Vaccines and vaccination, an introduction

Published: 05/01/2008
Last updated: 13/09/2015
Last reviewed: 23/06/2022

The present article was first published in the Journal of Feline Medicine and Surgery (2009) 15, 530-537 by Marian C. Horzinek and Etienne Thiry.


The ABCD has been created with the feline practitioner in mind: to provide him/her with a rational base for vaccine use and with advise for the management of feline infectious diseases.

However, the Board hopes to achieve more than that: to bring feline health issues to the forefront of clinical practice, which must be based on continuously updated research findings and confirmed medical evidence. The veterinary experts’ knowledge – rather than any unverified opinion and subjective experience – must be the trusted source of information for the lay public, a service to the animal-owning citizen, which reaches beyond the physical veterinary activities that feature so prominently in popular TV shows.

The Board’s Mission Statement reads: “The European Advisory Board on Cat Diseases (ABCD) aims to issue guidelines on the prevention and management of feline infectious disease in Europe, for the benefit of the health and welfare of cats. The guidelines are based on current scientific knowledge of the diseases and available vaccines concerned.”

This introduction is intended as a ‘light’ primer to vaccines and vaccination, and is of a very general nature; it is not intended to replace textbooks of immunology, which abound both in the veterinary and medical fields, and whose amount of detail may discourage the casual reader. This chapter is therefore not referenced.

The immune response: vaccination and infection

Vaccination is intended to protect individuals against disease caused by bacteria and viruses, but also to prevent infection and transmission of the agents within a population. Vaccination coverage – when exceeding ~70% of a population, as shown e.g. by antibody prevalence – leads to “herd immunity”, which is a safeguard against epidemics. When a wild-type virus infection occurs as immunity wanes, it may be without clinical signs, but it will still boost immunity. For endemic viruses, this frequently occurs in communities of high population density.

Immunity in a vertebrate is based on the multiplication of specialized cells, predominantly lymphocytes, which either kill virus-infected cells by direct contact or through proteins they secrete. Antibody proteins exist in several classes or isotypes. The IgA immunoglobulin isotype is important to prevent infection of mucosal surfaces, of various epithelia. Thus an orally administered, attenuated vaccine virus will replicate in the intestinal tract, leading to prolonged local synthesis of IgA antibody and providing an ‘antiseptic paint’ even on distant mucosal surfaces. By preventing infection, such a vaccine may allow eradication of the virus from the local population or even globally (as for poliomyelitis in humans).

Immunity to reinfection may be life-long for viruses that reach their target organs via systemic spread, through viraemic spread in the organism. Protection in this case is due to antibodies of the IgG class that neutralize virus to which an animal is re-exposed (e.g. feline panleukopenia virus). The objective of live virus vaccination is to mimic the natural infection, which usually elicits high concentrations (titers) of virus-neutralizing antibodies.

Special difficulties attend vaccination against viruses that cause persistent infections and disappear into latency, such as herpes- and retroviruses; a vaccine must be very effective to prevent latency, and indeed only few are able to achieve this. Attenuated viruses are generally more effective than inactivated viruses in eliciting cell-mediated immunity, which is the most effective arm of the immune system in modulating latent/persistent infections.

This summary is to illustrate one aspect of the natural history of infectious agents: there is no “one-fits-all” approach to protection, and the extended Guidelines – as well as the condensed Fact Sheets – are intended to provide the feline expert with the required detailed knowledge. Knowledge is not stationary, as practitioners experience at any continuing education meeting, and changing insights are difficult to convey to the lay public. It should be obvious: the Guidelines are not set in stone; they will have to be updated on a regular basis.

Duration of immunity

The custom of yearly revaccinations of companion animals – a veterinary specialty – dates back to the times of the first commercial products, when knowledge of immunology was modest. A conservative attitude, financial and safety concerns have prevented the industry from translating the growing vaccinological insight into rules for the practitioner. In ignoring the fundamental difference between a drug and an immunogen, the “more-helps-more” adage prevailed; it is indeed counter-intuitive to expect lifelong protection against deadly panleukopenia from a single injection of a few nanograms of a parvovirus preparation. On the other hand, it is unwarranted to extrapolate from this example to other infections, particularly bacterial, and again detailed knowledge about the agents and their immunogens is required.

