<< Back To Home

DEVELOPING VACCINES TO PREVENT SUSTAINED INFECTION WITH MYCOBACTERIUM TUBERCULOSIS: CONFERENCE PROCEEDINGS: NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASES, ROCKVILLE, MARYLAND USA, NOVEMBER 7, 2014

Monday, 8th of June 2015

This article is one of four on the subject appearing in the June 2015 issue of Vaccine, which those interested in TB vaccine development may wish to consult.

DEVELOPING VACCINES TO PREVENT SUSTAINED INFECTION WITH MYCOBACTERIUM TUBERCULOSIS: CONFERENCE PROCEEDINGS: NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASES, ROCKVILLE, MARYLAND USA, NOVEMBER 7, 2014

Vaccine

Volume 33, Issue 26, 12 June 2015, Pages 3056–3064

Vaccine Prevention of Sustained Mycobacterium tuberculosis Infection Summary Group¹
Aeras, Rockville, MD, USA

doi:10.1016/j.vaccine.2015.03.061

Abstract

On November 7, 2014, Aeras and the National Institute of Allergy and Infectious Diseases convened a conference entitled “Vaccine Prevention of Sustained Mycobacterium tuberculosis Infection.” The purpose of this meeting was to explore the biologic plausibility, potential public health and economic impact, and regulatory feasibility in attempting to develop a vaccine to prevent sustained infection with Mycobacterium tuberculosis (Mtb). Currently there are two main goals for tuberculosis (TB) vaccine development, to develop a vaccine that could serve as a booster to Bacille Calmette-Guérin (BCG) vaccination and prevent active TB in adolescents and adults, and to develop an improved vaccine to replace BCG in infants. Although prevention of sustained Mtb infection is being used as a proof of biological activity for vaccines in mid-Phase 2 development, there currently are no plans for pursuing a prevention of Mtb infection licensure indication for TB vaccines. Ultimately, pursuing a prevention of sustained Mtb infection indication for TB vaccines, in parallel with ongoing efforts to develop vaccines to prevent active TB disease, was deemed a potentially important effort, but would require further resources, particularly to improve diagnostic assays, to increase the regulatory feasibility of this endeavor.

Abbreviations

ADCC, antibody directed cellular cytotoxicity;
AIR, Airborne Infections Research Facility;
BCG, Bacille Calmette-Guérin;
CFP-10, 10 kDa culture filtrate protein;
CFU, colony forming units;
CM, cynomolgous macaques;
CMI, cell mediated immunity;
CT, computerized tomography;
ELISA, enzyme-linked immunosorbant assay;
ESAT-6, 6 kDa early secreted antigenic target;
FDA, Food and Drug Administration;
FDG, 18-F fluorodeoxyglucose;
GP, guinea pig;
IFN-γ, interferon-gamma;
IGRA,interferon-gamma release assay;
IP-10, IFN-gamma inducible protein 10;
IU, international units;
MAITs,mucosal associated invariant T-cells;
MDR, multidrug-resistant;
Mtb, Mycobacterium tuberculosis;
NHP,non-human primate;
NTM, non-tuberculous mycobacteria;
PET, positron emission tomography;
PoI,prevention of Mtb infection;
QFT, QuantiFERON Gold In-Tube interferon gamma release assay;
TB,tuberculosis;
TNF-1, tumor necrosis factor-1;
TPP, target product profile;
TST, tuberculin skin test;
VE,vaccine efficacy;
WHO, World Health Organization;
XDR, extensively drug-resistant
Tuberculosis vaccine;
Mycobacterium tuberculosis;
Vaccine regulation;
Vaccine endpoints

Keywords

1. Introduction

Dr. Lewis Schrager, Vice President, Scientific Affairs, Aeras. Rockville, Maryland, USA

