《学术英语（医学）》拓展阅读资料：a new TB vaccine fact or fiction

Comparative Immunology, Microbiology and Infectious Diseases 36(2013)287-294 Contents lists available at SciVerse Science Direct Comparative Immunology, Microbiology and infectious diseases ELSEVIER journalhomepagewww.elsevier.com/locate/cimid Review a new tb vaccine: Fact or fiction? Paul d. van helden*. Eileen g. hoal ivision of Molecular Biology and Human bosch University, P O Box 19063, Tygerberg 7505, South Africa ARTICLE INFO A B STRACT Vaccination has been spectacularly successful in eradicating or controlling some infectious diseases, and is particularly attractive as an approach to tackling other infectious diseases. Although vaccination against tuberculosis has been done for nearly 100 years, it is clearly not that successful since the disease still occurs at epidemic levels in animals and humans in many areas. New approaches to vaccination against TB in humans and animals are cur- rently in the pipeline, but none show either complete protection or sterilization. However there is evidence to suggest that vaccination may deliver some positive outcomes. Not only should we be investigating new vaccines, but also how vaccines and candidates are used and delivered. There are many reasons to think that this task will not be simple, or perhaps no possible in some cases We present different aspects of the development of vaccines against TB, outline some complications and suggest some new ways to consider this problem. O2012 Elsevier Ltd. All rights reserved. 1. Background/introduction 2. Prior or acquired immunity? 3. Animal models 4. Choice of vaccine candidates 6. Where should we target our intervention step? 7. Vaccine administration and delivery 8. Clinical trials, evaluation and end-points 8999 9. will vaccination make a significant impact? 10. Goal of vaccination ……291 11. Will a new vaccine merely shift the population structure of mycobacteria? 12. Future prospects Acknowledgments References 292 1. Background/ introduction generally not an option for animals, with the exception of mycobacteriosis, or tuberculosis(TB)is still common the occasional animal in captivity. For TB. as with other today in both humans and animals. At present, we can treat diseases, prevention is better than cure, and thus attempts the disease in humans with antibiotic treatment this is have been made to produce a vaccine against tb for over 100 years. This paper aims to briefly review the field in terms of human and animal reaction to vaccination against r.TeL:+270219389401;fax:+270219389863 TB and highlights some of the difficulties and progress in the field front matter o 2012 Elsevier Ltd. All rights reserved. 0.1016/ timid201207003

Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 Contents lists available at SciVerse ScienceDirect Comparative Immunology, Microbiology and Infectious Diseases jou rn al h om epage: www.elsevier.com/locate/cimid Review A new TB vaccine: Fact or fiction? Paul D. van Helden∗, Eileen G. Hoal DST/NRF Centre of Excellence for Biomedical Tuberculosis Research, MRC Centre for Molecular and Cellular Biology, Division of Molecular Biology and Human Genetics, Faculty of Medicine and Health Sciences, Stellenbosch University, P.O. Box 19063, Tygerberg 7505, South Africa a r t i c l e i n f o Keywords: Tuberculosis Humans Animals Vaccination Prevention Susceptibility a b s t r a c t Vaccination has been spectacularly successful in eradicating or controlling some infectious diseases, and is particularly attractive as an approach to tackling other infectious diseases. Although vaccination against tuberculosis has been done for nearly 100 years, it is clearly not that successful since the disease still occurs at epidemic levels in animals and humans in many areas. New approaches to vaccination against TB in humans and animals are cur￾rently in the pipeline, but none show either complete protection or sterilization. However, there is evidence to suggest that vaccination may deliver some positive outcomes. Not only should we be investigating new vaccines, but also how vaccines and candidates are used and delivered. There are many reasons to think that this task will not be simple, or perhaps not possible in some cases. We present different aspects ofthe development of vaccines against TB, outline some complications and suggest some new ways to consider this problem. © 2012 Elsevier Ltd. All rights reserved. Contents 1. Background/introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 2. Prior or acquired immunity? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 3. Animal models . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 288 4. Choice of vaccine candidates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 5. Possible complications? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 6. Where should we target our intervention step? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289 7. Vaccine administration and delivery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 8. Clinical trials, evaluation and end-points . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 9. Will vaccination make a significant impact? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290 10. Goal of vaccination . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 11. Will a new vaccine merely shift the population structure of mycobacteria? . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 12. Future prospects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 Acknowledgments . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292 1. Background/introduction Mycobacteriosis, or tuberculosis (TB) is still common today in both humans and animals. At present, we can treat the disease in humans with antibiotic treatment. This is ∗ Corresponding author. Tel.: +27 021 9389401; fax: +27 021 9389863. E-mail address: pvh@sun.ac.za (P.D. van Helden). generally not an option for animals, with the exception of the occasional animal in captivity. For TB, as with other diseases, prevention is better than cure, and thus attempts have been made to produce a vaccine against TB for over 100 years. This paper aims to briefly review the field in terms of human and animal reaction to vaccination against TB and highlights some of the difficulties and progress in the field. 0147-9571/$ – see front matter © 2012 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.cimid.2012.07.003

P.D. van Helden, E.G. Hoal/ Comparative ology, Microbiology and Infectious Diseases 36 (2013)287-294 It is often said that Mycobacterium bovis BCG, the current we also know that many species appear to be quite suscep- vaccine against tuberculosis, is arguably the most fre tible since TB is commonly reported in them, e.g. badgers, quently given human vaccine in the world today. It has been possums, cattle, buffalo, lions, lechwe, deer[13-16].On in use since the 1920s, despite ongoing argument about the other hand tb (or mycobacterial disease) has never ol its efficacy and complications in its usage [1, 2. Owing to rarely been reported in many other species, such as dogs this, some low incidence countries such as the USA, do not or horses. In some susceptible species, vaccination stud routinely vaccinate with BCG 3 One of the factors in this ies have been done, with some studies reporting various controversy is that there are now many different deriva levels of protection, using BCG or new candidate vaccines tives of BCG [ 4, which may not be comparable in their [17-26 However, it is not clear whether some animal effect, making any analysis of outcome extremely complex, strains or species or a proportion of any given species is and inter-trial or meta-analyses impossible owing to lack resistant to TB, or whether a proportion can acquire immu of power [1, 2,5]. Furthermore, BCG can be administered in nity under the appropriate conditions. different ways and at different ages 3, complicating an comparative analysis. Existing meta-analysis of BCG vacci- 3. Animal models nation suggests efficacy ranging from negative to positive effects, or protection preferentially against TB meningitis Various options exist, but the mouse model is still the and military TB, although it has been concluded that overall most frequently used Mice are relatively cheap and can be he risk of TB in humans are reduced by 50%[1, 2). Despite housed reasonably easily under containment for pathogen problems of analysis it cannot be claimed that BCG vaccina- challenge. Many well-characterized strains of mice are tion against TB is highly effective It is probably safe to say available, and there is a variety of reagents and kits avail- that BCG performs best at limiting disseminated disease able for mouse research. Few if any specific reagents or and mortality in children 1, 6]. It cannot reliably prevent lines exist for other animal models. However, there are infection or protect against pulmonary tB disease in adults some major differences between mice, other ani Thus, BCG as a vaccine is not sterilizing and there is a clear humans one notable difference is that mice do not form reason for development of new vaccine candidates (like he classic granuloma seen in most other vertebrates with wise, BCG is not a sterilizing vaccine in animals). Any nev TB. For this and many other reasons, this model may there candidate will have to be demonstrably better than the cur- fore have major deficiencies. Noting this, some researchers rent live BCG, in terms of efficacy and safety, the latter have invested in non-human primates as a model sys being a problem with BCG use, particularly in HIV posi- tem [27-29, but this is an expensive and time consuming route, albeit perhaps self-evidently better than others. Ide does not stop infection, latent TB or reactivation, or guar ally, potential vaccines for a particular species should be antee sterilizing immunity and we may perhaps find better tested in either the same animal, or at least a closely solutions using a post-exposure or therapeutic vaccine[8]. related species such as bovids for cattle or buffalo vaccines Perhaps the most difficult problem with TB or my Each existing model has its own problems and erial immunity is that the organism hides very successfully cies, the details of which are beyond the scope of this inside the macrophage or granuloma. For this reason, it paper to discuss in detail. However, a common limita- Is comm hought that T-cell immunity is critical and tion is that many lab-based models are not exposed to hat B-cells are of lesser importance 9-12]. Developing a the environmental stimuli experienced by free-living crea- vaccine to deliver immunity based on this premise is of tures, which include organisms such as other bacteria, fungi course complex, unlike the well-known successful humoral and parasites, which alone may render their hosts defi- response vaccines against viral diseases. cient as models. In addition. most animal models are inbred strains. If we postulate that genetics is a major determinant 2. Prior or acquired immunity? of resistance or susceptibility [30-33]. this alone makes such models tricky to understand since the strain may Since the overwhelming majority of infected humans be inherently resistant or susceptible. Even if not inbred never develop symptomatic (TB)disease, one may con- certain animal species may be innately resistant, where clude that they have inherent or innate immunity which vaccination is unnecessary, or innately susceptible, being an cope with theinfection, rendering them resistant. How- perhaps intractable to vaccination. This variation may of ever, it is also clear that at least in some communities, BCG course also be found within an out-bred population. due to genetic variation. Recognizing the limitations in labo- sonably conclude that vaccination can confer some degree ratory based animal models and the importance of dealing of acquired immunity. One may thus postulate that in the with research close to the real problem, a number of groups absence of immunosuppression, most humans are inhe have investigated larger animals such as deer or cattle for ently resistant via innate immunity: some may become trial vaccination. These animals are arguably a good choice resistant with the correct stimulus, but that some ma ince they are potentially susceptible, are at risk for disease, not be able to acquire immunity to TB In the absence of are economically important and impact on ecosystems. a sound body of research and data concerning diffe eren In addition, they present with large genetic and environ species of animals, we cannot make statements with the mental heterogeneity, which better reflects what we may same level of confidence for these species. However, w expect in trials under real life conditions. know that different mouse strains show varying levels of For this reason, apart from laboratory-based animal resistance or susceptibility to mycobacterial infection and models, a number of attempts to vaccinate free-living