Immune lymphocytes die, and antibodies wane – their limited life spans are known to science. However, if tests fail to demonstrate their presence, this does not imply that the organism is unprotected. An ‘anamnestic response’ kicks in when the immune system recognizes an antigen it has seen before – be it upon infection or revaccination – immune cells promptly start to divide, and antibodies are synthesized at high speed. Both ‘booster’ phenomena are based on the recruitment of memory cells, which are maintained, often for life, through the cytokine network in the nooks and crannies of the immune system.

The ‘duration of immunity’ (DOI) issue has dominated the veterinary literature during the last decade, the polemics having been inspired by industry claims aimed at achieving a marketing advantage. However, it is moot to discuss the benefits of 2-yearly vs. 3-yearly vaccination intervals when immunological knowledge, independent experimental data and medical vaccinology have evidenced lifelong protection for some viral antigens. After all, “childhood diseases” have been so named because of their temporal limitation, and adults are not revaccinated e.g. against measles.

There is a mirror picture to the DOI polemics, however, and it is a positive one. In their wake, the distinction was highlighted between viral ‘core’ vaccines and bacterial antigens, between modified live, attenuated and killed, inactivated preparations. The polemics have emphasized the need for detailed, agent- and preparation-specific information, and they were at the root of the ABCD initiative. In our publications, we have exposed the nonsense of yearly ‘core’ revaccination as eloquently as we have campaigned for more-than-yearly immunisations of risk groups using new Leptospira serovars. Also, Chlamydophila and Bordetella products require more frequent boosters for reliable protection.

The reproach that the recent recommendations in immunisation schedules will necessarily lead to reduced income from vaccination is unfounded. Quite on the contrary: with the increased sophistication in vaccine development, monitoring of the international scene and translation into local requirements and into the layman’s vocabulary must become an ongoing process. The vaccination interview, the yearly health check will have to become an accepted and invoiced commodity. Epidemiologic information must be collected, interpreted and disseminated. In anticipation of the expansion of viral, bacterial and protozoan diseases, of vector-borne infections (as a consequence of global warming), of zoonotic threats (e.g. SARS, avian influenza), links need to be established and entertained with the medical scene. This is in line with the “One Medicine – One Health” philosophy presently occupying center stage in the biomedical setting (

Vaccine safety, efficacy and tests predicting protection

Of all veterinary activities, vaccination probably stands alone in that its result is not systematically evaluated – certainly not by the vaccinating practitioner. The “shoot-and-trust” principle reigns, and trust is based on the claims of producers and on the registration files. After all, there is a licensing system in place that is operated by national and European agencies for veterinary medicinal products, which serves to ascertain vaccine efficiency, purity and safety. It must be said: mainly the latter. Whereas protection needs to be proven in lengthy, costly, at times scientifically doubtful and ethically controversial vaccination/challenge experiments, undesirable effects are more easily and often more quickly detected.

This is true for medicine and veterinary medicine alike. The so-called Phase I studies are intended to assess just this: the absence of side effects. Or rather: a tolerable degree, because there is always a compromise to be reached. In the case of e.g. live vaccines, the balance is between sufficient virus replication to produce enough antigenic mass, and restricted replication to avoid clinical signs. With killed preparations, the adjuvant may be the culprit, whose task it is to trigger the innate immune system by inducing a local inflammation. Again, this is usually contained, but it may get out of control, causing a transient cytokine rush with fever, nausea etc. Or even worse: like any chronic inflammation, it can develop into a tumor, as the injection-site fibrosarcomas in cats have shown. These are long-term side effects – the pharmaceutical industry has neither the patience nor the funds to test and exclude them; this is now the domain of pharmacovigilance.

