Critique of World Health Organization Recommendation of a Dengue Vaccine

Journal of Infectious Diseases, Dec 2016

Scott B. Halstead

A PDF file should load here. If you do not see its contents the file may be temporarily unavailable at the journal website or you do not have a PDF plug-in installed and enabled in your browser.

Alternatively, you can download the file locally and open with any standalone PDF reader:

http://jid.oxfordjournals.org/content/214/12/1793.full.pdf

Critique of World Health Organization Recommendation of a Dengue Vaccine

JID Critique of World Health Organization Recommendation of a Dengue Vaccine Scott B. Halstead 0 0 Department of Preventive Medicine and Biometrics, Uniformed Services University of the Health Sciences , Rockville, Maryland PERSPECTIVE dengue virus; vaccine; dengue vaccine; pathogenesis; antibody dependent enhancement (ADE); safety; vaccine adverse event - In mid-April 2016, the World Health Organization’s (WHO’s) Scientific Advisory Group of Experts on Immunization (SAGE) recommended that dengueendemic countries consider using Sanofi Pasteur’s chimeric yellow fever dengue vaccine (Dengvaxia) to immunize individuals aged 9–45 years in populations with high levels of dengue endemicity [1]. Dengvaxia was made by splicing yellow fever virus 17D genes with those of the 4 dengue virus (DENV) serotypes and is administered in 3 doses over 1 year [2]. The WHO’s Dengue Vaccine Working Group developed recommendations to the SAGE based on published year 1–3 data from phase 3 clinical trials involving >35 000 children aged 2–16 years in 10 dengue-endemic countries [3–6]. These data were supplemented with unpublished data supplied by the manufacturer from up to 6 years after vaccination [7]. During the first 2 years after immunization, compared with placebo controls, Dengvaxia reduced the prevalence of dengue, mild and severe, by 57%, with a lower efficacy against illnesses caused by DENV-1 and DENV-2, compared with DENV-3 and DENV-4 [3–5]. However, during the third year after vaccination, the protective efficacy dropped to 16.7% (65 cases among 22 177 vaccine recipients vs 39 cases among 11 089 placebo recipients) [6]. An analysis of year 3 breakthrough dengue found the vaccine to be asymmetrically protective and enhancing [8]. In Asian sites, the dengue hospitalization rate was significantly higher among vaccinated children aged ≤5 years (20 of 2029 [0.99%]) than among controls (2 of 1005 [0.2%]), with a relative risk of 4.95 (P = .03) [8]. Concerning hospitalizations in seronegative children aged ≥9 years, the SAGE wrote, “There are few data to support or refute any risk in seronegatives >9 years of age. . .. In CYD14 and CYD15, over 70% of the population in this age group was seropositive, and this increased with age up to 16 years. The relative risks were below 1 over time in this age group (consisting of both seropositives and seronegatives)” [7]. Because only 8%–19% of children enrolled in clinical trials underwent blood sampling before receiving the vaccine, complete data based on the serological status at the time of vaccination of all 65 hospitalized children are not available. In late 2015 and early 2016, Dengvaxia was licensed by the governments of Brazil, Costa Rica, El Salvador, Mexico, Paraguay, and the Philippines, where 1 million doses were purchased for nationwide vaccination of 9 year olds [9–11]. Although the SAGE concluded that Dengvaxia can be safely administered to seronegative individuals aged ≥9 years, several lines of evidence suggest that Dengvaxia, when given to seronegative individuals of any age, permitted hospitalizations due to breakthrough DENV infection during year 3. First, vaccination of seronegative individuals raises enhancing antibodies. Seronegative children composed a substantial portion of the total vaccinated individuals: the median DENV-seronegative prevalence varied from 54.2% among those aged 2–5 years to 21.5% among those aged 9–16 years [12]. These seronegative children responded to ≥1 dose of Dengvaxia by regularly developing DENV-1–4 neutralizing antibodies, although these children were poorly protected against dengue [3–5]. A similar protection outcome was observed in sera from 23 Singaporean adults who provided blood specimens 5 years after having received 3 phase 2 Dengvaxia doses. These individuals, who predominantly were seronegative when vaccinated, had low levels of circulating DENV neutralizing antibodies, which failed to protect mice against DENV-2 challenge [13]. The combination of poor protection against DENV infection of individuals with circulating DENV antibodies (monotypic immune equivalents) satisfies the known preconditions for antibody-dependent enhancement of infection [14]. Second, during year 3, based on serological status and DENV infection rates measured in children in Asian vaccination sites, 20 hospitalized 2–5-year-old children were estimated among 176 seronegative vaccinated children who experienced their first DENV infection, a hospitalization rate of 11.4% [8, 12]. In the corresponding placebo population, it was estimated that there were 60 individuals with DENV monotypic immunity who experienced