Fast-Track Zika Vaccine Development — Is It Possible?

Thursday, 29th of September 2016

This discussion is important, especially if other practical preventives against Zika virus cannot be found.

Perspective

Fast-Track Zika Vaccine Development — Is It Possible?

Best viewed at http://www.nejm.org/doi/full/10.1056/NEJMp1609300?query=featured_home

Stephen J. Thomas, M.D., Maïna LAzou, M.Sc., Alan D.T. Barrett, Ph.D., and Nicholas A.C. Jackson, Ph.D.

N Engl J Med 2016; 375:1212-1216September 29, 2016DOI: 10.1056/NEJMp1609300

Studies have demonstrated that various Zika virus (ZIKV) vaccine constructs generate protective immune responses in mice and nonhuman primates,1,2 and two DNA ZIKV vaccine candidates have entered phase 1 human safety testing (ClinicalTrials.gov numbers, NCT01099852 andNCT02840487). ZIKV vaccine development is advancing rapidly thanks to collaborations among academia, governments, and industry. Current knowledge gaps related to the properties, epidemiology, and pathology of ZIKV increase the complexity of vaccine development (see Table 1TABLE 1Current Knowledge of Zika Virus (ZIKV) Epidemiology and Related Questions for Vaccine Development.), but historical success in developing other flavivirus vaccines encourages optimism.

An ongoing epidemic in the Americas and the impact of ZIKV congenital syndrome (ZCS) necessitate rapid development of a safe, efficacious vaccine. As Ebola vaccine–development efforts taught us, conducting sequential, iterative preclinical studies followed by phase escalating human trials is suboptimal in an ongoing outbreak. Preclinical studies of ZIKV vaccine candidates need to continue in parallel with human trials, informing their design and the evolving target product profile (TPP), including dose level and schedule, delivery method, and primary vaccinee population. Newly minted ZIKV vaccinologists need to determine which questions will inform development plans (see Table 2TABLE 2Zika Virus (ZIKV) Vaccine Challenges.).

Defining the TPP of a vaccine for emergency or conditional use has been a complex exercise, and the World Health Organization (WHO) has made a proposal (www.who.int/immunization/research/development/zika/en). Considerations include indications for male and female vaccinees, a short immunization schedule, brisk induction of a protective immune response, an advantageous safety profile, and potential contraindications in pregnancy.

Though live or other replicating virus vaccine platforms would probably be less acceptable in emergencies, they might offer advantages for routine immunization, such as long-term protective immunity with minimal dosing. Given the need to protect girls before they reach childbearing age, a TPP should address vaccination starting at 9 years of age and in people of both sexes, given evidence of ZIKV in semen up to 6 months after infection.3This starting age would align with WHO recommendations3 and with the precedent set by human papillomavirus vaccines (target group, girls 9 to 13 years old).

Prospective cohort studies can elucidate infection and disease attack rates, incidence of adverse pregnancy or neurologic outcomes, and contributions of different transmission modes to the overall epidemiology, helping to define routine immunization strategies. However, they wont provide information in the short term. Developers of first-generation vaccine candidates will need to measure performance against historical data and experiences with licensed flavivirus vaccines.

Preclinical development investigates vaccine safety, immunogenicity, and potential efficacy in animal models. Zikas interaction with other flaviviruses, role in pregnancy, and neurologic effects must be explored, and researchers are studying mouse and nonhuman primate ZIKV disease models in an effort to do so. Models in which sexual transmission (intravaginal challenges), maternofetal infection, and ZCS are reproduced will allow vaccine developers to assess the potential for preventing or attenuating infection or disease, the safety of vaccination during pregnancy, and the ability to protect the fetus or newborn from ZCS; there is no animal model for Guillain–Barré syndrome (GBS) that would permit the exploration of neurologic disease and the immune responses (to vaccine or infection) driving these outcomes.

Its important to elucidate how cocirculating flaviviruses may interact and affect a Zika vaccines performance. The vaccine performance profile may differ in flavivirus-immune and nonimmune recipients. Mechanisms of immune enhancement and the potential association with severe Zika outcomes are being explored in vitro and in animal models. However, human field studies in a setting with cocirculating flaviviruses would be more informative and definitive.

Demonstrating a vaccines safety and clinical benefit during an epidemic is a key challenge. Before clinical testing, regulators must ensure volunteer safety by assessing the manufacturing process and supporting preclinical data and clinical plans. Many vaccine candidates fail to transition from preclinical to clinical testing because of a lack of sound manufacturing capabilities. Accelerating the production process for Zika vaccine requires numerous early development activities to occur in parallel with clinical evaluation.

Early-phase clinical studies primarily assess safety, but Zikas characteristics necessitate expanded early-phase studies, including assessment of age ranges associated with childbearing potential, volunteers with and without previous flavivirus exposure, and measurement of the longevity of the immune response. Adjuvants may be tested in early phases in efforts to optimize immune responses, but will further complicate clinical studies and potentially delay licensure.