288 P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 Itis often said that Mycobacterium bovis BCG,the current vaccine against tuberculosis, is arguably the most fre￾quently givenhumanvaccine inthe worldtoday.Ithas been in use since the 1920s, despite ongoing argument about its efficacy and complications in its usage [1,2]. Owing to this, some low incidence countries such as the USA, do not routinely vaccinate with BCG [3]. One of the factors in this controversy is that there are now many different deriva￾tives of BCG [4], which may not be comparable in their effect, making any analysis of outcome extremely complex, and inter-trial or meta-analyses impossible owing to lack of power [1,2,5]. Furthermore, BCG can be administered in different ways and at different ages [3], complicating any comparative analysis. Existing meta-analysis of BCG vacci￾nation suggests efficacy ranging from negative to positive effects, or protection preferentially against TB meningitis and military TB, although it has been concluded that overall the risk of TB in humans are reduced by 50% [1,2]. Despite problems of analysis it cannot be claimed that BCG vaccina￾tion against TB is highly effective. It is probably safe to say that BCG performs best at limiting disseminated disease and mortality in children [1,6]. It cannot reliably prevent infection or protect against pulmonary TB disease in adults. Thus, BCG as a vaccine is not sterilizing and there is a clear reason for development of new vaccine candidates (like￾wise, BCG is not a sterilizing vaccine in animals). Any new candidate will have to be demonstrably better than the cur￾rent live BCG, in terms of efficacy and safety, the latter being a problem with BCG use, particularly in HIV posi￾tive neonates [7]. BCG is also a pre-exposure vaccine which does not stop infection, latent TB or reactivation, or guar￾antee sterilizing immunity and we may perhaps find better solutions using a post-exposure or therapeutic vaccine [8]. Perhaps the most difficult problem with TB or mycobac￾terial immunity is thatthe organismhides very successfully inside the macrophage or granuloma. For this reason, it is commonly thought that T-cell immunity is critical and that B-cells are of lesser importance [9–12]. Developing a vaccine to deliver immunity based on this premise is of course complex, unlike the well-knownsuccessfulhumoral response vaccines against viral diseases. 2. Prior or acquired immunity? Since the overwhelming majority of infected humans never develop symptomatic (TB) disease, one may con￾clude that they have inherent or innate immunity which cancope withthe infection, rendering themresistant.How￾ever, it is also clear that at least in some communities, BCG vaccination provides some protection, thus one may rea￾sonably conclude that vaccination can confer some degree of acquired immunity. One may thus postulate that in the absence of immunosuppression, most humans are inher￾ently resistant via innate immunity; some may become resistant with the correct stimulus, but that some may not be able to acquire immunity to TB. In the absence of a sound body of research and data concerning different species of animals, we cannot make statements with the same level of confidence for these species. However, we know that different mouse strains show varying levels of resistance or susceptibility to mycobacterial infection and we also know that many species appear to be quite suscep￾tible since TB is commonly reported in them, e.g. badgers, possums, cattle, buffalo, lions, lechwe, deer [13–16]. On the other hand, TB (or mycobacterial disease) has never or rarely been reported in many other species, such as dogs or horses. In some susceptible species, vaccination stud￾ies have been done, with some studies reporting various levels of protection, using BCG or new candidate vaccines [17–26]. However, it is not clear whether some animal strains or species or a proportion of any given species is resistant to TB, or whether a proportion can acquire immu￾nity under the appropriate conditions. 3. Animal models Various options exist, but the mouse model is still the most frequently used. Mice are relatively cheap and can be housed reasonably easily under containment for pathogen challenge. Many well-characterized strains of mice are available, and there is a variety of reagents and kits avail￾able for mouse research. Few if any specific reagents or lines exist for other animal models. However, there are some major differences between mice, other animals and humans. One notable difference is that mice do not form the classic granuloma seen in most other vertebrates with TB. For this and many other reasons, this model may there￾fore have major deficiencies. Noting this, some researchers have invested in non-human primates as a model sys￾tem [27–29], but this is an expensive and time consuming route, albeit perhaps self-evidently better than others. Ide￾ally, potential vaccines for a particular species should be tested in either the same animal, or at least a closely related species such as bovids for cattle or buffalo vaccines. Each existing model has its own problems and deficien￾cies, the details of which are beyond the scope of this paper to discuss in detail. However, a common limita￾tion is that many lab-based models are not exposed to the environmental stimuli experienced by free-living crea￾tures, whichinclude organisms suchas other bacteria,fungi and parasites, which alone may render their hosts defi- cient as models. In addition, most animal models are inbred strains. If we postulate that genetics is a major determinant of resistance or susceptibility [30–33], this alone makes such models tricky to understand, since the strain may be inherently resistant or susceptible. Even if not inbred, certain animal species may be innately resistant, where vaccination is unnecessary, or innately susceptible, being perhaps intractable to vaccination. This variation may of course also be found within an out-bred population, due to genetic variation. Recognizing the limitations in labo￾ratory based animal models and the importance of dealing with research close to the real problem, a number of groups have investigated larger animals such as deer or cattle for trial vaccination. These animals are arguably a good choice, since they are potentially susceptible, are at risk for disease, are economically important and impact on ecosystems. In addition, they present with large genetic and environ￾mental heterogeneity, which better reflects what we may expect in trials under real life conditions. For this reason, apart from laboratory-based animal models, a number of attempts to vaccinate free-living