secondary DENV infection, of whom 2 (3.3%) were hospitalized, which similar to hospitalization rates reported for second DENV infections in many prospective cohort studies [8, 14]. This hospitalization rate ratio of 3.5 may persist into older age groups, although with rates dictated by age-specific risk to the DENV vascular permeability syndrome [8, 15]. Third, hospitalizations in vaccinated and placebo groups occurred in different populations. Among placebo recipients, children aged ≥9 years were hospitalized nearly twice as frequently (3.1 hospitalizations/1000 placebo recipients; 27 hospitalizations/17 346 recipients) as those aged 2–5 years (1.6/1000; 27/8703) [6]. This higher hospitalization rate for older children, compared with younger children, corresponds to the age-specific hospitalization rates due predominantly to secondary DENV infections in open populations in Asia, peaking at 9–11 years [8, 16]. In the vaccine cohort, children aged 2–5 years were hospitalized at a rate of 9.9 events/1000 children (20 hospitalizations/2029 vaccine recipients), compared with 1.6/1000 for children aged ≥9 years (27/17346) [6]. This 6-fold difference in the hospitalization rate among younger children, compared with older children, mirrors the higher intrinsic susceptibility to hospitalizations during secondary DENV-2 infections in Cuba of young children, compared with older children [15]. Fourth, standard vaccine efficacy calculations compare disease attack rates in placebo groups to those in vaccinated groups. But when a vaccine has the potential to sensitize a recipient to immune enhancement, a classical efficacy calculation is not relevant. In the vaccine group, seronegative individuals may be converted by vaccine to so-called monotypic-immune equivalents who experience disease/hospitalizations during breakthrough primary DENV infections, whereas hospitalizations in the placebo groups occur during secondary DENV infections. Thus, DENV hospitalizations in vaccinees and placebo recipients occur in completely different denominator groups. These rates cannot be compared. Fifth, the observed increased rate of hospitalization among vaccinated 2–5year-old children was labeled by the SAGE as a “theoretical risk,” causally attributed among other possibilities to “temporal clustering” [7]. However, in Cuba, following transmission of DENV1 in 1977, a clustered outbreak of DENV-2 in 1981 produced hospitalizations during secondary infections, while seronegative children experienced unapparent or mild disease [17]. It must be concluded that, among seronegative vaccinees, breakthrough dengue disease, and certainly hospitalizations, that occur at rates significantly greater than those during primary dengue DENV infections when normalized for age should be scored as serious adverse events. Clustering of cases due to the vaccination of large numbers in cohorts is irrelevant. DISCUSSION An important element contributing to the SAGE’s conclusion on Dengvaxia safety was that the risk of hospitalization for 2–5-year-old children peaked at year 3 and subsequently diminished or “dissipated.” This conclusion is refuted by data from year 6 of the phase 2b trial. Table 8 from the Background Paper on Dengue Vaccines shows that there were 37 hospitalizations among 2131 vaccinated children (compared with 13 among 1072 placebo recipients) [7]. The rates by age group when vaccinated are 2.8% (11 of 393) among children aged 2–5 years:, 1.6% (15 of 945) among those aged 6–8 years, and 1.4% (11 of 793) among those aged ≥9 years (denominators are from Table 1 in the article by Hadinegoro et al [6]). Calculations for vaccinated susceptible individuals would yield rates that are 2–5 times higher. The year 6 overall hospitalization rate of 1.7% (37 of 2131) is higher than that for year 3 (1.0% [22 of 2131]). Hospitalizations are not dissipating. This increase in the hospitalization rate may parallel increases in dengue severity observed with increasing intervals between the first and second DENV infections [18]. While the SAGE has recognized that “the greatest benefit would be expected in those who are seropositive at the time of vaccination,” it was concluded that, “[c]urrently, there is no rapid, point-ofcare test to establish serostatus in order to allow for this kind of targeted vaccination,” and that “[a]ny requirement to conduct seroprevalence studies prior to vaccine implementation will be new to the EPI programme” [7, pp 33]. The SAGE thus failed to rise to the unique challenge posed by the need to identify and avoid the immunological risks imposed by the possible occurrence of vaccine-enhanced disease. Point-of-care lateral flow tests are on the market that would permit rapid identification of individuals with circulating anti-DENV immunoglobulin G (IgG) antibodies. Further, vaccine can be administered in a 2-step process whereby children have their enzyme-linked immunosorbent assay–based DENV IgG status determined to be deemed vaccine eligible or ineligible. The report notes that “a statistically significant increased risk of hospitalized dengue was seen in (2–5 year-old) vaccine recipients in the third year after the first dose” [7, pp 3] but