site personnel for their contributions to the study. In addition, thanks are due to the following people within Sanofi Pasteur: Dany de Grave and Branda Hu, for their contributions in the development and improvement of immunological assays, and the Global Clinical Immunology Department for conducting the immunological assays; Linda Urcuyo and Martin Sanchez-Ruiz for study management and logistics; and Fernando Noriega, Alain Bouckenooghe, Thomas Papa, Simonetta Viviani, Melanie Saville, Nadia Tornieporth and Jean Lang for their contributions to the study design and development; and Lixbeth Carmona for program management.
1. Burke DS, Monath TP. Knipe DM, Howley PM. Flaviviruses. Fields Virology. 2001; Vol 1:4th ed Philadelphia, PA Lippincott Williams & Wilkins 1043–1125
3. San Martín JL, Brathwaite O, Zambrano B, et al. The epidemiology of dengue in the americas over the last three decades: a worrisome reality. Am J Trop Med Hyg. 2010; 82:128–135
4. Hurtado-Díaz M, Riojas-Rodríguez H, Rothenberg SJ, et al. Short communication: impact of climate variability on the incidence of dengue in Mexico. Trop Med Int Health. 2007; 12:1327–1337
5. Barclay E. Is climate change affecting dengue in the Americas? Lancet. 2008; 371:973–974
7. Guirakhoo F, Weltzin R, Chambers TJ, et al. Recombinant chimeric yellow fever-dengue type 2 virus is immunogenic and protective in nonhuman primates. J Virol. 2000; 74:5477–5485
8. Guirakhoo F, Arroyo J, Pugachev KV, et al. Construction, safety, and immunogenicity in nonhuman primates of a chimeric yellow fever-dengue virus tetravalent vaccine. J Virol. 2001; 75:7290–7304
9. Guy B, Guirakhoo F, Barban V, et al. Preclinical and clinical development of YFV 17D-based chimeric vaccines against dengue, West Nile and Japanese encephalitis viruses. Vaccine. 2010; 28:632–649
10. Guy B, Barrere B, Malinowski C, et al. From research to phase III: preclinical, industrial and clinical development of the Sanofi Pasteur tetravalent dengue vaccine. Vaccine. 2011; 29:7229–7241
11. Qiao M, Shaw D, Forrat R, et al. Priming effect of dengue and yellow fever vaccination on the immunogenicity, infectivity, and safety of a tetravalent dengue vaccine in humans. Am J Trop Med Hyg. 2011; 85:724–731
12. Guirakhoo F, Kitchener S, Morrison D, et al. Live attenuated chimeric yellow fever dengue type 2 (ChimeriVax-DEN2) vaccine: Phase I clinical trial for safety and immunogenicity: effect of yellow fever pre-immunity in induction of cross neutralizing antibody responses to all 4 dengue serotypes. Hum Vaccin. 2006; 2:60–67
13. Morrison D, Legg TJ, Billings CW, et al. A novel tetravalent dengue vaccine is well tolerated and immunogenic against all 4 serotypes in flavivirus-naive adults. J Infect Dis. 2010; 201:370–377
14. Capeding RZ, Luna IA, Bomasang E, et al. Live-attenuated, tetravalent dengue vaccine in children, adolescents and adults in a dengue endemic country: randomized controlled phase I trial in the Philippines. Vaccine. 2011; 29:3863–3872
15. Poo J, Galan F, Forrat R, et al. Live-attenuated tetravalent dengue vaccine in dengue-naïve children, adolescents, and adults in Mexico City: randomized controlled phase 1 trial of safety and immunogenicity. Pediatr Infect Dis J. 2011; 30:e9–e17
16. Lanata CF, Andrade T, Gil AI, et al. Immunogenicity and safety of tetravalent dengue vaccine in 2-11 year-olds previously vaccinated against yellow fever: randomized, controlled, phase II study in Piura, Peru. Vaccine. 2012; 30:5935–5941
17. Sin Leo Y, Wilder-Smith A, Archuleta S, et al. Immunogenicity and safety of recombinant tetravalent dengue vaccine (CYD-TDV) in individuals aged 2–45 y: Phase II randomized controlled trial in Singapore. Hum Vaccin Immunother. 2012; 8:1259–1271
18. Mantel N, Aguirre M, Gulia S, et al. Standardized quantitative RT-PCR assays for quantitation of yellow fever and chimeric yellow fever-dengue vaccines. J Virol Methods. 2008; 151:40–46
19. Russell PK, Nisalak A, Sukhavachana P, et al. A plaque reduction test for dengue virus neutralizing antibodies. J Immunol. 1967; 99:285–290
20. Roehrig JT, Hombach J, Barrett AD. Guidelines for plaque-reduction neutralization testing of human antibodies to dengue viruses. Viral Immunol. 2008; 21:123–132
21. Huhtamo E, Hasu E, Uzcátegui NY, et al. Early diagnosis of dengue in travelers: comparison of a novel real-time RT-PCR, NS1 antigen detection and serology. J Clin Virol. 2010; 47:49–53
22. Clopper C, Pearson ES. The use of confidence or fiducial limits illustrated in the case of the binomial. Biometrika. 1934; 26:404–413
23. Campbell GL, Grady LJ, Huang C, et al. Laboratory testing for West Nile virus: panel discussion. Ann N Y Acad Sci. 2001; 951:179–194
24. Houghton-Triviño N, Montaña D, Castellanos J. Dengue-yellow fever sera cross-reactivity; challenges for diagnosis. Rev Salud Publica (Bogota). 2008; 10:299–307
25. Mansfield KL, Horton DL, Johnson N, et al. Flavivirus-induced antibody cross-reactivity. J Gen Virol. 2011; 92:(Pt 12)2821–2829
27. Sabchareon A, Wallace D, Sirivichayakul C, et al. Protective efficacy of the recombinant, live-attenuated, CYD tetravalent dengue vaccine in Thai school children: a randomised, controlled phase IIb trial. Lancet. 2012; 380:1559–1567