Developing new vaccines for tuberculosis (TB) remains a global health priority. Currently there are two main goals for TB vaccine development, to develop a vaccine that could serve as a booster to Bacille Calmette-Guérin (BCG) vaccination and prevent active TB in adolescents and adults, and to develop an improved vaccine to replace BCG in infants. Although prevention of sustained Mycobacterium tuberculosis(Mtb) infection is being used as a proof of biological activity for vaccines in mid-Phase 2 development, there currently are no plans for pursuing a prevention of Mtb infection licensure indication for any TB vaccines. One reason for this is that the biological plausibility of such an effort has been questioned given that, in general, vaccines do not create sterilizing immunity and thereby prevent infection, but more often prevent disease after a targeted microorganism has initiated infection and gone through at least a few replication cycles. Additionally, it has been questioned whether a vaccine to prevent Mtb infection would actually prove to be useful from a public health or economic perspective, as approximately ninety percent of Mtb-infected individuals never develop active TB disease. There also have been concerns as to the regulatory feasibility of seeking a prevention of infection licensure indication for TB vaccines, as the definitive endpoint, Mtb infection, would be determined by the results of a diagnostic assay, rather than by a clinical outcome as generally preferred by regulatory authorities, given that Mtb infection cannot be identified based on clinical symptomatology.

This conference sought to address each of these issues. The first part of the initial session was dedicated to defining what is meant by prevention of sustained Mtb infection, differentiating this concept from efforts to create sterilizing immunity against Mtb via vaccination. Modeling of the likely impact of exposure intensity and magnitude on vaccine efficacy, and of the potential public health and economic impact of vaccines capable of preventing sustained Mtb infection, was followed by presentations addressing the biological plausibility of this approach, with information coming from human natural history and epidemiology studies, and from animal models of Mtb infection. The second session focused upon the challenges to developing improved assays to diagnose Mtb infection, the regulatory pathways potentially available if a prevention of sustained Mtb infection were to be pursued for a TB vaccine, and the attributes of an optimal vaccine to prevent sustained Mtb infection were explored. A concluding panel discussion sought to summarize the conference discussions regarding the biological plausibility, potential public health impact, and regulatory feasibility of pursuing a prevention of sustained Mtb infection indication for a TB vaccine.

2. Session 1: prevention of sustained infection with Mycobacterium tuberculosis: definitions, impact and feasibility

Prevention of SustainedMycobacterium tuberculosisInfection: Defining the Term. Dr. Edward Nardell, Division of Global Health Equity, Brigham and Womens Hospital, Harvard Medical School, Boston, Massachusetts, USA

Dr. Edward Nardell addressed the question of how best to define “sustained Mtb infection,” a critically important issue when attempting to develop vaccines for this purpose.

In contrast to the widely-held concept that infection with Mtb is life-long, Dr. Nardell noted multiple examples of transient Mtb infection in humans, as reflected by tuberculin skin test (TST) or interferon-gamma release assay (IGRA) reversions [1] as well as reversions seen in the natural transmission human-to-guinea pig model [2]. Laboratory exposure to aerosol delivery of 20–50 colony forming units (CFU) of Mtb in the highly susceptible guinea pig model resulted in uniform infection and death that varied by the virulence of the strain [3]. A natural infection model in guinea pigs, however, resulted in a spectrum of outcomes including apparent latency, reversions and reinfection [4]. This natural infection model was established in the Airborne Infections Research Facility (AIR) in South Africa, a six-bed, multidrug-resistant tuberculosis inpatient hospital ward in which 362 guinea pigs were continuously exposed to patient exhaust air over a period of four months. Mtb infection was assessed with monthly serial TSTs, where the cutoff point for a positive TST was greater than or equal to six millimeters (mm) followed by evaluation of histological disease progression at necropsy. After four months, 91 (25%) of the animals had no evidence of infection, with 271 (75%) infected. Reversion from a positive to negative TST was seen in 53 (15%) of the animals, and was only seen in animals with a TST reaction of less than 14 mm. None of those animals progressed to active disease. Thirty-three (62%) of the 53 reverted animals became re-infected by the end of the study due to ongoing exposure. Fifty-four (15%) of all infected guinea pigs subsequently developed active TB disease.

The results from the AIR study are consistent with the spectrum of possible outcomes following human Mtb exposure [5]. These outcomes may be due to elimination of infection via innate immunity, prior to the priming of antigen-specific T-cells; elimination of infection in association with engagement of acquired immunity and associated T-cell priming; control of infection with some bacteria persisting in non-replicating forms; bacteria replication maintained at a subclinical level by appropriate immune responses; and active disease.