So far for the safety concerns, and what about the reason why vaccines are developed in the first place: their efficacy? There is unfortunately no easy way to demonstrate protection of a person or an animal. In medicine, the evidence is mainly statistical, population-based, retrospective. In veterinary medicine, the evidence can be obtained experimentally, by exposing an immunized host to the virulent agent. Such vaccination/challenge expectations have governed the DOI discussion, with the implicit understanding that they will never be fulfilled by industry for longer than the yearly intervals, and so there would be no need to change a cherished habit. However, this pessimism proved to be unwarranted, as recent developments have shown. After a memorable meeting in Prague (2005), at which opinion leaders in European companion animal medicine had been exposed to state-of-the-art vaccinology, acceptance of longer revaccination intervals has grown. Protection has indeed been proven, as indicated above, in lengthy and costly vaccination/challenge experiments – for a three-year interval for a few viruses of the “core” category.

Population (“herd”) immunity

 Vaccination has advantages not only for the vaccinated hosts but also for the unvaccinated part of the population. The latter is indirectly protected because the opportunities for transmission of the virus are reduced (“herd immunity”, a term also applied to humans, cats etc.).

The threshold is defined by the condition Ro=1. If Ro>1 each primary case will, on average, produce more than one secondary case and the infection will spread through the host population, leading to an epidemic. Conversely, if Ro <1 each primary case will, on average, produce less than one secondary case and, although some secondary cases may occur, the infection will die out. An epidemic is possible only if Ro ≥1.

The value of Ro for a given virus in a given host population is determined by e.g. the transmissibility of the virus, the period over which an infected host is infectious, and the population density. The details of these relationships depend on the mode of transmission (e.g., direct contact, from fomites). Equations have been developed to take account of the impact of host life expectancy, duration of protection due to maternal antibody, and the duration of protection afforded by a vaccine.

While these parameters have not been applied to epizootiological predictions in companion animals with their different breed susceptibilities, lifestyles, population densities, contact frequencies etc., they may eventually. After all, protection of the individual is the objective both in human and companion animal vaccination, and population immunity aspects therefore apply to both.

Types of vaccine

Vaccination is the most effective way of preventing viral diseases. While attenuated and inactivated virus vaccines are still the “work horses” of veterinary practice, third generation products are now complementing and, in some cases, replacing them.

There are two major strategies for vaccine design, one employing avirulent live virus, and the other inactivated virions. Modified live vaccine virus replicates in the recipient, and in so doing the antigenic mass presented to the host’s immune system is amplified. The vaccine virus thus mimics an infection, with little or no clinical signs. This means that the host’s immune response is more similar to that taking place after a natural attack by virulent virus than is the case with inactivated or subunit vaccines. With “killed” preparations, the chemical or physical treatment used to inactivate the infectivity may damage or modify the immunogenicity, which usually results in an immune response of shorter duration, a narrower antigenic spectrum, and weaker cell-mediated and mucosal immune responses. However, inactivated vaccines are safe and widely used.

Modified live virus (MLV) vaccines

When proven safe, attenuated viruses have been the best of all vaccines. MLV products are injected subcutaneously or intramuscularly, but some are delivered using natural (mucosal) routes.

Most MLV vaccines have been derived empirically by serial virus passage in cultured cells of heterologous host origin. Adaptation is fortuitously accompanied by loss of virulence for the natural host. This is initially assayed in a laboratory animal, often the mouse, before being confirmed by clinical trials in the target species. The vaccine virus must not be so attenuated that it fails to replicate satisfactorily in vivo, and it is sometimes necessary to compromise, by using a strain that may occasionally induce mild clinical signs. Despite the exceptional success of empirically derived MLV vaccines, research is currently aimed at replacing this “genetic roulette” with rationally designed, engineered vaccines. In these, the mutations associated with attenuation are known and stable.

Recombinant DNA techniques also allow foreign genes to be introduced into the genome of a vector virus. Cells in which the vector replicates will express the foreign protein (containing the relevant antigenic determinants) and the animal will mount both humoral and cell-mediated immune responses to it. Thus vaccinia virus-vectored rabies vaccines have protected red foxes when incorporated into baits that had been laid out in their biotopes. Fowlpox virus is a logical choice as a vector for avian vaccines, and surprisingly is also a functional vector in mammals. Even though this virus, and the closely related canarypox virus, does not replicate to produce infectious particles in mammals, the inserted genes are expressed and induce a vigorous immune response. Also adeno-, herpes- and parvoviruses have been explored as vectors and may have advantages in terms of long-term antigen expression.

