Current and New Technologies for Vaccines and Vaccination Decisions

1. Introduction

The purpose of this paper is to review recent advances in vaccinology aimed at the control of infectious diseases in horses. Particular emphasis will be placed on advances in the formulation and the assessment of efficacy of these vaccines as well as their use for maximum effect. Influenza and West Nile Virus (WNV) provide the best examples of how and where recent advances have been made in development and use. Effective commercial equine vaccines available in North America will also be discussed.

2. Recent Advances-Influenza

Over the past 10 - 15 yr, vaccine manufacturers and the veterinary community have become increasingly interested in gaining a better understanding of vaccine efficacy and the strategic use of vaccines in the prevention of infectious diseases in horses. During this time, we have gained a better understanding of the epidemiology and control of infectious respiratory diseases of the horse through the use of detailed epidemiologic investigations, mathematical modelling techniques, and results from a variety of randomized, controlled, field, and challenge trials. These studies have given us a better understanding of the efficacy of current and recently introduced vaccines and how they may be used to the greatest effect. The publication of these studies has stimulated a great deal of interest in the veterinary community. It has encouraged manufacturers to employ both new and improved methods of vaccine production as well as fund the publication of a number of important in-house and independent studies aimed at assessing vaccine efficacy. In a little over 10 years, we have gone from doubting the efficacy of the equine influenza vaccines marketed in North America to knowing that most provide at least some degree of protection. These advances are the result of a significant improvement in vaccines produced by time-honored techniques as well as the introduction of vaccines formulated by new methods. Today, veterinarians and their clients have several options to employ in the control of influenza. These developments have been followed, and to some degree paralleled, by the remarkable response of the vaccine industry to the introduction of WNV to North America in 1999. Since then, three entirely different and efficacious vaccines have been developed for the prevention of this disease in horses. Results related to the use and effectiveness of these vaccines have been published in the peer-reviewed literature, and all three vaccines have been registered for use in the U.S.

Significant in this chain of events was the publication of the results of a large epidemiologic study and randomized-controlled field trial [1,2]; this information described a series of three yearly outbreaks of equine influenza in a large population of Thoroughbred racehorses. Important findings from these investigations were that outbreaks occurred once a year at about the same time, that outbreaks took at least 4 wk to progress through the population of horses with varied degrees of immunity, that outbreaks seemed to start immediately after the first detection of viral shedding by a horse in the population, and that each outbreak had a profound impact on the economics of racing at that facility. In the third year of the study, a randomized, controlled, double-blind study of the efficacy of the most modern vaccine on the market (i.e., holding the greatest share of the vaccine market at the time) showed that vaccination had little or no impact on the occurrence of disease [3]. In addition to these findings, it was recognized that there was no published data showing efficacy of any of the other equine influenza vaccines on the market in North America at that time.

A landmark development occurred in 1999 when Heska Corporation marketed a modified live, temperature-sensitive, intranasal vaccine [4-7]. Before its release, this vaccine had been subjected to series of challenge studies at the University of Kentucky, the University of Wisconsin, and the University of Saskatchewan. In the studies carried out at the University of Saskatchewan, naïve horses were given a single vaccination and then challenged 5 wk, 6 mo, and 12 mo later [4]. Subsequent to these studies, the vaccine has been shown to be cross-protective against a European strain of the virus and to provide protection for naïve horses within 7 days of a single application. These studies provided the first evidence of efficacy of an equine influenza vaccine marketed in North America, and they may have provided some impetus toward the improvement of existing killed vaccines. In 2003, we presented results at the AAEP Convention that showed that all of the killed vaccines marketed in North America were capable of producing at least some degree of protection against experimental challenge [8]. Significantly, a vaccine developed, formulated, and tested for efficacy according to the strict guidelines applied in the European Common Market seemed to be the most efficacious of the killed vaccines available at the time of our investigations.

Increased understanding of the epidemiology of equine influenza and the development of efficacious vaccines was complimented by the development and application of rapid-test technology [9-12]; with this technology, influenza can be diagnosed in affected animals within 30 min of infection. This important development means that, in combination with the periodic updating of virus strains contained in the vaccines, it should be possible to contain if not prevent outbreaks of this disease in large groups of horses through early identification and isolation of affected horses. Additionally, this information may aid in the institution of appropriate biosafety standards and the vaccination of susceptible animals (animals not recently vaccinated with a vaccine of known efficacy) in the face of a threatened outbreak of the disease.