P.D. van Helden, E.G. Hoal/ Comparative Immunology, Microbiology and Infectious Diseases 36(2013)287-294 larger mammals have been made [17-26, 34 Some suc- Immune system stimulants, which directly affect cytokine cess in badgers, possums, deer and cattle has been claimed response for example 5. 9. 11, 12, 45, or alternatively, target but total protection has not been evident. The advantage the phagosome [9, 35, or interfere with the host cell cycle is that in these cases, a controlled challenge can be given The use of a simple antigen may seem counterintuitive. to directly assess protection. However, it can always be since it is widely believed that humoral immunity may not argued that the challenge is very unlikely to reflect the real, be important in mycobacterial infection or disease. How field condition and therefore the evaluation outcome of ever, we do not know this for certain and immune boosters vaccination may not reflect the true efficacy of the vaccine. or vaccination with antigens alone have shown promising Currently, TB in animals is either dealt with by a test-and- effects [17, 18, 36, 42]. Hence the rationale to vaccinate first slaughter policy or, as is generally the case with wildlife with a live vaccine and follow with a boost of antigen Tri- in many parts of the globe, left unmanaged and untreated. als currently underway in primates [46]and humans may These two approaches are both unsatisfactory for a variety show whether this approach suggests any advantage. of reasons which will not be the subject of discussion in this paper. At this time, it would seem that the best theo etical options for control would be test and slaughter, or 5. Possible complications? vaccination. Neither is simple nor clear-cut, but neverthe- less they are options to consider. Since M. bovis is generally Perhaps one of the most distressing problems is that n introduced oralien invader"pathogen, it may exe the preferred location of pathogenic mycobacteria is intra- serious negative effects on ecosystems, which does requir attention. This is particularly important in free-ranging ani normal macrophage function. It is thus difficult to imag- mals, where treatment is not possible, particularly in the ine how we may overpower these hidden organisms. Quite case of wildlife apart from this problem, there are also many other prob lems to consider. For example the Koch phenomenon 4. Choice of vaccine candidates suggests possible complications for a good vaccine uberculosis: in endemic areas, many individuals m For nearly a century, the vaccine given(bCg) has been a been infected and if they are exposed to a strong antigen, live one. There are good intellectual reasons to consider the ensuing host response may result in the exacerbation a live vaccine, not least of which is that we know that of occult disease leading to severe toxicities [47] he live vaccine bcg has some effect and that it reflects If we use a new live vaccine, what danger will this po the pathogen in that it is an intracellular organism. Newer live vaccine candidates may come in the form of mod- individuals? We already know BCG is problematic in this ified BCG (for example strongly expressing a particular regard. antigen)[11,35-37, or attenuated Mycobacterium tuber- Even if we overcome these problems, there is evidence culosis constructs(deletion mutants)[20, 38-40 In this that any protection will wane[48, 49], which then begs the case, it is thought to be necessary to knock out at least 2 question of when to vaccinate for maxim genes to negate any possibility that a pathogenic strain may [10 and whether a booster will be necessary or effective. regain virulence. However, live vaccines may also be the All these difficulties are compounded by the lack of com- source of adverse effects: in immunocompromised humans prehensive information on the effect of dosage [23 ,use for example, a live vaccine can cause disease(BCgosis i of adjuvants and timing of prime or boost vaccination. In seen frequently in HIV positive neonates)7. However, addition, it seems that there will be species-specific dif- the same situation may not be the case in immunocom- ferences, making extrapolation risky. Finally, it is possible promised individuals vaccinated with double knock-out that vaccination may interfere with some diagnostic tests live attenuated vaccines. It was shown that such a vac- for TB, complicating diagnosis [50]. This is particularly rel- cine candidate elicited good responses in FIV+ and FIv evant for control of TB in domestic and production animals cats, albeit weaker respons in those that were flv+ at present. with no adverse events noted. This work suggests that live vaccination may be possible in immunocompromised hosts [41. Despite this finding there are good reasons 6. where should we target our intervention step? o explore non-viable preparations for a vaccine. Much ttention has been given to this, ranging from individ Generally, vaccination is considered to be a preventa- ual antigens[11, 12, 18, 42. DNa or heat killed organisms tive activity. However, one may envisage the possibilities or crude preparations of mycobacterial extracts (24, 43. of a pre-exposure vaccine as is currently done, or a post which may also lead to adverse responses (tried by robert exposure vaccine a therapeutic vaccine or one that will Koch, where the adverse reaction has become known suppress reactivation disease. The use of Mycobacterium the Koch phenomenon). Most work has probably been vaccae as a post-exposure vaccine has been tried, but not devoted to protein antigens, perhaps because these can be been convincingly demonstrated to be of importance for loned, expressed and produced relatively easily. However general use [51], although there is some evidence to sug there is published evidence in animal models to suggest gest improved conversion during antibiotic treatment in that lipids or carbohydrates also hold potential as car humans 51. but no efficacy on its own. One might also didates [22, 42, 44 ]. Finally, a new-generation vaccine that envisage that if we find that all active cases of TB lack targets hostresponse can be an option. Such vaccines can be an adequate concentration of a bi ecule. such as a

P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 289 larger mammals have been made [17–26,34]. Some suc￾cess in badgers, possums, deer and cattle has been claimed, but total protection has not been evident. The advantage is that in these cases, a controlled challenge can be given to directly assess protection. However, it can always be argued thatthe challenge is very unlikely to reflectthe real, or field condition and therefore the evaluation outcome of vaccination may not reflect the true efficacy of the vaccine. Currently, TB in animals is either dealt with by a test-and￾slaughter policy or, as is generally the case with wildlife in many parts of the globe, left unmanaged and untreated. These two approaches are both unsatisfactory for a variety of reasons which will not be the subject of discussion in this paper. At this time, it would seem that the best theo￾retical options for control would be test and slaughter, or vaccination. Neither is simple nor clear-cut, but neverthe￾less they are options to consider. Since M. bovis is generally an introduced or “alien invader” pathogen, it may exert serious negative effects on ecosystems, which does require attention. This is particularly importantinfree-ranging ani￾mals, where treatment is not possible, particularly in the case of wildlife. 4. Choice of vaccine candidates For nearly a century, the vaccine given (BCG) has been a live one. There are good intellectual reasons to consider a live vaccine, not least of which is that we know that the live vaccine BCG has some effect and that it reflects the pathogen in that it is an intracellular organism. Newer live vaccine candidates may come in the form of mod￾ified BCG (for example strongly expressing a particular antigen) [11,35–37], or attenuated Mycobacterium tuber￾culosis constructs (deletion mutants) [20,38–40]. In this case, it is thought to be necessary to knock out at least 2 genes to negate any possibility that a pathogenic strain may regain virulence. However, live vaccines may also be the source of adverse effects:in immunocompromised humans for example, a live vaccine can cause disease (BCGosis is seen frequently in HIV positive neonates) [7]. However, the same situation may not be the case in immunocom￾promised individuals vaccinated with double knock-out live attenuated vaccines. It was shown that such a vac￾cine candidate elicited good responses in FIV+ and FIV− cats, albeit weaker responses in those that were FIV+, with no adverse events noted. This work suggests that live vaccination may be possible in immunocompromised hosts [41]. Despite this finding there are good reasons to explore non-viable preparations for a vaccine. Much attention has been given to this, ranging from individ￾ual antigens [11,12,18,42], DNA or heat killed organisms or crude preparations of mycobacterial extracts [24,43], which may also lead to adverse responses (tried by Robert Koch, where the adverse reaction has become known as the Koch phenomenon). Most work has probably been devoted to protein antigens, perhaps because these can be cloned, expressed and produced relatively easily. However, there is published evidence in animal models to suggest that lipids or carbohydrates also hold potential as can￾didates [22,42,44]. Finally, a new-generation vaccine that targetshost response canbe anoption. Suchvaccines canbe immune system stimulants, which directly affect cytokine response for example [5,9,11,12,45], or alternatively,target the phagosome [9,35], or interfere with the host cell cycle. The use of a simple antigen may seem counterintuitive, since it is widely believed that humoral immunity may not be important in mycobacterial infection or disease. How￾ever, we do not know this for certain, and immune boosters or vaccination with antigens alone have shown promising effects [17,18,36,42]. Hence the rationale to vaccinate first with a live vaccine and follow with a boost of antigen. Tri￾als currently underway in primates [46] and humans may show whether this approach suggests any advantage. 5. Possible complications? Perhaps one of the most distressing problems is that the preferred location of pathogenic mycobacteria is intra￾cellular and pathogen survival is achieved by subverting normal macrophage function. It is thus difficult to imag￾ine how we may overpower these hidden organisms. Quite apart from this problem, there are also many other prob￾lems to consider. For example, the Koch phenomenon suggests possible complications for a good vaccine against tuberculosis: in endemic areas, many individuals may have been infected and if they are exposed to a strong antigen, the ensuing host response may result in the exacerbation of occult disease leading to severe toxicities [47]. If we use a new live vaccine, what danger will this pose to immunocompromised individuals, such as HIV positive individuals? We already know BCG is problematic in this regard. Even if we overcome these problems, there is evidence that any protection will wane [48,49], which then begs the question of when to vaccinate for maximum protection [10] and whether a booster will be necessary or effective. All these difficulties are compounded by the lack of com￾prehensive information on the effect of dosage [23], use of adjuvants and timing of prime or boost vaccination. In addition, it seems that there will be species-specific dif￾ferences, making extrapolation risky. Finally, it is possible that vaccination may interfere with some diagnostic tests for TB, complicating diagnosis [50]. This is particularly rel￾evant for control of TB in domestic and production animals at present. 6. Where should we target our intervention step? Generally, vaccination is considered to be a preventa￾tive activity. However, one may envisage the possibilities of a pre-exposure vaccine as is currently done, or a post￾exposure vaccine, a therapeutic vaccine, or one that will suppress reactivation disease. The use of Mycobacterium vaccae as a post-exposure vaccine has been tried, but not been convincingly demonstrated to be of importance for general use [51], although there is some evidence to sug￾gest improved conversion during antibiotic treatment in humans [51], but no efficacy on its own. One might also envisage that if we find that all active cases of TB lack an adequate concentration of a biomolecule, such as a