Non-replicating (“killed”, inactivated) Vaccines

“Killed” vaccines are usually made by selectively destroying the infectivity of a virulent virus while maintaining the immunogenicity of its proteins. The most commonly used inactivating agents are formaldehyde, β-propiolactone and ethylenimine. These vaccines are safe, but a large antigenic mass is required to elicit an antibody response commensurate with that attainable by the tiny dose of an attenuated virus. For the manufacturer, this is a costly procedure. The primary vaccination course usually consists of two or three injections, and “booster” doses are required at intervals to maintain immunity.

Using recombinant-DNA technology, large amounts of protein can be produced, readily purified and formulated into subunit vaccines. Once the protein conferring protection has been identified, its gene may be cloned into a plasmid and expressed in any of several cell systems. Useful eukaryotic expression systems include yeast, insect and various mammalian cells, but even the procaryote E.coli has been used to express an immunogenic viral surface protein (gp70 of FeLV, in its unglycosylated p40 form). In general, mammalian cells are used since they possess the machinery for correct post-translational processing, including glycosylation and secretion of viral proteins.

Common to all non-replicating vaccine preparations is their immunologic pathway involving the MHC type II molecule. If injected alone, the proteins lack the intracellular signalling required for a broad response, and support must be provided by adjuvants. These potentiate the immune response, both humoral and cellular, so that less antigen and/or fewer doses will suffice. Adjuvants differ in their chemistry and mode of action, which includes 1) the prolongation of antigen release from the site of injection, 2) macrophage activation leading to secretion of lymphokines and attraction of lymphocytes, and 3) mitogenicity for lymphocytes. Aluminium salts and mineral oils are widely used in animal vaccines; a promising new adjuvant is muramyl dipeptide, which can be coupled to synthetic antigens or incorporated into liposomes.

Viral chimeras

The principle of chimeric viruses was well established even before the advent of recombinant DNA technology. Chimeric influenza viruses had been produced by reassortment (segment swapping – the influenza genome comes in 8 segments) by co-cultivation of an existing vaccine strain with the new isolate. Reassortant viruses with the desirable growth properties of the vaccine strain and the immunogenic properties of the isolate were selected, cloned and used as vaccine. In view of the emerging role of carnivores in influenza virus epidemiology, we may see more of this strategy in the future.

“DNA vaccines”

Instead of an immunogenic protein, isolated viral DNA can be used as a vaccine. Recombinant plasmids containing the relevant genes are injected into a vertebrate, which results in the transfection of cells and protein expression. An immune response ensues that simulates the response elicited by the viral infection. DNA vaccines have shown protection, but only in some virus/host systems.

Advantages of DNA vaccines include their achievable purity, their physiochemical stability, simplicity and low cost of production, distribution and delivery. Multiple genes can be engineered into a single plasmid, and antigens will be expressed in their native form, with MHC type I processing and presentation to the immune system. Repeated injections can be given without interference from the immune system, and both cellular and humoral responses are elicited. One interesting aspect of DNA immunisation is that it may induce immunity even in the presence of maternal antibodies.

Kitten vaccination

Kittens from regularly vaccinated queens or from those having experienced field infections are protected by maternally acquired antibodies during the first few weeks of life. In general, this passive immunity will have waned by 8 to 12 weeks of age to a level that allows active immunisation. Pups with little antibody may be vulnerable (and capable of responding to vaccination) at an earlier age, while others may possess them at such high titres that they are incapable of responding until about 16 weeks of age. No single primary vaccination policy will therefore cover all possible situations.

In view of these facts, the first vaccination must be followed by a second immunisation 3 to 4 weeks later, and by a third vaccination between 14 to 16 weeks of age. Some vaccine data sheets recommend an initial course of only two injections, but this may be insufficient, depending upon the local vaccination habits and the degree of pathogen circulation, as discussed above.

In immunological terms, the repeated vaccinations of kittens in their first year of life do not constitute boosters. They are rather attempts to induce a primary immune response by injecting the attenuated virus into an animal in a time window, where it is devoid of neutralising antibody. Only then, an attenuated virus can multiply and be processed by antigen presenting cells, eventually stimulating the expansion of antigen-specific T- and B-lymphocytes. In the case of inactivated vaccine products, maternally derived antibody may interfere with successful immunisation by binding to and ‘masking’ the relevant antigens.