Dr. Nardell emphasized that much of the clinical disease attributed to reactivation of latent infection in high exposure settings is likely to be due to exogenous reinfection, as shown by studies done in the high burden settings of homeless shelters [6] and gold mines [7]. Exogenous reinfection is an alternative pathway versus endogenous reactivation to cavitary TB, and continuous ongoing transmission, repeated exposure and reinfection in high burden settings represents a major obstacle to tuberculosis control [8] and [9]. Studies of reinfection suggest that prior TB disease is a risk factor for reinfection, with recurrence rates found to be much higher than incidence rates for new TB infections [10], [11] and [12]. Accordingly, developing a vaccine capable of preventing the establishment of sustained Mtb infection, thereby eliminating the cycle of infection leading to disease and increased susceptibility to re-infection among those who become infected with Mtb, would represent a major contribution to Mtb control efforts.

Vaccine Prevention ofMycobacterium tuberculosisInfection: the Potential Impact of Exposure Intensity and Magnitude. Dr. Steven Self, Department of Biostatistics, Vaccine and Infectious Disease Institute, Fred Hutchinson Cancer Research Center, University of Washington, Seattle, Washington, USA

Dr. Steven Self discussed a mathematical modeling analysis of the potential impact of Mtb exposure intensity and magnitude on vaccine efficacy (VE) at the population level. While Mtb infection is clearly not a surrogate for active TB disease, Dr. Self noted that it may represent a more valid endpoint for assessing TB vaccine biological activity than immunogenicity endpoints in the absence of defined correlates of protection from Mtb infection or TB disease.

The determination of VE can become complicated when the vaccine effect is heterogeneous across the vaccinated population [13], as is the case for TB. Not every Mtb infection results in disease, and it is unclear whether a vaccine developed to prevent infection with Mtb would protect those who would otherwise naturally advance to active disease as opposed to those who if infected would never advance to active disease.

Dr. Self suggested that the effects of different levels of exposure to the pathogen on VE at the population level may be more subtle and nuanced than previously recognized. A model calibrated to the known Mtb infection rate in a population could provide insight about possible vaccine-induced reduction of Mtb transmission in situations of differing degrees of Mtb exposure. In this model, exposure intensity and exposure magnitude are key variables that help describe the nature of Mtb exposure. Exposure intensity reflects the number of repeated exposure events over time. Exposure magnitude refers to the variability of infectious potential per event, which is related to variations in the number of droplet nuclei deposited in the lung as well as infectious differences between Mtb strains. A family of curves can be generated that represent the probability of infection with different magnitudes of exposure, depending on a parameter that defines the probability that a single exposure event of a certain magnitude will lead to a sustained TB infection. This model suggests that higher levels of Mtb exposure in a population may not substantially reduce the potential efficacy of a vaccine developed to prevent Mtb infection in that population [15]. This outcome runs counter to concerns that vaccines developed to prevent Mtb infection would be likely to have diminished effect in populations with elevated Mtb transmission rates, and provides support for developing vaccines for this purpose.

Prevention of SustainedMycobacterium tuberculosisInfection: Modeling Epidemiologic and Economic Impact. Dr. Richard G. White, TB Modeling Group, Department of Infectious Disease Epidemiology, London School of Hygiene and Tropical Medicine, London, UK

What are the potential public health and economic benefits of a novel vaccine licensed for the prevention of sustained Mtb infection? Dr. Richard White presented his review of modeling and economic publications that addressed these questions.

Ziv et al. [14] explored the potential public health impact of pre- and post- infection vaccines in high TB incidence countries. They demonstrated that pre-infection vaccines that reduced susceptibility to infection would more quickly reduce the number of new infections than post-infection vaccines, while post-infection vaccines would be more likely to have a more rapid impact on the reduction of TB disease. Within 30 years, however, pre-and post-exposure vaccines may be equally effective at reducing TB disease. For a vaccine to contribute to global control of TB, the authors suggested that both pre- and post-infection protective functions are needed.

Dye and Williams [15] modeled the combination of diagnostics, drugs and vaccines needed to achieve the WHO target of TB elimination as a public health problem by 2050. Their models predict that an effective pre-infection vaccine would have a large impact, but that a post-infection vaccine or treatment that reduces progression to disease in latently infected individuals would be essential to reach the 2050 WHO TB elimination goal.