P.D. van Helden, E.G. Hoal/ Comparative Immunology, Microbiology and Infectious Diseases 36(2013)287-294 cytokine for example, then a vaccine that will deliver this A major problem for research in a slow onset disease could be considered as a therapeutic such as TB is the time needed for any trial. This suggests 7. Vaccine administration and delivery How do we evaluate vaccine efficacy in candidate vaccines without a huge trial taking years to measure out We do not yet know how changing given parameters comes? We know very little about immunity to TB. For of administration of any given candidate tB vaccine will xample, why/how does BCG protect, if and when it does? impact on its efficacy. We are not even certain that cur- There are difficult technical issues with trials, e.g. how long rent practices with respect to BCG vaccination are optimal do we wait to see if disease results? Can we use a surrogate For example, giving BCG by the oral, intra-dermal route marker[55], and if so how do we find one or more? Furthermore, if certain individuals are largely immune take"considerably. Although BCG is given as a live vac- vaccination is unnecessary in them. Forexample, the90%of cine, many bacilli in the vials as packaged are not live at the time of administration. thus, there is no exact controlled measure efficacy in a trial in humans or any animal species that is similar. the numbers that will need to be tested bacilli we have little idea of the effect of this uncontrolled may be massive, particularly so if we are trying to measure mix. Should we continue to give it as we do currently with BCG, i.e. a single administration at birth or later [ 6, 10, 52. 100% coverage because of logistical difficulties for exa or do we follow a"prime-boost"using ple, how do we identify vulnerable twice [52]? This approach has failed with BCG in humans although this will not be relevant to all species, particu- for those time points tried 52, but does not guarantee larly perhaps in free living animals. Finally, if individuals that it will fail at all time points. There is some evidence for species)are innately vulnerable, will vaccination ever (from a deer model (50) that a second vaccination of BCG work? may have strongly positive effects, but that the timing is Important. One of the strategies being tried at present is a prime- 9. Will vaccination make a significant impact? boost design(see Fig. 1). Given the strong possibility that at least for humans there will be enormous resistance or reluctance to abandon BCG entirely, the strategy can absence of immunosuppression, about 90% of infected or include priming with BCG, followed by boosting with an exposed humans will not develop active TB. This suggests rather, over-abundant choices, in deciding what the timing should be and what adjuvant to use. In this context, very quately with this infection. The key question therefore is little work on the effect of adjuvants on TB vaccination has is this innate and inviolate, or is it adaptive? If the latter been done, but this may be extremely important [42, 43 The results of this work suggest equivalence of non-live hen the future looks less promising unless we can mimic andidates or improved protection to live bCg under the this in susceptible individuals. At present, our knowledge conditions tested. This is extremely encouraging, since it uggests that susceptibility or resistance to TB is a func- ffers a route for non-live vaccines which makes manu tion of genetics, maternal exposure, exposure to other facture, quality control and logistics of delivery simpler, as agents such as helminths or non-tuberculous environmen- well as posing far less risk for immunocompromised indi- tal mycobacteria [56] or other diseases which affect the viduals(e.g HIV-positive humans and FIv-positive felines). immune system, such as diabetes. Therefore, there must as well as any potentially highly vulnerable animal species be some form of protective immunity in by far the major In addition, simultaneous vaccination may be an option. ity of humans, which is most likely largely innate and Although no enhancement of protection was obtained in most humans who experience an active episode of b dis- ported 53 ease can be cured by antibiotic treatment. However, we bserve that these individuals are not protected by this prior episode and remain highly susceptible and likely to 8. Clinical trials, evaluation and end-points develop a subsequent episode [57 We may hypothesize that this is a subset of inherently highly susceptible indi There are currently at least 12 candidate vaccines in trial viduals, who remain so. Given that no protection is gained [54 Some are subunit vaccines, using fusion or recom- after an episode of disease caused by the pathogen itself, binant proteins. Some have been shown to boost BCG there may be reason to doubt that we may be able to protect mmunity"in preclinical studies. One is a recombinant such highly vulnerable individuals by a vaccine. However BCG that produces a better CD4 or CDS response. a quite there is likely to be a further subset of individuals who different approach is to provide a vaccine that will affect are susceptible but can be protected by acquiring immu macrophage function(such as by enzymatic membrane- nity. The relative proportion of such subsets may differ erforation)[9, 35, lead to apoptosis, cell death, and across geographic regions or ethnic groups [30,58. which hereby"immunity", since evidence suggests that apopto- may explain the known differences in BCG efficacy and will is may be an important step in resistance to TB confound vaccine studies in different regions and groups

290 P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 cytokine for example, then a vaccine that will deliver this could be considered as a therapeutic. 7. Vaccine administration and delivery We do not yet know how changing given parameters of administration of any given candidate TB vaccine will impact on its efficacy. We are not even certain that cur￾rent practices with respect to BCG vaccination are optimal. For example, giving BCG by the oral, intra-dermal route or by multi-puncture can vary success rate of vaccination “take” considerably. Although BCG is given as a live vac￾cine, many bacilli in the vials as packaged are not live at the time of administration. Thus, there is no exact controlled dose oflive bacilli and simultaneous administration of dead bacilli. We have little idea of the effect of this uncontrolled mix. Should we continue to give it as we do currently with BCG, i.e. a single administration at birth or later [6,10,52], or do we follow a “prime-boost” using the same vaccine twice [52]? This approach has failed with BCG in humans for those time points tried [52], but does not guarantee that it will fail at all time points. There is some evidence (from a deer model [50]) that a second vaccination of BCG may have strongly positive effects, but that the timing is important. One of the strategies being tried at present is a prime￾boost design (see Fig. 1). Given the strong possibility that, at least for humans, there will be enormous resistance or reluctance to abandon BCG entirely, the strategy can include priming with BCG, followed by boosting with an antigen [18]. Even here there are complications, or perhaps rather, over-abundant choices, in deciding what the timing should be and what adjuvant to use. In this context, very little work on the effect of adjuvants on TB vaccination has been done, but this may be extremely important [42,43]. The results of this work suggest equivalence of non-live candidates or improved protection to live BCG under the conditions tested. This is extremely encouraging, since it offers a route for non-live vaccines which makes manu￾facture, quality control and logistics of delivery simpler, as well as posing far less risk for immunocompromised indi￾viduals (e.g. HIV-positive humans and FIV-positive felines), as well as any potentially highly vulnerable animal species. In addition, simultaneous vaccination may be an option. Although no enhancement of protection was obtained in cattle [17], in humans enhanced immunization has been reported [53]. 8. Clinical trials, evaluation and end-points There are currently atleast 12 candidate vaccines in trial [54]. Some are subunit vaccines, using fusion or recom￾binant proteins. Some have been shown to boost BCG “immunity” in preclinical studies. One is a recombinant BCG that produces a better CD4 or CD8 response. A quite different approach is to provide a vaccine that will affect macrophage function (such as by enzymatic membrane￾perforation) [9,35], lead to apoptosis, cell death, and thereby “immunity”, since evidence suggests that apopto￾sis may be an important step in resistance to TB. A major problem for research in a slow onset disease such as TB is the time needed for any trial. This suggests the question: How do we evaluate vaccine efficacy in candidate vaccines without a huge trial taking years to measure out￾comes? We know very little about immunity to TB. For example, why/how does BCG protect, if and when it does? There are difficult technical issues with trials, e.g. how long do we wait to see if disease results? Can we use a surrogate marker [55], and if so how do we find one or more? Furthermore, if certain individuals are largely immune, vaccination is unnecessary in them. For example,the 90% of humans who may be infected but never develop disease. To measure efficacy in a trial in humans or any animal species that is similar, the numbers that will need to be tested may be massive, particularly so if we are trying to measure and improvement over current BCG. If we cannot achieve 100% coverage because of logistical difficulties for exam￾ple, how do we identify vulnerable individuals to vaccinate, although this will not be relevant to all species, particu￾larly perhaps in free living animals. Finally, if individuals (or species) are innately vulnerable, will vaccination ever work? 9. Will vaccination make a significant impact? What do we observe to give us hope or conversely suggest that a vaccination strategy may not work? In the absence of immunosuppression, about 90% of infected or exposed humans will not develop active TB. This suggests that the immune system of these individuals copes ade￾quately with this infection. The key question therefore is: is this innate and inviolate, or is it adaptive? If the latter situation exists, then we can move forward. If the former, then the future looks less promising unless we can mimic this in susceptible individuals. At present, our knowledge suggests that susceptibility or resistance to TB is a func￾tion of genetics, maternal exposure, exposure to other agents such as helminths or non-tuberculous environmen￾tal mycobacteria [56], or other diseases which affect the immune system, such as diabetes. Therefore, there must be some form of protective immunity in by far the major￾ity of humans, which is most likely largely innate and may include some acquired characteristics. We note that most humans who experience an active episode of TB dis￾ease can be cured by antibiotic treatment. However, we observe that these individuals are not protected by this prior episode and remain highly susceptible and likely to develop a subsequent episode [57]. We may hypothesize that this is a subset of inherently highly susceptible indi￾viduals, who remain so. Given that no protection is gained after an episode of disease caused by the pathogen itself, there may be reason to doubtthat we may be able to protect such highly vulnerable individuals by a vaccine. However, there is likely to be a further subset of individuals who are susceptible but can be protected by acquiring immu￾nity. The relative proportion of such subsets may differ across geographic regions or ethnic groups [30,58], which may explain the known differences in BCG efficacy and will confound vaccine studies in different regions and groups