The Vaccination Guidelines Group (VGG), an expert committee of the World Small Animal Veterinary Association, has used the term “core vaccine” to identify the components that all cats should receive, regardless of the circumstances. Core vaccines should protect animals from severe, life-threatening diseases. Core vaccines for cats are those that protect against feline parvovirus (FPV), calicivirus (FCV) and herpesvirus (FHV) infections. Their global distribution is not part of the definition – as the rabies example shows: there are countries, where vaccination is neither mandatory nor sensible, because they have officially been labelled as rabies-free. However, in areas of the world where rabies is endemic, vaccination should be considered core, even when there are no legal requirements.

The VGG has defined non-core vaccines as components for animals whose geographical location, local environment or lifestyle places them at risk of contracting the specified infections. The VGG has also classified some vaccines as not recommended (where there is insufficient scientific evidence to justify their use) and has disregarded minority products, which are not universally available.

Passive immunisation

Short-term, instantaneous protection can be achieved by the subcutaneous administration of antibody, in the form of immune serum or immunoglobulin. Homologous immunoglobulin is preferred, because foreign protein may provoke a hypersensitivity response. Pooled immunoglobulin from healthy donors contains sufficient antibody against the common viruses that cause systemic disease in the respective species. Serum from animals that have recovered from infection or have been hyperimmunized by repeated vaccinations possess higher titers and are to be preferred, if available. Passive immunisation is not used as a routine procedure.

A more common practice is to vaccinate the queen before breeding. This provides the offspring with passive (maternal) immunity via antibodies present in colostrum mainly, and in milk to a lesser extent. This is particularly important for severe diseases occurring during the first few weeks of kitten life, when active immunisation cannot be accomplished. However, the better a queen is immunized, the longer she will give antibody to her kittens, and the later these can be successfully immunized.

Vaccination policy in populations

An optimal vaccination program depends on both the characteristics of the vaccine and the epidemiology of the agent. Vaccine characteristics include the proportion of protected vaccinates, the duration of protection and the coverage achieved in a population by the program (herd immunity). The most important parameter is the reproduction figure Ro (as explained above). Rabies in foxes has been used as a guide to the design of vaccination programs and the general usefulness of epidemiological modelling.

Mathematical theory can integrate this information to address questions like

whether it is at all possible to eliminate an infection

what percentage of a host population must be vaccinated to achieve this

at what age an individual should first be vaccinated

at what intervals should individuals be revaccinated (if at all required).

A microorganism can be eliminated from a population when the herd immunity level reaches or exceeds 67%, which is attained nowhere in cats. Persistently infected, shedding animals, ‘spill-over’ of the agent from other host species, its long-time presence in the environment due to an extreme resistance to inactivation all enter the equation. For the cat, this means that elimination of most feline infectious conditions is not possible. Containment and control, on the other hand can be achieved for most of them. – The term ‘eradication’ should not be used in this context, since it refers to the global elimination of a disease and its agent, as has been achieved for variola major, smallpox in man; similar objectives pertain to poliomyelitis and measles.

Vaccination in crowded cat communities

A distinction must be made between, shelters, multiple-cat households and boarding catteries. The latter are facilities accepting fully vaccinated animals for short periods of time, e.g. when owners are on vacation. Entry into any such facility requires that the individual cat be fully vaccinated with core products; non-core vaccines against respiratory infections are appropriate. The same goes for multiple-cat households, where the owners must be educated in this sense.

A shelter is a holding facility for animals usually awaiting adoption, rescue, or reclaim by owners. A source population of unknown origin, an unknown vaccination history, a high population turnover, and a high risk of infection characterize these facilities. Sanctuaries that possess a stable population, facilities that admit hundreds of animals per day, rescue and foster homes that care for many cats or litters at any given time are all “crowded cat communities”. Just as the approach to vaccination varies with individual pets, there is no universal, standard vaccination strategy for animals under such conditions.

