Feline leukaemia virus and vaccination

The immune response is one of the crucial factors determining the outcome of infection with FeLV, but the mechanisms of naturally acquired immunity to FeLV are incompletely understood. Many investigations of the protective immune response have focused on the importance of virus neutralising (VN) antibodies and anti-feline oncornavirus-associated cell membrane antigen (anti-FOCMA) antibodies which are thought to protect against persistent viraemia (PV) and FeLV-associated neoplasia, respectively.

FeLV vaccines

An understanding of the immune response to FeLV infection helps to resolve some of the controversies surrounding FeLV vaccination. There are currently several commercial FeLV vaccines available, and some are marketed by more than one company (under different labels). The vaccines are all inactivated, but some differ in other aspects. Most are adjuvanted (using different products), there are whole virus vaccines and multiple subunit and recombinant vaccines. Also they incorporate different sources and subgroups of FeLV.

Assessment of vaccine efficacy

An ideal vaccine would provide protection against both persistent and transient viraemia and thus also prevent latent infections and the development of FeLV-related diseases. However, none of the currently available vaccines have been shown to produce sufficient mucosal immunity to routinely prevent transient viraemia following exposure. Table 1 summarises the results of the FeLV major vaccine efficacy studies relating to the currently available products published in the veterinary literature. Table 1 lists both studies evaluating a single product and also comparative studies but, as indicated, where studies included evaluation of experimental vaccines or vaccines no longer commercially available, the results have been omitted.

Table 1.

Results of the studies of commercially available FeLV vaccines

References	Vaccine(s)	Number of cats		Preventative fraction
		Vaccine	Controls
Sebring et al (1991)	Fel-O-Vax	4	4	100%
	Leukocell 2	4		50%
	VacSYN	4		25%
Sebring et al (1991)	Fel-O-Vax a,b	90	58	95.1%
Lafrado (1994)	Leukocell 2	26	26	100%
Haffer et al (1990)	Leukocell 2	25	10	53.3%
Clark et al (1991)	Genetivac	20	20	78.6%
Hines et al (1991)	Fevaxyn b	144	45	90.4%
Legendre et al (1991)	Fel-O-Vax	12	11	100%
	Leukocell 2	12		34.5%
	VacSYN	12		21.4%
Lehmann et al (1991)	Genetivac c	18	12	93.3%
Pedersen & Johnson (1991)	VacSYN a,d	18	12	39.4%
Pollock & Haffer (1991)	Leukocell 2	148	81	74.8%
Pollock & Haffer (1991)	Leukocell 2	14	5	88.1%
Tizzard & Bass (1991)	Leukocell 2 a	10	18	44.6%
York & York (1991)	VacSYN	43	22	92.7%
Pedersen (1993)	Fevaxyn	10	10	100%
Jarrett & Ganiere (1996)	Leucat	12	8	−14.3%
	Leucogen	12		52.4%
	Leukocell 2	12		4.8%
Jarrett & Ganiere (1996)	Leucogen	6	6	80%

^aOther (unlicensed) vaccines included in the study.

^bComposite figure given for the results of several reported trials.

^cFifty percent of vaccinates and controls FIV-infected, but FIV status reported not to affect vaccine efficacy.

^dStudy performed when VacSYN was only licensed (under a different name) in one state of the USA.

An important concept in the evaluation of FeLV vaccine efficacy is that of the ‘preventable fraction’ (PF). This is designed to give a more accurate reflection of vaccine efficacy than simply looking at the proportion of vaccinated cats protected against viraemia, since it takes into account the fact that often considerably less than 100% of control (non- or sham-vaccinated cats) develop PV. The PF is therefore defined as the proportion of cats protected by vaccination in excess of that protected by natural resistance and is calculated as:

$$\text{PF}\left(\%\right)\text{=}\frac{\%\text{Controls with PV}-\text{\%Vaccinates with PV}}{\%\text{ Controls with PV}}\times 100$$

Table 1 lists the calculated PF from the various studies cited. Although assessment of the PF is the recommended way of evaluating vaccine efficacy, it is important that other factors are also considered, in particular the number of cats used for the study, and the number of controls developing PV. For example, in the study reported by Sebring et al (1999) comparing the efficacy of Fel-O-Vax, Leukocell 2 and VacSYN, there were only four cats in each group (vaccinates and controls), thus a 25% difference in the reported PF was accounted for by a single cat developing or resisting PV. Also in the study evaluating Leukocell 2, reported by Lafrado (1994), the PF was calculated as 100%, but this reflected the development of PV in only one of the 26 control cats, and none of the vaccinates. This emphasises the need to examine otherfactors in addition to the PF to fully evaluate FeLV vaccine studies, particularly the number of cats used, and the proportion of controls that develop PV.