P.D. van Helden, E.G. Hoal/Comparative Immunology, Microbiology and Infectious Diseases 36(2013 )287-294 Protected Vaccinate Partial protection Vaccine Active TB case Infecte accine Exposed Boost not Not protected Protecte Immune Active TB Therapeutic ystem I factors not shown here are the type of vaccine used at any given a at any given administration, the timing involved or the dose The figure also there there is no active disease. These matters are discussed in the text of people. This variability may apply similarly to different as disease. However, in the case of possums, an alien species nimal species in New Zealand for example, we may want a vaccine that Once we have some idea of vaccine protection, we can presents transmission and not be concerned about the indi sk the next question: what coverage will be needed for the vidual animal and its health status. Our need for protection vaccination to work under the different scenarios? This will in wildlife and or domestic stock may differ, as might our also depend on the prevalence of disease and ultimate goal approach to maintenance hosts or spill-over species. Thus, of vaccination, which in turn will depend on its efficacy we may aim at eradication under some scenarios should a for a given individual, herd or species. It is highly unlike vaccine confer 100% protection, or control, which would b that we will achieve 100% coverage or protection, but at a an improvement over the current situation. The cost of vac certain point we will probably be able to attain adequate ination would have to be considered against potential gain herd immunity to reduce prevalence incidence and achieve in the DACYs in humans, agricultural losses in livestock or control or perhaps even eradication. ecosystem 10. Goal of vaccination 11. Will a new vaccine merely shift the population structure of mycobacteria? The goal or aim of vaccination may be quite different for different species. For humans, a highly mobile species, we We know that the relative proportion of certain strains would optimally want a vaccine that offers individuals full of M tuberculosis are waxing or waning in different regions protection against disease and preferably infection as well of the globe 59]. furthermore, there is good evidence to

P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 291 Fig. 1. A simplified scheme illustrating key points to consider for vaccination against TB. It is assumed here that at least some protection is conferred by vaccination. Additional factors not shown here are the type of vaccine used at any given administration, the timing involved, or the dose. The figure also does not illustrate the complication of a post-exposure vaccination event, where there is no active disease. These matters are discussed in the text. of people. This variability may apply similarly to different animal species. Once we have some idea of vaccine protection, we can ask the next question: what coverage will be needed for the vaccination to work under the different scenarios? This will also depend on the prevalence of disease and ultimate goal of vaccination, which in turn will depend on its efficacy for a given individual, herd or species. It is highly unlikely that we will achieve 100% coverage or protection, but at a certain point we will probably be able to attain adequate herd immunity to reduce prevalence incidence and achieve control or perhaps even eradication. 10. Goal of vaccination The goal or aim of vaccination may be quite different for different species. For humans, a highly mobile species, we would optimally want a vaccine that offers individuals full protection against disease and preferably infection as well asdisease.However,inthe case ofpossums, analienspecies in New Zealand for example, we may want a vaccine that presents transmission and not be concerned aboutthe indi￾vidual animal and its health status. Our need for protection in wildlife and/or domestic stock may differ, as might our approach to maintenance hosts or spill-over species. Thus, we may aim at eradication under some scenarios should a vaccine confer 100% protection, or control, which would be an improvement over the current situation. The cost of vac￾cination would have to be considered against potential gain in the DACYs in humans, agricultural losses in livestock or ecosystem damage in wildlife. 11. Will a new vaccine merely shift the population structure of mycobacteria? We know that the relative proportion of certain strains of M. tuberculosis are waxing or waning in different regions of the globe [59]. Furthermore, there is good evidence to

P.D. van Helden, E.G. Hoal/ Comparative Immunology, Microbiology and Infectious Diseases 36(2013)287-294 enable us to conclude that various species and strains of be the case, then is there any motivation to continue this mycobacteria cause variable responses in hosts [4, 30]. We line of research? On the other hand, if the bacterial load do not know for certain whether vaccination with BCG is a is drastically reduced, maybe mortality will be reduced if driver of this change. This is partly as a result of changing the animal can eventually overcome the infection itself and vaccination policies, incomplete coverage of vaccination perhaps transmission can be halted. Either way, whilst the and strain analysis and the use of different BCG strains, individual may not benefit much, from a public health pe all of which makes any analysis or comparison difficult or spective, it may be a justifiable intervention. Overall, whilst invalid. However, it is reasonable to suppose that vaccina- there is no clear winner in sight and the field is complex, tion may offer variable protection against different species there is reason to be opportunistic and continue to invest or strains of mycobacteria. Should this be the case, we in tB vaccines. may anticipate initial success with vaccination, followed by emergence or re-emergence of new dominant strains Acknowledgments to replace those out-competed by the vaccine. Thus, good surveillance after introduction of new vaccination strate Ms Louise Vos with invaluable help with the manuscript and colleagues for many stimulating discussions. 12. Future prospects References The omics"driven investigation of TB, the host [1] Colditz GA, Brewer TF, Berkey CS, Wilson ME, Burdick E, Fineberg HV. and host-pathogen interactions will improve our under- tanding of the disease dynamics, which is still poorly understood. Many of the newly elucidated perhaps fu [21 Rodrigues LC, Diwan VK, Wheeler JG. Protective effect of BCG against crum points of the disease process may present target a meta-analysis. or vaccine intervention, also of the therapeutic type. up1921541 data sets may be amenable to a systems biology approach 3] Zwerling A, Behr MA. Verma A, Brewer TF, Menzies D, Pai M. The [60], to study relevant pathways. Responses to TB infection are highly complex and it is unlikely that a simple solution es of different bcg strains in a murine model vaccine ill be found 2011:29:1519-26. Even if no dramatically improved new vaccine cand [51 date is devel The influence of BCG vaccine strain on the immune response agains vaccines is not yet fully exploited for humans and animals. We have not exhaustively explored options in terms of tim- [6] Pereira SM, Barreto ML, Pilger D, Cruz AA. 6.10485253], dosage[19,23].adj 224243], prime-boost or simultaneous vaccination for example. The in test (BCG-REVAC trial): a cluster-ran problem with newer candidates is the same: we should not 2011:12:300-6. top exploring delivery options too soon if we find some- thing that shows improvement over current BCG. Even if we use only one candidate, we need to fully explore HIV-uninfected children. Clinical Infectious Diseases 2006:42:548-58. parameters to optimize it. Amajor problem will be funding [8] Aagaard C, Hoang T, Dietrich J. Cardona P, Izzo A, Dolganov G, et al It is encouraging that BCG has some protective efficacy efore and after exposure. Nature Medicine 2011: 17: 189-94. Fact and fiction in tuberculosis vaccine researche On the other hand, recent work in a mouse model sug- gests that there is initial protection over the short term. [10 Kagina BMN. Abel B, Bowmaker M, Scriba T]. Gelderbloem S. Smit then later T-cells show markers characteristic of exhaus- nay result in an enhanced memory CD4 T cell response. Vaccine tion, followed by a bacterial load increase(l Orme, personal comm ) This may help to explain our findings whick [11] Dey B, Jain R, Khera A, Gupta UD, Katoch VM, Ramanathan VD, et al. showed increased vulnerability in humans with prior dis ease 57]. This is a potentially serious flaw in the idea that ing the dynamics of pulmonary vaccination in humans can work over the long term. Fur- [121 Parlane NA, Grage K Mifune J Basaraba R), Wedlock DN, Rehm BHA, thermore, a typical statement from the literature may read al. Vaccines displaying mycobacterial proteins on biopolyester this new candidate offers more efficient containment of late stage infection". In other words, there is a lower bac- 113] Tompkins DM, Ra DSL, Cross ML Aldwell FE, de Lisle gw terial load compared to controls at that same time point ination reduces the incidence of tuberculo- Thus far. the best candidates offer a reduced bacterial load sis in free-living brushtail possums. Veterinary Immunology and initially, but not protection[8]. Part of the problem with [14] Parsons SD. Cooper D, McCall AJ, McCall WA, Streicher EM, le Maitre vaccine candidate testing is that the bacterial load in the experimental animal is evaluated at some arbitrary time the diagnosis of Mycobacterium bovis infection in African buffaloes Syncerus caffer). Veterinary Immunology and Immunopathology point, which also means that the animal is euthanized and 2011:142:113-8 o subsequent measurements can take place. We have no [15] Munyeme M, Munanga M, Muma JB, D way of knowing whether the bacterial load will reach the gating effects of parasite ondition of the Kafue lechwe( Kobus leche kafuensis) in the Kafue same as the controls, but simply take longer. If this should sin. BMC Research Notes 2010: 3: 346