While infectious agents cannot be eliminated from shelters, the spread of infections can be minimized and the health of yet uninfected individuals maintained. Modified live virus vaccine preparations are preferable. If unambiguous documentation of vaccination is provided for an animal at the time of admission to a shelter, there is no reason to revaccinate.

Adverse reactions to vaccines

As discussed above, immunity to viruses of the core category has been agreed to last for about three years (being the longest time span tested), but e.g. bacterial components in a combined vaccine may not provide such long protection. In companion animal practice the cost of vaccination is comparatively small – and if revaccination does no harm, it may be justified to make it part of a routine health check. In many countries the annual revaccination has become a cornerstone of health care programs, and a source of regular income; however, the time is ripe to reconsider.

The “no-harm” way of thinking was disturbed in the mid-1990ies by reports of subcutaneous fibrosarcomas in cats (often in the shoulder region), corresponding to actual or suspected vaccine injection sites. After more focused observation, also other injectables have been incriminated in the condition, and evidence is now pointing towards repeated local irritations with chronic inflammation that may be at the root of the problem. The suggestive term “vaccination-site fibrosarcoma” has been used a little too indiscriminately and has biased the veterinary and lay communities alike.

Many reported side effects fulfill the criteria of selective observations, and although some have achieved prominence in the scientific literature – like the fibrosarcoma – they are comparatively rare. However, also some infections have become rare, and it must be anticipated that vaccination will be discontinued when the disease to protect against is no longer around. The scenario is similar to that during the final phase of smallpox vaccination, when the sporadic side effects (less than 1 in a million vaccinated) still exceeded the number of natural disease cases.

Factors affecting vaccine efficacy and safety

Vaccines are produced under Good Manufacturing Practice (GMP) guidelines. As a minimum requirement, the licensing authorities insist on safety tests for residual infectious virus in inactivated preparations. Other safety problems are unique to attenuated virus vaccines – an important consideration being that attenuation is host-dependent; it is only certified for the species in which the safety tests have been performed. Off-label use of MLV vaccines e.g. in exotic carnivores has incidentally lead to dramatic losses.

Also attenuated viruses may produce clinical signs – in effect, a mild disease. Further attenuation by additional cell culture passages has led to lower titers in the vaccine, with a corresponding loss of immunogenicity. Any MLV vaccine therefore represents a compromise between virulence and attainable antigenic mass. – Although vaccine strains may revert to virulence during replication in the recipient or in contact animals, this has never been observed for feline products.

Since vaccine viruses are grown in animal cells, there is a risk that a product is contaminated with an adventitious virus from that animal. The risk from contaminating viruses is greatest with MLV vaccines, but it has so far not been reported for feline products.

Vaccines containing live viruses are not recommended for use in pregnant animals, since they may cause abortions or malformations, like e.g. feline panleukopenia virus products. Adverse effects are usually seen upon primary immunisation of pregnant animals. Breeders that wish to boost immunity in pregnant queens, to provide high maternal antibody levels, should be advised to have them vaccinated before mating.

Reasons for poor vaccine performance

The most common reason for vaccination failure is the neutralisation/masking of vaccinal antigen by colostral and/or milk antibodies acquired from the queen. When the last vaccine dose is given at ≥12 weeks of age, however, these should have decreased to a low level, and active immunisation will succeed in most kittens.

Other reasons for vaccination failure may be extrinsic (concomitant immunosuppressive retrovirus infections) or intrinsic (the animal is a poor responder, its immune system fails to recognize the vaccinal antigens). If this is the case after repeated revaccinations, the cat should be considered a non-responder. Because immunological non-responsiveness is genetically controlled in other species, certain breeds (e.g. Siamese) have been suspected to be poor-responders.

Poor immunogenicity of a vaccine preparation may be due to many factors between manufacture and administration. This has been more often suspected than proven. However, the virus strain, its passage history, production errors, flawed manufacture of a particular batch etc. have all been recognized as causes of vaccine failure. More commonly, post-manufacture causes such as incorrect storage, transportation (interrupted cold chain) and handling (disinfectant use) in the veterinary practice may result in inactivation of a modified live virus product. Inactivated vaccines are less vulnerable.

This might also interest you