In addressing the potential economic benefits from novel TB vaccines, Dr. White cited work by Bishai and Mercer [16] which examined the global economic benefits of a novel TB vaccine with combined pre-and post-infection efficacy, capable of reducing the risk of Mtb reinfection and progression to TB disease. These authors found that improved TB vaccines would have immediate substantial financial value to most populations in the world, with benefits to 1 billion people in prevented medical spending at a cost of $25 per vaccine dose, and a prediction of large societal benefits to 3.5 billion people even at a cost of $135 per vaccine dose.

Also cited was modeling of the impact and cost-effectiveness of new TB vaccines in low-and middle-income countries, undertaken by Knight and colleagues [17]. Over the 2024–2050 time horizon, vaccines targeted at adolescents and adults would have a far greater impact on TB control than vaccines targeted at infants. Even with relatively high vaccine prices and low efficacy and short duration, TB vaccines for adolescents and adults could be cost effective. Ditkowsky and Schwartzman [18] demonstrated that, in a very high burden South Africa setting, an effective pre-infection BCG booster vaccine would result in both societal cost savings and a reduction in active tuberculosis cases, even at the lowest efficacy considered.

In conclusion, Dr. White noted that both pre-infection Mtb vaccines aimed at preventing sustained Mtb infection, and post-Mtb infection vaccines geared to prevent active TB disease, will be needed to reach the WHO goals of elimination of TB as a public health problem by 2050. Additionally, economic modeling studies consistently show that improved TB vaccines would result in substantial global cost savings.

Prevention of SustainedMycobacterium tuberculosisInfection: Lessons from Human Natural History and BCG Studies. Dr. Ajit Lalvani, Faculty of Medicine, National Heart and Lung Institute, Imperial College, London, UK

Natural history studies of individuals who do not develop apparent infection with Mtb despite ongoing exposure may provide an important opportunity to understand the factors that protect humans from infection, and pursue these to vaccine development. Dr. Ajit Lalvani focused attention on two distinct epidemiological settings where individuals have remained uninfected. The first setting involves intense, sustained exposure, usually to a single tuberculosis index case, for a finite period of time (weeks to months) e.g. within the household [19] and [20] or during institutional outbreaks [21] and [22]. Here, between 30% and 60% of individuals become TST positive [19] and [23]. The second setting involves repeated occupational exposure to different index cases over many years, which may occur for health care workers [24]. In the latter situation, approximately 10% to 30% of exposed individuals convert to TST positivity [25].

Dr. Lalvani described a prospective, community-based study of children in Turkey that provided the first evidence that BCG immunization protects against Mtb infection [20]. Risk factors for Mtb infection in 979 children from Istanbul, Turkey with household TB contacts were investigated using the IGRA and TST. Here, independent risk factors for Mtb infection were found to be an increasing number of index patients in the household, being the child of the index patient, increasing age up to 16 years, and the absence of a BCG scar. The presence of a BCG scar, definitive evidence of past BCG immunization, was associated with a relative risk reduction of Mtb infection of 24% and a relative risk reduction of 92% for tuberculosis disease [20]. These results represented the first evidence that a TB vaccine can provide resistance to Mtb infection.

In a 2014 systematic review and meta-analysis of 14 studies on the effect of BCG vaccination in children with a total of 3855 participants, BCG again was found to protect against Mtb infection as well as progression from infection to disease [26]. When VE was determined using the six highest quality studies, a VE of 27% as detected by IGRA against Mtb infection was found, as compared with a VE of 71% against active TB [26]. Evidence of BCG-induced protection, however, only was seen in studies conducted above 40° latitude.

Clearance or containment of Mtb with little or no involvement of the adaptive immune system, either through the action of innate immune responses occurring at the mucosal barrier or locally in the pulmonary parenchyma, represents a type of immune response to Mtb exposure that may lead to resistance to infection while maintaining a negative IGRA response. Clearance and containment of Mtb infection, which may involve transient peripheral adaptive priming, could lead to an IGRA positive to negative reversion. In comparison, it is through the development of a sustained Mtb infection, accompanied by persistent peripheral adaptive priming, that an IGRA positive, latent TB infection develops. Multiple risk factors may determine the outcome of exposure, including two interesting recently implicated factors, Vitamin D levels and host genetics. Sufficient Vitamin D levels may be a protective factor for Mtb infection [27]. Genome-wide association studies in a South African cohort revealed a genetic locus associated with persistent TST negativity in an endemic area [25].