292 P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 enable us to conclude that various species and strains of mycobacteria cause variable responses in hosts [4,30]. We do not know for certain whether vaccination with BCG is a driver of this change. This is partly as a result of changing vaccination policies, incomplete coverage of vaccination and strain analysis and the use of different BCG strains, all of which makes any analysis or comparison difficult or invalid. However, it is reasonable to suppose that vaccina￾tion may offer variable protection against different species or strains of mycobacteria. Should this be the case, we may anticipate initial success with vaccination, followed by emergence or re-emergence of new dominant strains to replace those out-competed by the vaccine. Thus, good surveillance after introduction of new vaccination strate￾gies will be important. 12. Future prospects The “omics” driven investigation of TB, the host and host–pathogen interactions will improve our under￾standing of the disease dynamics, which is still poorly understood. Many of the newly elucidated, perhaps ful￾crum points of the disease process may present targets for vaccine intervention, also of the therapeutic type. Huge data sets may be amenable to a systems biology approach [60], to study relevant pathways. Responses to TB infection are highly complex and it is unlikely that a simple solution will be found. Even if no dramatically improved new vaccine candi￾date is developed, in our opinion the efficacy of current vaccines is not yet fully exploited for humans and animals. We have not exhaustively explored options in terms oftim￾ing [6,10,48,52,53], dosage [19,23], adjuvants [22,42,43], prime-boost or simultaneous vaccination for example. The problem with newer candidates is the same: we should not stop exploring delivery options too soon if we find some￾thing that shows improvement over current BCG. Even if we use only one candidate, we need to fully explore parameters to optimize it. A major problem will be funding, however. It is encouraging that BCG has some protective efficacy. On the other hand, recent work in a mouse model sug￾gests that there is initial protection over the short term, then later, T-cells show markers characteristic of exhaus￾tion,followed by a bacterial load increase (I. Orme, personal comm.). This may help to explain our findings which showed increased vulnerability in humans with prior dis￾ease [57]. This is a potentially serious flaw in the idea that vaccination in humans can work over the long term. Fur￾thermore, a typical statement from the literature may read “this new candidate offers more efficient containment of late stage infection”. In other words, there is a lower bac￾terial load compared to controls at that same time point. Thus far, the best candidates offer a reduced bacterial load initially, but not protection [8]. Part of the problem with vaccine candidate testing is that the bacterial load in the experimental animal is evaluated at some arbitrary time point, which also means that the animal is euthanized and no subsequent measurements can take place. We have no way of knowing whether the bacterial load will reach the same as the controls, but simply take longer. If this should be the case, then is there any motivation to continue this line of research? On the other hand, if the bacterial load is drastically reduced, maybe mortality will be reduced if the animal can eventually overcome the infection itself and perhaps transmission can be halted. Either way, whilst the individual may not benefit much, from a public health per￾spective, it may be a justifiable intervention. Overall, whilst there is no clear winner in sight and the field is complex, there is reason to be opportunistic and continue to invest in TB vaccines. Acknowledgments Ms. LouiseVos withinvaluablehelp withthemanuscript and colleagues for many stimulating discussions. References [1] Colditz GA, Brewer TF, Berkey CS, Wilson ME, Burdick E, Fineberg HV, et al. Efficacy of BCG vaccine in the prevention of tuberculosis. Meta￾analysis of the published literature. Journal of the American Medical Association 1994;271:698–702. [2] Rodrigues LC, Diwan VK, Wheeler JG. Protective effect of BCG against tuberculous meningitis and miliary tuberculosis: a meta-analysis. International Journal of Epidemiology 1993;22:1154–8. [3] Zwerling A, Behr MA, Verma A, Brewer TF, Menzies D, Pai M. The BCG world atlas: a database of global BCG vaccination policies and practices. PLoS ONE 2011;8:e1001012. [4] Kozak R, Behr MA. Divergence of immunologic and protective responses of different BCG strains in a murine model. Vaccine 2011;29:1519–26. [5] Ritz N, Dutta B, Donath S, Casalaz D, Connell TG, Tebruegge M, et al. The influence of BCG vaccine strain on the immune response against tuberculosis: a randomised trial.American Journal of Respiratory and Critical Care Medicine 2011;185:213–22. [6] Pereira SM, Barreto ML, Pilger D, Cruz AA, Sant’anna C, Hijjar MA, et al. Effectiveness and cost-effectiveness of first BCG vaccination against tuberculosis in school-age children without previous tuber￾culin test (BCG-REVAC trial): a cluster-randomised trial. Lancet 2011;12:300–6. [7] Hesseling AC, Rabie H, Marais BJ, Manders M, Lips M, Schaaf HS, et al. Bacille Calmette-Guérin vaccine-induced disease in HIV￾infected and HIV-uninfected children. Clinical Infectious Diseases 2006;42:548–58. [8] Aagaard C, Hoang T, Dietrich J, Cardona P, Izzo A, Dolganov G, et al. A multistage tuberculosis vaccine that confers efficient protection before and after exposure. Nature Medicine 2011;17:189–94. [9] Kaufmann SHE. Fact and fiction in tuberculosis vaccine research: 10 years later. Lancet 2011;11:633–40. [10] Kagina BMN, Abel B, Bowmaker M, Scriba TJ, Gelderbloem S, Smit E, et al. Delaying BCG vaccination from birth to 10 weeks of age may result in an enhanced memory CD4 T cell response. Vaccine 2009;27:5488–95. [11] Dey B, Jain R, Khera A, Gupta UD, Katoch VM, Ramanathan VD, et al. Latency antigen -crystalline based vaccination imparts a robust protection against TB by modulating the dynamics of pulmonary cytokines. PLoS ONE 2011;6:e18773. [12] Parlane NA, Grage K, Mifune J, Basaraba RJ, Wedlock DN, Rehm BHA, et al. Vaccines displaying mycobacterial proteins on biopolyester beads stimulate cellular immunity and induce protection against tuberculosis. Clinical and Vaccine Immunology: CVI 2012;19:37–44. [13] Tompkins DM, Ramsey DSL, Cross ML, Aldwell FE, de Lisle GW, Buddle BM. Oral vaccination reduces the incidence of tuberculo￾sis in free-living brushtail possums. Veterinary Immunology and Immunopathology 2009;276:2987–95. [14] Parsons SD, Cooper D, McCall AJ, McCall WA, Streicher EM, le Maitre NC, et al.Modificationofthe QuantiFERON-TBgold (in-tube) assay for the diagnosis of Mycobacterium bovis infection in African buffaloes (Syncerus caffer). Veterinary Immunology and Immunopathology 2011;142:113–8. [15] Munyeme M, Munang’andu HM, Muma JB, Nambota AM, Biffa D, Siamudaala VM. Investigating effects of parasite infection on body condition of the Kafue lechwe (Kobus leche kafuensis) in the Kafue basin. BMC Research Notes 2010;3:346.