Demonstrating efficacy will require careful planning and substantial resources. First, the relevant clinical end point and its measurability in a reasonably powered and sized study will need to be determined. Historical data indicate that clinical disease develops in approximately 20% of people infected with ZIKV, but this disease is overwhelmingly of a mild phenotype. The current epidemic has been associated with ZCS and GBS, which may be easier to detect but probably have low incidences and will require complex evaluations in order to confirm causal relationships. Additional data are needed to clarify the true incidence of these outcomes; current estimates are that the risk of ZCS after infection during the first trimester of pregnancy ranges from 0.88 to 13.2%4 and that there are 0.24 GBS cases per 1000 ZIKV infections.5 Developers must decide whether to design studies to detect mild disease or increase sample sizes to identify rarer outcomes. Mild disease could also be the primary end point, with rare outcomes as secondary end points collected over time. Alternatively, rarer outcomes could be assessed during postlicensure studies.

During clinical trials, adverse events of special interest, such as GBS, have to be prospectively defined, and any cases detected must be monitored. However, rare events will also probably require postlicensure assessment as part of a risk-management plan to address causality. If ZIKV transmission is highly seasonal, herd immunity reduces transmission, or epidemics occur sporadically, the window for conducting a clinical end-point efficacy trial during the current epidemic may be narrow. The field requires more epidemiologic information to properly design advanced clinical trials (see Table 1).

Quickly advancing from small-scale manufacturing to levels required to support advanced trials or deploy vaccine in settings where disease is endemic requires substantial expertise and resources. Technical requirements may make scale-up problematic, delaying advanced development and licensure. Annual production of 100 million doses, a projected requirement, would be challenging even for the manufacturers of most licensed vaccines. Transferring technology from one organization to another, scaling up manufacturing, and securing approvals to use the vaccine could take years.

Regulatory and licensing strategies are guided by the TPP and the data generated to support the vaccines intended use. Regulatory agencies will pay particular attention to preclinical safety and toxicity studies and assessments of unexpected adverse events during clinical trials and after licensure. The case for licensure may be established through traditional clinical efficacy trials, but declining case counts or an urgent need for intervention may necessitate a different pathway. Alternatives include using efficacy data from studies in animals combined with human immunogenicity data or bridging to an as-yet-undefined immune correlate of protection. Human challenge studies have been proposed in order to augment information from efficacy trials, assist in exploring immune correlates of protection, or generate efficacy data if natural transmission substantially declines. In the absence of a clear understanding of the frequency of adverse neurologic outcomes or the persistence of ZIKV in biologic fluids, however, human ZIKV challenge is ethically complex.

Other flavivirus vaccines have been licensed, including those against yellow fever (live attenuated), Japanese encephalitis (inactivated, live chimeric, live attenuated), tickborne encephalitis (inactivated), and dengue (live chimeric). Some have validated surrogates of protection, and all are based on neutralizing antibody. A neutralizing antibody titer of 1 in 10 is the surrogate of protection for the Japanese and tickborne encephalitis vaccines; for yellow fever, the titer is between 1 in 10 and 1 in 50. Preclinical ZIKV studies suggest that a titer of 1 in 10 for mice and approximately 1 in 100 for nonhuman primates protected against ZIKV challenge.1,2 If these figures translate to humans, developing a ZIKV vaccine is very feasible.

The time required to develop a safe, efficacious ZIKV vaccine will be determined by prior experience with the selected technology, the continuation of outbreaks, and the required scale-up of manufacturing. Ultimately, developing, licensing, and deploying a vaccine capable of affecting the current epidemic will require seamless coordination among developers, regulatory agencies, the WHO, and national health authorities, along with a robust monetary commitment from governments and funding agencies.

Disclosure forms provided by the authors are available at NEJM.org.

The views expressed in this article do not necessarily represent those of the U.S. Army or the Department of Defense.

SOURCE INFORMATION

From the Walter Reed Army Institute of Research, Silver Spring, MD (S.J.T.); Sanofi Pasteur, Lyon, France (M.L., N.A.C.J.); and the Sealy Center for Vaccine Development and Department of Pathology, University of Texas Medical Branch, Galveston (A.D.T.B.).

Notes from Discussion with RC external monitors,

W.H.O. Office, Nairobi, 25 May 2016

In Baringo, there were six subcounties, with populations as noted below. Observers were able to visit 5 of the 6 subcounties.

The data from Baringo are broken down by age break. With 95,000 <5s among 276,000 total, the under-fives appear to be well represented.

County	Sub-county	Target pop	Total vaccinated, 9-59 months	Total vaccinated, 5-15 years	Total vaccinated	Admin coverage
Baringo	Baringo Central	41412	14248	28727	42975	104%
	Baringo North	43032	14594	31991	46585	108%
	East Pokot	60826	25630	39742	65372	107%
	Koibatek	50958	16705	32731	49436	97%
	Marigat	35996	13693	24633	38326	106%
	Mogotio	29508	10439	22926	33365	113%
	County Total	261732	95309	180750	276059	105%