The fact that substantial advances have been made in the control of influenza does not seem to have dulled scientific interest in further advancement. Investigators in the United Kingdom have been developing a better understanding of the epidemiology of the disease through the application of advanced mathematical-modelling procedures; these procedures help to estimate the true effect of all known risk factors, including timing and frequency of vaccination, serum antibody concentrations, and vaccine strain heterology on disease occurrence [13-16]. The efficacy of mucosal and skin delivery of a DNA vaccine has been used in challenge studies [17,18], and Merial [a,b] has developed a canarypox-vectored vaccine that has also been shown to be protective against experimental challenge.

3. West Nile Virus Vaccines

Even more remarkable than the progress that has been made in the control of equine influenza has been the development, testing, and reporting of challenge-trial results of three vaccines against WNV within ~4 yr of the virus' isolation in North America. In an unparalleled accomplishment in equine vaccinology, Davis et al. [19] submitted the first evidence of vaccine efficacy against WNV for publication on November 15, 2000; this was just 14 mo after the initial isolation of the virus from a horse in the state of New York. Using technology that they had already perfected in the development of a vaccine against Japanese encephalitis, Davis et al. [19] produced a DNA vaccine by transforming COS-1 cells with a plasmid that expressed WNV proteins. These cells secreted high levels of WNV pre-membrane and envelope proteins that were subsequently purified and formulated into a vaccine. When administered intramuscularly to horses, this vaccine provided protection against viremia in experimentally infected, naïve horses after a single injection. Four vaccinated and eight non-vaccinated control horses were challenged through exposure to infected mosquitoes 38 days after vaccination. Viremia was detected in seven of eight non-vaccinates compared with zero of four vaccinates (p = 0.01). One control horse developed clinical signs consistent with WNV encephalitis and was euthanized. All vaccinated horses were euthanized on post-challenge day 14. Gross and histopathologic examination revealed no evidence of WNV infection in these animals. These results were of great importance and provided the first published data demonstrating efficacy of vaccination against WNV in horses. In August 2005, Ft. Dodge Laboratories announced the licensure of a WNV vaccine based on this technology, the first DNA vaccine to be approved for commercial use in animals or humans.

The first commercial equine WNV vaccine was produced by Ft. Dodge Animal Health, Overland Park, Kansas, and provisionally licensed by the United States Department of Agriculture (USDA) in August 2001. During the licensing process, the manufacturer conducted challenge trials aimed at showing the vaccine's protection against viremia after IM injection of the live virus 6 and 12 mo after administration of two doses of vaccine. The results of the 6-mo study showed protection against WNV, but they were not published. The results of the 12-mo study have, however, been published, and they showed a significant reduction in the occurrence of viremia in challenged horses [20]. Viremia was observed in 9 of 11 control horses compared with 1 of 19 vaccinates (p < 0.0001). The results of these challenge studies were very encouraging and suggested that field use of the vaccine should provide vaccinated horses with a significant degree of protection against the development of clinical disease.

The killed vaccine was widely adopted by veterinarians and horse owners as soon as it was released. Although there are confirmed reports of the disease in fully vaccinated animals, the results of three large-scale epidemiologic studies provide excellent evidence of efficacy in the field. Salazar et al. [21] assessed data from 484 clinical cases of WNV disease in horses from Colorado and Nebraska in 2002. They showed that the odds of survival were 2.1 times greater among horses that had received at least one administration of the vaccine than among horses that had not been vaccinated (95% confidence interval [CI] = 1.0 - 4. 5; p = 0.04). Schuler et al. [22] assessed the outcome of 484 clinical cases of WNV in horses in North Dakota in the same year and also found that vaccination reduced death losses in these animals. In this population, the odds of survival of vaccinated horses were at least 3 times greater than those of non-vaccinates (crude odds ratio [OR] = 3; 95% CI = 1.8 - 5.2; p < 0.0001). Epp [c] in her case-control study conducted in 2003 in Saskatchewan, provides the first field evidence on the ability of the vaccine to protect against the development of clinical signs of disease. Based on a study of 300 horses located on 23 case properties and 23 control properties, the odds of remaining free of clinical signs of disease were 22.5 times greater in vaccinated animals compared with non-vaccinated horses (95% CI = 3.0 - 168; p = 0.002). During the outbreak, the overall prevalence of infection of horses in the province was 55.7% (95% CI = 44.9 - 65.8%). Thirteen percent (95% CI = 8 - 18%) of infected animals developed clinical signs of disease, and the case fatality rate was 43.8% (95% CI = 35.2 - 52.4).