From the limited data available (see Table 1), the whole cell vaccines (Fel-O-Vax and Fevaxyn) appear to show most consistent protection against FeLV challenge. However, it can be clearly seen that the reported efficacy for any individual product varies, sometimes greatly, between different studies. Several factors help to explain this, and also make direct comparison between studies very difficult. There are important differences between the studies such as the method of viral challenge, the strain of virus used for the challenge and the age of cats at the time of challenge. Although studies involving ‘natural exposure’ of cats (ie, control and vaccinated cats housed together with persistently viraemic cats) will mimic field exposure to the virus most closely, many studies employ an artificial challenge system. This usually involves the administration of virus via the intraperitoneal or the oro-nasal route, frequently with concurrent immunosuppression provided by administration of corticosteroids. These changes produce a much higher proportion of infected control cats, reducing the overall number of cats needed for efficacy studies. However, they leave open the question as to whether such studies truly reflect the efficacy of a product under natural conditions where the virus is usually spread by prolonged close contact between cats. Furthermore, it is obviously difficult to compare studies using different challenge protocols, and even where the same virus and challenge route were used for different studies there were often other differences such as in the dose of challenge virus, the frequency of virus administration or the protocol for inducing immunosuppression.

Another important difference between the studies is the post-challenge sampling protocol and definition of PV, which may have an important impact on the final results. All this serves to emphasise the difficulties in comparing results from the reported studies, and the importance of trying to standardise an approach to FeLV vaccine efficacy trials. Single comparative trials involving several (or all) available vaccines are preferable for assessment of efficacy, but no one study would be able to address all of the variables that could have an impact on the results. Also, most of the vaccine efficacy studies have been either supported by, or performed by, one of the vaccine manufacturers/distributors. This does not imply criticism of these particular studies, and if adequate details of the study design are provided they can be judged on their own merit, however, fully independent trials where commercial interests are not an issue are clearly preferable. Although independent, natural-challenge studies where several (or all) vaccines are compared simultaneously are likely to provide the most convincing efficacy data, such trials areexpensive to undertake, and this also raises the issue of how such studies can be funded.

Results from published studies clearly suggest that none of the vaccines provide 100% protection against transient viraemia, but again comparison between studies is extremely difficult due, particularly, to the different post-challenge sampling protocols and the definitions of persistent (and therefore transient) viraemia. As transient viraemia has been detected in a proportion of vaccinated cats, it is not surprising that latent infections have also been identified in a variable proportion of vaccinated cats in some studies.

FeLV subgroups and vaccination

While obviously not deleterious, there is no evidence that inclusion of subgroups B and C in a vaccine is of any benefit. There is no evidence of natural horizontal spread of FeLV-B or C between cats and even if it were to occur, as these virus subgroups are defective, it would require the concomitant transmission of FeLV-A to establish infection. Protection (primarily through the induction of VN antibodies) against FeLV-A should therefore be all that is required of a vaccine. Furthermore, the heterogeneity of FeLV-B and C isolates means VN antibodies induced by one isolate would not necessarily provide cross-protection against another. The sufficiency of including just FeLV-A derived antigens in a vaccine has also been confirmed in two different studies, where cats immunised with Leucogen or a prototype vaccine containing just FeLV-A were shown to be protected against challenge with a mixture of both FeLV-A and B viruses.

Although FeLV-A strains are monotypic with cross-reacting VN antibodies, other differences between isolates confer altered infectivity and pathogenicity and this is one reason why full evaluation of vaccine efficacy requires challenge exposure to different FeLV isolates.

VN antibodies and vaccination

Although CMI and induction of antibodies to other viral proteins may play a secondary role in the protection of cats, it is the induction of VN antibodies to prevent viraemia that is considered of prime importance to vaccination. However, examining VN antibody responses post-vaccination cannot assess vaccine efficacy. It is clear from several studies that while vaccination may confer solid protection to an individual, this is not necessarily reflected in high VN antibody titres. In many cases, vaccination appears to ‘prime’ the cat, and appreciable VN antibody titres may not be observed until after subsequent challenge with FeLV.

A practical consequence of the importance of the VN antibody response is that it is generally considered to be possible to booster vaccinate a cat using a different vaccine than that used for primary vaccinations.

The importance of FOCMA in FeLV vaccines

Of the vaccines available, only some claim to incorporate FOCMA, although it has been suggested that due to the way in which the vaccines are produced, all except Leucogen may in fact contain FOCMA. However, there is no evidence that the inclusion of FOCMA in the vaccines is of any benefit to the cat. If a vaccine protects against infection with FeLV, it will also protect the cat from the development of FeLV-related disease, and there is no evidence that the inclusion of FOCMA has any role in protecting cats against infection.