concluded simply that “this increased risk is currently not understood” [7, pp 282]. Why not? The possibility that vaccine-enhanced disease occurred in individuals older and younger than 9 years should have been subjected to careful scrutiny prior to vaccine licensure. To overcome the vacuum left by incomplete serological sampling before vaccination, the immune status at the time of vaccination can be identified retrospectively in vaccinated children who acquire a subsequent DENV infection. For example, the presence of anti-DENV envelope IgG but the absence of dengue NS1 IgG antibodies in late convalescent sera characterizes a first DENV infection on a background of Dengvaxia yellow fever vaccine–derived immunity. Clarifying how populations with different immunological backgrounds handle DENV infection after Dengvaxia administration is crucial for regulatory authorities who must direct Dengvaxia delivery, whether via public or private channels, to optimize benefit and minimize risk. The SAGE Dengue Vaccine Working Group should consider the special assessments required to identify viral vaccine safety and measure efficacy in which the vaccine itself may sensitize to antibody-dependent enhancement of infection. Potential conflict of interest. S. B. H. certifies no potential conflict of interest. The author has submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed. References 1. World Health Organization . SAGE meeting of April 2016 . http://www.who.int/immunization/ sage/meetings/2016/april/en/. Accessed 22 April 2016 . 2. Guy B , Briand O , Lang J , Saville M , Jackson N. Development of the Sanofi Pasteur tetravalent dengue vaccine: one more step forward . Vaccine 2015 ; 33 : 7100 - 11 . 3. Sabchareon A , Wallace D , Sirivichayakul C , et al. Protective efficacy of the recombinant, live-attenuated , CYD tetravalent dengue vaccine in Thai schoolchildren: a randomised, controlled phase 2b trial . Lancet 2012 ; 380 : 1559 - 67 . 4. Capeding MR , Tran NH , Hadinegoro SR , et al. Clinical efficacy and safety of a novel tetravalent dengue vaccine in healthy children in Asia: a phase 3, randomised, observer-masked, placebo-controlled trial . Lancet 2014 ; 384 : 1358 - 65 . 5. Villar L , Dayan GH , Arredondo-Garcia JL , et al. Efficacy of a tetravalent dengue vaccine in children in Latin America . N Engl J Med 2014 ; 372 : 113 - 23 . 6. Hadinegoro SR , Arredondo-Garcia JL , Capeding MR , et al. Efficacy and long-term safety of a dengue vaccine in regions of endemic disease . N Engl J Med 2015 ; 373 : 1195 - 206 . 7. SAGE Working Group on Dengue Vaccines and WHO Secretariat. Background paper on dengue vaccines . Geneva: World Health Organization , 2016 . 8. Halstead SB , Russell PK . Protective and immunological behavior of chimeric yellow fever dengue vaccine . Vaccine 2016 ; 34 : 1643 - 7 . 9. Sanofi Pasteur . Dengvaxia, world's first dengue vaccine, approved in Mexico [news release] . 9 December 2015 . http://www.sanofipasteur. com/en/articles/ dengvaxia-world-s-first-dengue-vaccine-approvedin-mexico .aspx. Accessed 24 December 2015 . 10. Herriman R. El Salvador becomes the 4th country to approve dengue vaccine . Outbreak News Today. Posted 11 February 2016 . http://outbreaknewstoday. com/ el-salvador-becomes-the-4th-country-to-approvedengue-vaccine-53510/ . Accessed 14 February 2016 . 11. Maron DF . First dengue fever vaccine gets green light in 3 countries . Scientific American . 30 December 2015 . http://www.scientificamerican. com/ article/first-dengue-fever-vaccine-gets-green-lightin-three-countries/ . Accessed 5 January 2016 . 12. Coudeville L , Baurin N , Vergu E. Estimation of parameters related to vaccine efficacy and dengue transmission from two large phase III studies . Vaccine 2016 ; In press. 13. Velumani S , Toh YX , Balasingam S , et al. Low antibody titers 5 years after vaccination with the CYD-TDV dengue vaccine in both pre-immune and naive vaccinees . Hum Vaccin Immunother 2016 ; 12 : 1265 - 73 . 14. Halstead SB . Neutralization and antibody dependent enhancement of dengue viruses . Adv Virus Res 2003 ; 60 : 421 - 67 . 15. Guzman MG , Kouri G , Bravo J , Valdes L , Vazquez SHalstead SB . Effect of age on outcome of secondary dengue 2 infections . Intern J Infect Dis 2002 ; 6 : 118 - 24 . 16. Anders KL , Nguyet NM , Chau NV , et al. Epidemiological factors associated with dengue shock syndrome and mortality in hospitalized dengue patients in Ho Chi Minh City, Vietnam . Am J Trop Med Hyg 2011 ; 84 : 127 - 34 . 17. Guzman MG , Kouri GP , Bravo J , Soler M , Vazquez SMorier L. Dengue hemorrhagic fever in Cuba, 1981: a retrospective seroepidemiologic study . Am J Trop Med Hyg 1990 ; 42 : 179 - 84 . 18. Guzman MG , Kouri G , Valdes L , Bravo J , Vazquez SHalstead SB . Enhanced severity of secondary dengue 2 infections occurring at an interval of 20 compared with 4 years after dengue 1 infection . PAHO J Epidemiol 2002 ; 81 : 223 - 7 .


This is a preview of a remote PDF: http://jid.oxfordjournals.org/content/214/12/1793.full.pdf

Scott B. Halstead. Critique of World Health Organization Recommendation of a Dengue Vaccine, Journal of Infectious Diseases, 2016, 1793-1795, DOI: 10.1093/infdis/jiw340