Complexity in Assessing the Benefit vs Risk of Vaccines - Experience With Rotavirus and Dengue

Complexity in Assessing the Benefit vs Risk of Vaccines - Experience With Rotavirus and Dengue Virus Vaccines. H. Cody Meissner. JAMA. October 17, 2019. doi:10.1001/jama.2019.16206

The remarkable contribution of immunization programs to public health is recognized by most people, including those who express vaccine hesitancy. Vaccine hesitancy often is predicated on the concern of an individual or family members regarding the risk of an adverse event following immunization. But for each licensed vaccine, the relative risk of an untoward event, such as contracting the disease, is greater among those who remain unvaccinated.1 Before US Food and Drug Administration (FDA) licensure, vaccine safety must be determined by demonstration that the benefit from disease prevention exceeds the risk of adverse reactions associated with the vaccine. However, this assessment is not always straightforward and the societal perspective of regulatory and advisory bodies may conflict with the individual perspective of the patient or parent. Experience with rotavirus vaccines and, more recently, the dengue virus vaccine provides insight into the complexity of this assessment.


Rotavirus Vaccine

Prior to licensure of the first oral rotavirus vaccine, rhesus-human rotavirus reassortant-tetravalent vaccine (RRV-TV), rotavirus gastroenteritis was associated with more than 50 000 hospitalizations annually and 20 to 60 deaths among children younger than 5 years in the United States. RRV-TV prevented 70% to 100% of severe infections and 48% to 68% of rotavirus diarrheal episodes.2 During prelicensure trials of RRV-TV, 5 cases of intussusception occurred among 10 054 vaccinated individuals and 1 case of intussusception occurred among 4633 unvaccinated control individuals, a difference that was not statistically significant.2 Considering the rate of intussusception among participants did not exceed the expected rate, the assessment was made by the FDA that an association between intussusception and vaccination was unlikely; nonetheless, intussusception was included as a potential adverse reaction in the package insert. RRV-TV was licensed and routine immunization was recommended by the Advisory Committee for Immunization Practices (ACIP) and the American Academy of Pediatrics (AAP) in August 1998. Less than 1 year later, in July 1999, 15 cases of intussusception occurring after immunization had been reported to the Vaccine Adverse Events Reporting System.3 When these results became available, the ACIP and the AAP stopped the recommendation for vaccine use and the manufacturer of RRV-TV voluntarily withdrew the vaccine from the market, despite the large benefit in disease prevention from the vaccine.3
Today it is recognized that the rate of intussusception is approximately 35 cases per 100 000 children younger than 1 year, and the 2 currently available rotavirus vaccines are estimated to result in an additional 1 to 5 cases of intussusception per 100 000 children younger than 1 year (about one-tenth to one-half of the rate of 10 per 100 000 individuals following administration of RRV-TV).3 Current data suggest that approximately 40 to 120 infants in the United States develop intussusception annually after administration of currently available rotavirus vaccines.3 Because of this low association with intussusception and the ability to manage most intussusception cases with nonsurgical procedures, the overall benefit in reduction of disease burden from immunization is judged by the ACIP and the AAP to exceed the small risk.
Based on the intention to offer the greatest benefit to the greatest number of people, continued use of RRV-TV in 1999 would have been more appropriate than discontinuation of the recommendation for vaccine use. Removal of the vaccine from the market resulted in more rotavirus-related illness for a larger number of infants and their family members than would have occurred with continued use of the vaccine. The more than 5-year interval between discontinuation of RRV-TV in July 1999 and availability of a human-bovine reassortant rotavirus vaccine in February 2006 likely resulted in more than 250 000 hospitalizations and 100 deaths related to rotavirus that could have been avoided with continued use of RRV-TV (based on a calculation of 5 times the number of annual cases in the prevaccine era). The number of additional cases of intussusception following continued use of RRV-TV would have been small relative to the reduction in the disease.