P.D. van Helden, E.G. Hoal/Comparative Immunology, Microbiology and Infectious Diseases 36(2013 )287-294 16 Keet DF, Michel AL, Bengis RG, Becker P, van Dyk DS, van Vuuren and protection against challenge with with Mycobacterium bovis. [36] Dey B, Jain R, G an VD, Tyagi AK. A 17 DN, Aldwell FE, Vordermeier HM, Hewinson RG, Buddle and confers enhanced protection against BM. Protection against bovine tuberculosis induced by oral vac- 2011:6:e23360. ination of cattle with My m bovis bcg is not enhanced [37] Sweeney KA. by co-administration of atis induces potent bactericidal immunity against [18] Maue AC, Waters WR, Pa /C, etal. An ESAT-6: CFPl0 DNA vaccine ad [38] Waters WR, Palmer MV, Nonnecke B]. Thacker TC, Scherer CFC, Estes with virulent M. bovis vaccine 2007: 25: 4735-46. [19 de Klerk L Michel AL Bengis RG, Kriek NPl, Godfroid ]. Be accination failed Iffer) against experimental intratonsillar challenge with Mycobac [39] Sambandamurthy VK, Derrick SC, Hsu T, Chen B, Larsen MH, Estes DM, et al. Efficacy and immunogenicity of Mycobac DeltaRD1 against aerosol M. bovis infection in neonat a [40] Desel C, Dorhoi ental BCG 21 Lopez-Valencia G, Renter ta T, Williams JdJ. stimulating ination of type l and type 17 cytokine the protective efficacy F Mycob vaccine [41] Zimmerman DM, Waters WR, Lyashchenko KP Nonn lin veterinary Science 2010: 88: [22] Cross ML Henderson R], Lambeth MR, Buddle BM, Aldwell FE sinc in domes tability, efficacy, and palatability to brushtail possums (Tri- 42]H ladle M. Pawlowski A Schroder U, williams A Hatch ecula)in New Zealand. Journal of wildlife Diseases ulosis arabinomannan-protein conju- cine2003;21:4081-9 23 Buddle BM, Aldwell FE, de Lisle GW, vordermeier HM, Hewinson [43] Haile M, Schroder U. Hamasur B, Pawlowski A, Jaxmar T, Kallenius RG, Wedlock DN. Low oral BCG doses fail to protect cattle again nization with heat-killed Mycobacterium bovis bacille an experimental challenge with Mycobacterium bovis. Tuberculosis Calmette-Guerin(BCG) in Eurocine L3 adjuvant protects against 4 Garrido JM, an-Beck B, Minguijon E, Ballesteros C, lindo RC, et al. Protection against tuberculosis in Eurasian wild edades Infecciosas y Microbiologia Clinica 2011; 29(SuppL. ONE2011:6:e24905 [45] Lahey T, Mitchell BK, Arbeit RD, ShethS, Matee M, Horsburgh CR, et al. [25 Palmer MV, Thacker TC, Waters WR. Vaccination of white-tailed deer protection (Odocoileus virginianus) with Mycobacterium bovis bacillus Calmette Guerin Vaccine 2007: 25: 6589-97 ion. PLos one 2011: 6 e22074. [26] Buddle BM, Wedlock DN, Denis M. Progress in the development of [46] Rahman S, Magalhaes L Rahman J. Ahmed RK Sizemore DR, Scanga tuberculosis vaccines for cattle and wildlife. Infection and immunit 2006:112:191-200 cytolytic T cell response 27 Larsen MH, Biermann K Chen B, Hsu T, Sambandamurthy VK, Lack infected primates. Molecular Medicine 2012: 18: 647-58. [47] Rook GA, Stanford JL The Koch phenom sis vaccine candidates in ogy of tuberculosis. Current Topics in Microbiology and Immunology 28 Capuano Sv, Croix DA, Pawar S, Zinovik A, Myers A, Lin PL, [48 Rodrigues LC, Mangtani P, Abubakar L How does the level of BCG vac- uman M. tuberculosis infection Infection and Immunity 2003: 71: [49] Weir RE, Gorak-Stolinska P, Floyd S, Lalor MK, Stenson S, Branson K, induced by BCG vaccina 29 Lin PL Rodgers M, Smith L, Bigbee M, Myers A, Bigbee C, et al. Quanti- [50] Stringer LA. Wilson PR, Heuer C, Hunnam JC. Mackintosh CG Effect of 30] Di Pietrantonio T, Correa JA, Orlova M, Behr MA, Schurr E. Joint paratuberculosis on specificity of diagne effects of host genetic background and mycobacterial pathog usceptibility to infection. Infection and Immunity 2011: 79 ulosis in farmed red deer(Cervus elaphus). New Zealand veterinary 372-8. [51] Yang X, Chen O, liY, wu S Mycobacterium vaccae as adjuvant ther- [52] Randor sis in Malawi. Karonga Prevention Trial Grou in human genetic susceptibility to tuberculosis Tuberculosis(Edin- [53] Tchilian EZ, Ronan EO, de Lara C, Lee IN, Franken KLMC, Vorde ant CI. Simkin L Black GF, Stanley K, Hughes J, PLOS ONE2011:6:e27477 hyperendemic for tuberculosis. Journal of Experimental Medicine [54] Kupferschmidt K. Taking a new shot at a TB vaccine. Science 4 Waters W, Palmer M, Buddle B, Vordermeier H Bovine tuberculo- [55 Walzl G, Ronacher K, Hanekom W, Scriba T). Zumla A Immuno- arkers of tubercu [35] Sun R, Skeiky YAW, Izzo A, Dheenadhayalan V, Imam Z, Pen [56] Weir RE, Black GF, Nazareth B, Floyd S. Stenson S, Stanley C, et al. E, et al. Novel recombinant BCG expressing perfringolysin O and The influence of previous exposure to environmental mycobacte- he over-expression of key immunodominant antigens: pre-clinical ria on the interferon-gamma response to Bacille Calmette-Guerin