Based on these estimates, it is evident that vaccination has had a marked impact on the incidence of disease and death of horses exposed to WNV. Using all the data collected during the outbreak (876 animals), we calculated a crude (therefore, slightly inflated) estimate of vaccine efficacy as 95% (95% CI = 90 - 97%).

Since the licensure of the killed vaccine by Ft. Dodge Animal Health, Merial has marketed a vaccine that uses a recombinant canarypox virus expressing pre-membrane and envelope proteins of WNV. The efficacy of this vaccine has been assessed in three trials where horses were challenged with infected mosquitoes in the biosecurity level 3 containment building at Colorado State University. In the first trial, horses were vaccinated twice (5 wk between doses) and challenged 2 wk after the second dose. Viremia was not detected in any of 9 vaccinates compared with 8 of 10 controls (p = 0.0007). In a second trial, horses were challenged 12 mo after two vaccinations. Viremia was detected in 1 of 10 vaccinates and 5 of 6 controls (p = 0.008) [23]. In a third study, naïve horses were given a single vaccination and then challenged 26 days later. In this study, 1 of 10 vaccinates and 8 of 10 controls became viremic (p = 0.006) [24].

Statistically, there is no difference in efficacy between the killed and recombinant vaccines when horses were challenged 12 mo after vaccination (Mantel-Haenszel Analysis; p = 0.75). Based on our current data, there is a strong relationship between the WNV challenge trials using viremia as the outcome and protection against the development of clinical disease in the field. Therefore, it is reasonable to expect that the recombinant vaccine will prove to be highly effective in the field. A direct study of the three-vaccination regimen (i.e., two vaccinations with killed or recombinant vaccine or one administration of the recombinant vaccine) will likely require a large field trial involving naïve horses and a strong natural challenge. Given the widespread exposure of horses to both the virus and the vaccines, the circumstances required for such a study are unlikely to present themselves like they did in the years 2002 and 2003.

4. Deciding When to Vaccinate

An important aspect of vaccine use for the control of all infectious diseases of the horse, including influenza and WNV, is the timing of vaccination and the frequency of revaccination. The timing of initial vaccination seems straightforward, although more data on the earliest age at which foals will respond adequately to vaccination and the impact of maternal antibody interference and protection for each of the available vaccines are still needed. The risk period for both influenza and WNV are reasonably well understood in the various regions of the continent. Vaccination of naïve animals should be completed 14 - 21 days before the onset of the risk period. The duration of adequate protection, based on challenge and epidemiologic data, is ~6 mo for the influenza vaccines. Challenge data for the WNV vaccines indicate a duration of protection of at least 12 mo, although current recommendations published on the AAEP website suggest a smaller duration of protection.

Although the above data form the basis for recommending the interval between vaccinations throughout life, there are no published studies that clearly address this question. Little is known about the immunologic response of primed horses to revaccination at various times throughout life or the degree and duration of protection achieved after revaccination. Further challenge and field trials are needed to address these issues. Although there is a strong relationship between antibody response, killed influenza vaccines (as measured by single radial hemolysis [SRH]), and protection against experimental challenge [25-31], the resistance of an individual horse to a viral challenge cannot be accurately predicted based on serologic tests. Data derived from a large challenge study of killed vaccines performed in our laboratory suggest that ~17% of horses with suboptimal serum antibody concentrations before challenge will prove to be protected. It is also known that horses vaccinated with the modified live, cold-adapted, intranasal influenza vaccine do not produce significant amounts of antibody but are protected against challenge [4]. Importantly, serologic studies have proven useful in predicting immune status of horses at the herd level. With further research, it may prove possible to develop reliable predictive models that provide insight into the decisions being made about the vaccination of groups of horses as opposed to individual animals. The situation with respect to WNV vaccination is much less clear. Based on current test technology, the serologic status of horses with respect to WNV has not been shown to predict protection of horses at either the herd or individual level. The difficulty in determining protection of horses against infectious disease is not limited to WNV. With the exception of what is known about influenza, there is no good evidence to show how serologic tests may be used to accurately determine whether horses are protected against any infectious disease other than equine influenza.

Footnotes

1. Merial, Lyon, France.
2. Minke JM. Unpublished data. 2000.
3. Epp T. Unpublished data. 2003.