Dengue Vaccine

Dengue is a mosquito-borne disease caused by a flavivirus that has spread to most tropical and many subtropical countries. The disease is caused by 4 serotypes transmitted mainly by the bite of a female Aedes aegypti or Aedes albopictus mosquito. Although most dengue infections are asymptomatic, an estimated 100 million symptomatic infections occur annually resulting in 500 000 hospitalizations and 20 000 deaths worldwide, primarily in children.4 Dengue is endemic in certain territories of the United States including Puerto Rico, Guam, American Samoa, and the US Virgin Islands. No specific therapy is available for management of dengue and disease prevention has been limited to vector control.
In May 2019 the FDA licensed the first dengue vaccine (chimeric yellow fever–dengue, tetravalent dengue vaccine [CYD-TDV]), offering an important advance in disease control. Prelicensure trials in the flavivirus-endemic Latin America, Puerto Rico, and Asia-Pacific region demonstrated a vaccine efficacy of approximately 76% against virologically confirmed symptomatic dengue cases and approximately 80% against hospitalization in vaccinated individuals aged 9 through 16 years who received 3 doses of the vaccine and were seropositive at the time of vaccination.4 Among participants seronegative for dengue, vaccine effectiveness was 38%.4 In contrast to these benefits, an increased rate of severe dengue beginning about 30 weeks after the first dose occurred among a small number of vaccinated persons who were seronegative at the time of vaccination. Immunization with CVD-TDV of an immunologically naïve person created a similar immune response to that of a seronegative person infected with a dengue virus. A second infection by a different strain is associated with a risk of immune enhancement, an incompletely understood immunopathologic response resulting in an increased risk of dengue hemorrhagic fever and dengue shock syndrome.5 The rate of hospitalization due to dengue virus infection among children 9 through 16 years of age who were seronegative at time of vaccination was 1.09% among controls and 1.57% among vaccine recipients with a hazard ratio of 1.41 (95% CI 0.74-2.68.6 Based on these data, the Vaccines and Related Biologic Advisory Committee recommended licensing of CYD-TDV only for individuals in the well-studied age group (9-16 years) living in an endemic area with a laboratory documented history of dengue infection.
Prelicensure trials demonstrated that approximately 20% of participants aged 9 to 26 years living in dengue endemic areas were seronegative for dengue at baseline. Reliable identification of individuals seronegative for dengue is difficult in regions where multiple flaviviruses circulate because of cross-reacting antibodies and limited access to a reliable serologic assay. Therefore, a recommendation for routine use of CYD-TDY without screening will result in inadvertent vaccination of some seronegative people.
The World Health Organization (WHO) recommends prevaccination screening before vaccine administration. If prescreening is not feasible, WHO suggests consideration of routine use of CYD-TDV without prescreening in countries where seropositivity for dengue is greater than 80% by 9 years of age.5 One estimate of attributable risk for a cohort of 1 million people aged 9 through 16 years with greater than 80% rate of seropositivity for dengue over 5 years projected that 12 000 hospitalizations would be avoided but an additional 1000 hospitalizations would occur, primarily among individuals who were seronegative for dengue at the time of vaccination.6 Using this estimate, a recommendation for use of CVD-TDV in areas of high dengue seroprevalence where reliable serologic testing is not available could potentially benefit as many as 12 times more people than those who may experience dengue as a consequence of vaccination.
At what threshold does a large reduction in disease burden for society justify the small risk of an adverse reaction to an individual? Considerations by the FDA when licensing a candidate vaccine and by groups that issue vaccine recommendations include the prevalence and severity of the disease being prevented, the frequency and severity of an adverse reaction following immunization, and the effectiveness of the vaccine in disease prevention in the intended target population. But considerations may differ for individuals who are more concerned about an adverse reaction developing in themselves or their children and less concerned about societal benefits. They may be misinformed or unaware that remaining unvaccinated places their child at greater risk of disease than the risk of an adverse event following vaccination. Parents who choose not to immunize their children do not fulfill their parental obligation to protect children from disease. For instance, in California, because of increasing numbers of children who were incompletely vaccinated, legislation was enacted in 2016 to eliminate personal belief exemptions to vaccination. This legislation together with educational programs has been associated with a decline in the risk of a kindergarten student having contact with an inadequately vaccinated classmate from 10% in 2013 to 5% in 2017, enhancing the health of all students.7
Although risk of a rare vaccine reaction cannot be eliminated completely, well-established monitoring and compensation programs address this issue. Surveillance for safety of all licensed vaccines is monitored in a constant process. The United States uses several programs to monitor vaccine safety, including the Vaccine Adverse Events Reporting System, the Vaccine Safety Datalink, the Post-licensure Rapid Immunization Safety Monitoring System, and the Clinical Immunization Safety Assessment Project. Any untoward event reported after vaccination is assessed carefully for a causal relationship. The Vaccine Injury Compensation Program ensures equitable and timely compensation to those who may be injured by a vaccine.8 Although these programs may not reassure all vaccine-hesitant individuals, greater awareness of these programs is important. Balancing the potential of current and future vaccines in regard to disease control (even disease elimination has been achieved globally for wild polio virus types 2 and 3) vs an untoward event is overwhelmingly favorable for individuals and society.

References and full text at the DOI above.