P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 293 [16] Keet DF, Michel AL, Bengis RG, Becker P, van Dyk DS, van Vuuren M, et al. Intradermal tuberculin testing of wild African lions (Pan￾thera leo) naturally exposed to infection with Mycobacterium bovis. Veterinary Microbiology 2010;144:384–91. [17] Wedlock DN, Aldwell FE, Vordermeier HM, Hewinson RG, Buddle BM. Protection against bovine tuberculosis induced by oral vac￾cination of cattle with Mycobacterium bovis BCG is not enhanced by co-administration of mycobacterial protein vaccines. Veterinary Immunology and Immunopathology 2011;144:220–7. [18] Maue AC, Waters WR, Palmer MV, Nonnecke BJ, Minion FC, Brown WC, et al.An ESAT-6:CFP10 DNAvaccine administered in conjunction with Mycobacterium bovis BCG confers protection to cattle chal￾lenged with virulent M. bovis. Vaccine 2007;25:4735–46. [19] de Klerk L, Michel AL, Bengis RG, Kriek NPJ, Godfroid J. BCG vaccination failed to protect yearling African buffaloes (Syncerus caffer) against experimental intratonsilar challenge with Mycobac￾terium bovis. Veterinary Immunology and Immunopathology 2010;137:84–92. [20] Waters WR, Palmer MV, Nonnecke BJ, Thacker TC, Scherer CFC, Estes DM, et al. Efficacy and immunogenicity of Mycobacterium bovis DeltaRD1 against aerosol M. bovis infection in neonatal calves. Vac￾cine 2009;27:1201–9. [21] Lopez-Valencia G, Renteria-Evangelista T, Williams JdJ, Licea￾Navarro A, Mora-Valle A De L, Medina-Basulto G. Field evaluation of the protective efficacy of Mycobacterium bovis BCG vaccine against bovine tuberculosis. Research in Veterinary Science 2010;88: 44–9. [22] Cross ML, Henderson RJ, Lambeth MR, Buddle BM, Aldwell FE. Lipid-formulated bcg as an oral-bait vaccine for tuberculosis: vaccine stability, efficacy, and palatability to brushtail possums (Tri￾chosurus vulpecula) in New Zealand. Journal of Wildlife Diseases 2009;45:754–65. [23] Buddle BM, Aldwell FE, de Lisle GW, Vordermeier HM, Hewinson RG, Wedlock DN. Low oral BCG doses fail to protect cattle against an experimental challenge with Mycobacterium bovis. Tuberculosis (Edinburgh) 2011;91:400–5. [24] Garrido JM, Sevilla IA, Beltrán-Beck B, Minguijón E, Ballesteros C, Galindo RC, et al. Protection against tuberculosis in Eurasian wild boar vaccinated with heat-inactivated Mycobacterium bovis. PLoS ONE 2011;6:e24905. [25] Palmer MV, Thacker TC, Waters WR. Vaccination of white-tailed deer (Odocoileus virginianus) with Mycobacterium bovis bacillus Calmette Guerín. Vaccine 2007;25:6589–97. [26] Buddle BM, Wedlock DN, Denis M. Progress in the development of tuberculosis vaccines for cattle and wildlife. Infection and Immunity 2006;112:191–200. [27] Larsen MH, Biermann K, Chen B, Hsu T, Sambandamurthy VK, Lack￾ner AA, et al. Efficacy and safety of live attenuated persistent and rapidly cleared Mycobacterium tuberculosis vaccine candidates in non-human primates. Vaccine 2009;27:4709–17. [28] Capuano SV, Croix DA, Pawar S, Zinovik A, Myers A, Lin PL, et al. Experimental Mycobacterium tuberculosis infection of cynomol￾gus macaques closely resembles the various manifestations of human M. tuberculosis infection. Infection and Immunity 2003;71: 5831. [29] Lin PL, Rodgers M, Smith L, Bigbee M, Myers A, Bigbee C, et al. Quanti￾tative comparisonof active andlatenttuberculosis inthe cynomolgus macaque model. Infection and Immunity 2009;77:4631. [30] Di Pietrantonio T, Correa JA, Orlova M, Behr MA, Schurr E. Joint effects of host genetic background and mycobacterial pathogen on susceptibility to infection. Infection and Immunity 2011;79: 2372–8. [31] Kleinnijenhuis J, Oosting M, Joosten LAB, Netea MG, Van Crevel R. Innate immune recognition of Mycobacterium tuberculosis. Clinical & Developmental Immunology 2011, http://dx.doi.org/10.1155/2011/405310. [32] MöllerM, Hoal EG.Currentfindings, challenges andnovel approaches in human genetic susceptibility to tuberculosis. Tuberculosis (Edin￾burgh) 2010;90:71–83. [33] Cobat A, Gallant CJ, Simkin L, Black GF, Stanley K, Hughes J, et al. Two loci control tuberculin skin test reactivity in an area hyperendemic for tuberculosis. Journal of Experimental Medicine 2009;206:2583–91. [34] Waters W, Palmer M, Buddle B, Vordermeier H. Bovine tuberculo￾sis vaccine research: historical perspectives and recent advances. Vaccine 2012;30:2611–22. [35] Sun R, Skeiky YAW, Izzo A, Dheenadhayalan V, Imam Z, Penn E, et al. Novel recombinant BCG expressing perfringolysin O and the over-expression of key immunodominant antigens; pre-clinical characterization, safety and protection against challenge with mycobacterium tuberculosis. Vaccine 2009;27:4412–23. [36] Dey B, Jain R, Gupta UD, Katoch VM, Ramanathan VD, Tyagi AK. A booster vaccine expressing a latency-associated antigen augments BCG induced immunity and confers enhanced protection against tuberculosis. PLoS ONE 2011;6:e23360. [37] Sweeney KA, Dao DN, Goldberg MF, Hsu T, Venkataswamy MM, Henao-Tamayo Marcela, et al. A recombinant Mycobacterium smeg￾matis induces potent bactericidal immunity against Mycobacterium tuberculosis. Nature Medicine 2011;17:1261–8. [38] Waters WR, Palmer MV, Nonnecke BJ, Thacker TC, Scherer CFC, Estes DM, et al. Failure of a Mycobacterium tuberculosis DeltaRD1 Delta￾panCD double deletion mutant in a neonatal calf aerosol M. bovis challenge model: comparisons to responses elicited by M. bovis Bacille Calmette Guerin. Vaccine 2007;25:7832–40. [39] Sambandamurthy VK, Derrick SC, Hsu T, Chen B, Larsen MH, Jalap￾athy KV, et al. Mycobacterium tuberculosis DeltaRD1 DeltapanCD: a safe and limited replicating mutant strain that protects immuno￾competent and immunocompromised mice against experimental tuberculosis. Vaccine 2006;24:6309–20. [40] Desel C, Dorhoi A, Bandermann S, Grode L, Eisele B, Kaufmann SHE, et al. DeltaureC hly+ induces superior protection over parental BCG by stimulating a balanced combination oftype 1 and type 17 cytokine responses. Journal of Infectious Diseases 2011;204:1573–84. [41] Zimmerman DM, Waters WR, Lyashchenko KP, Nonnecke BJ, Arm￾strong DL, Jacobs Jr WR, et al. Safety and immunogenicity of the Mycobacterium tuberculosis DeltalysA DeltapanCD vaccine in domes￾tic cats infected with feline immunodeficiency virus. Clinical and Vaccine Immunology: CVI 2009;16:427–9. [42] Hamasur B, Haile M, Pawlowski A, Schröder U, Williams A, Hatch G, et al. Mycobacterium tuberculosis arabinomannan–protein conju￾gates protect against tuberculosis. Vaccine 2003;21:4081–93. [43] Haile M, Schröder U, Hamasur B, Pawlowski A, Jaxmar T, Källenius G, et al. Immunization with heat-killed Mycobacterium bovis Bacille Calmette-Guerin (BCG) in Eurocine L3 adjuvant protects against tuberculosis. Vaccine 2004;22:1498–508. [44] Martín Montanés C, Gicquel B. New tuberculosis vaccines. Enfermedades Infecciosas y Microbiologia Clinica 2011;29(Suppl. 1):57–62. [45] Lahey T, Mitchell BK,Arbeit RD, Sheth S, Matee M, Horsburgh CR, et al. Polyantigenic interferon- responses are associated with protection from TB among HIV-infected adults with childhood BCG immuniza￾tion. PLoS ONE 2011;6:e22074. [46] Rahman S, Magalhaes I, Rahman J, Ahmed RK, Sizemore DR, Scanga CA, et al. Prime-boost vaccination with rBCG/rAd35 enhances CD8+ cytolytic T cell responses in lesions from Mycobacterium tuberculosis infected primates. Molecular Medicine 2012;18:647–58. [47] Rook GA, Stanford JL. The Koch phenomenon and the immunopathol￾ogy of tuberculosis. Current Topics in Microbiology and Immunology 1996;215:239–62. [48] Rodrigues LC, Mangtani P, Abubakar I. How does the level of BCG vac￾cine protection against tuberculosis fall over time? British Medical Journal 2011;343:d5974. [49] Weir RE, Gorak-Stolinska P, Floyd S, Lalor MK, Stenson S, Branson K, et al. Persistence of the immune response induced by BCG vaccina￾tion. BMC Infectious Diseases 2008;8:9. [50] Stringer LA, Wilson PR, Heuer C, Hunnam JC, Mackintosh CG. Effect of vaccination and natural infection with Mycobacterium avium subsp. paratuberculosis on specificity of diagnostic tests for bovine tuber￾culosis in farmed red deer (Cervus elaphus). New Zealand Veterinary Journal 2011;59:218–24. [51] Yang X, Chen Q, Li Y, Wu S. Mycobacterium vaccae as adjuvant ther￾apy to anti-tuberculosis chemotherapy in never-treated tuberculosis patients: a meta-analysis. PLoS ONE 2011;6:e23826. [52] Randomised controlled trial of single BCG repeated BCG or combined BCG and killed Mycobacterium leprae vaccine for prevention of lep￾rosy and tuberculosis in Malawi. Karonga Prevention Trial Group. Lancet 1996;348:17–24. [53] Tchilian EZ, Ronan EO, de Lara C, Lee LN, Franken KLMC, Vorder￾meier MH, et al. Simultaneous immunization against tuberculosis. PLoS ONE 2011;6:e27477. [54] Kupferschmidt K. Taking a new shot at a TB vaccine. Science 2011;334:1488. [55] Walzl G, Ronacher K, Hanekom W, Scriba TJ, Zumla A. Immuno￾logical biomarkers of tuberculosis. Nature Reviews Immunology 2011;11:343–54. [56] Weir RE, Black GF, Nazareth B, Floyd S, Stenson S, Stanley C, et al. The influence of previous exposure to environmental mycobacte￾ria on the interferon-gamma response to Bacille Calmette-Guérin

P.D. van Helden, E.G. Hoal/ Comparative Immunology, Microbiology and Infectious Diseases 36(2013)287-294 land and Northern malawi. clinical and ntal Immunology 2006: 146: 390-9. [571 Verver S, Warren RM, Beyers N, Richardson M, van der Spuy GD. [59] van der Spuy GD, Kremer K, N, Dunbar R. nerican Journal lade-specific path haracteristics. Tuberculosis of Respiratory and Critical Care Medicine 2005: 171: 1430-5 [58 Lalor MK, Ben-Smith A, Gorak-Stolinska P, Weir RE, Floyd S. Blitz [60 Rappuoli R, Aderem A A 2020 vision for vaccines against HIV tuber- R, et al. Population differences in immune responses to Bacill culosis and malaria. Nature 2011: 473: 463-9

294 P.D. van Helden, E.G. Hoal / Comparative Immunology, Microbiology and Infectious Diseases 36 (2013) 287–294 vaccination in Southern England and Northern Malawi. Clinical and Experimental Immunology 2006;146:390–9. [57] Verver S, Warren RM, Beyers N, Richardson M, van der Spuy GD, Borgdorff MW, et al. Rate of reinfection tuberculosis after successful treatment is higher than rate of new tuberculosis. American Journal of Respiratory and Critical Care Medicine 2005;171:1430–5. [58] Lalor MK, Ben-Smith A, Gorak-Stolinska P, Weir RE, Floyd S, Blitz R, et al. Population differences in immune responses to Bacille Calmette-Guérin vaccination in infancy. Journal of Infectious Dis￾eases 2009;199:795–800. [59] van der Spuy GD, Kremer K, Ndabambi SL, Beyers N, Dunbar R, Marais BJ, et al. Changing Mycobacterium tuberculosis population highlights clade-specific pathogenic characteristics. Tuberculosis 2009;89:120–5. [60] Rappuoli R, Aderem A. A 2020 vision for vaccines against HIV tuber￾culosis and malaria. Nature 2011;473:463–9