Dr. Lalvani stated that, as of now, it is unclear if IGRA reversion from positive to negative represents abrogation of an initial Mtb infection prior to its establishment. IGRA reversion typically is associated with negative or borderline positive TST results and a low initial IGRA response. The possibility of persistent Mtb infection in IGRA-reverted humans, however, cannot be ruled out. Alternative explanations of IGRA reversion may include persistent Mtb infection with TB-specific T-cell frequencies either at frequencies below the level of detection (<20 per million) and/or confined to the site of latent infection. Ultimately, only long-term follow-up of cohorts to determine the risk of progression to TB disease among IGRA reverters will conclusively answer this important question.

Feasibility of Preventing SustainedMycobacterium tuberculosisInfection: Clues from Animal Models. Dr. JoAnne Flynn, Department of Microbiology and Molecular Genetics, University of Pittsburgh School of Medicine, Pittsburgh, Pennsylvania, USA

Dr. JoAnne Flynn described efforts to develop new non-human primate (NHP) models of acute Mtb infection to more accurately assess the feasibility of developing vaccines to prevent sustained Mtb infection. Most human infections result from repeated exposures to very low inocula of Mtb over time. In contrast, most vaccine challenge studies in NHPs have utilized very high, non-physiologic doses of Mtb, often in the range of 500–3000 CFUs, to assess the protective effect of vaccine candidates. These overwhelming doses provide little chance to accurately assess the ability of vaccines to prevent either sustained Mtb infection or TB disease.

Efforts are ongoing to develop NHP models of Mtb infection following low-inocula Mtb exposure, sometimes as low as 10 CFUs inoculated either intrabronchially via bronchoscope or through aerosol exposure, more closely approximating physiological Mtb exposure in humans. These models are being used both to study the natural history of the establishment of Mtb infection and to serve as challenge models to assess the potential efficacy of new Mtb vaccines. Much of this work has been done in cynomolgous macaques (CMs), given their propensity for developing the full range of Mtb-related outcomes seen in humans, ranging from latency to full-blown pulmonary and extrapulmonary tuberculosis[28]. Important information already has been garnered from low dose natural history studies in CMs. One key finding was that bacterial killing within individual granulomas varies widely within the same host and is not seen before four weeks after infection [29]. Additionally, while granulomas are capable of killing Mtb, even in CMs that ultimately develop active TB, granulomas taken from CMs that develop sustained, latent TB infection are significantly more likely to be sterile than those obtained from monkeys that developed active TB [29]. In both CMs that developed latent infection and those that went on to active TB, increased numbers of sterilized granulomas developed over time, indicating that nearly every CM has the ability to sterilize granulomas. The first six weeks of infection appears to be a critical time in Mtb infection and may predict the outcome of future infection and disease. Animals with latent infection have fewer lesions at three weeks and do not develop new lesions during the following three weeks [30].

Dr. Flynn described the potential for utilizing positron emission tomography (PET) in combination with computerized tomography (CT) imaging in vaccine trials to assess the events occurring in the lung and regional lymph nodes following the administration of an Mtb challenge. Utilizing combined PET/CT imaging holds the potential of eliciting possible outcomes of an effective vaccine including fewer granulomas established in the early phase of infection, no dissemination of granulomas, and lower overall inflammation as measured by 18-F fluorodeoxyglucose (FDG) avidity. She and her collaborators currently are testing PET/CT to assess efficacy in NHP vaccine trials.

Dr. Flynns preliminary data suggest that protection by vaccines can result in lower inflammation in the lungs, with reduced dissemination of early granulomas, compared to unvaccinated NHPs. Her previously published work indicated that CMs administered a subunit vaccine did not reactivate after anti-TNF treatment [31], and that this was associated with sterilized lymph nodes. Future studies, with more aggressive vaccination regimens, perhaps including an aerosol component along with an injected vaccine, may hold the potential for further interruption of the early events in the establishment of Mtb infection.

Establishment ofMycobacterium tuberculosisInfection: Human to Guinea Pig Natural Infection Model. Dr. Diane Ordway, Assistant Professor, Mycobacteria Research Laboratory, Department of Microbiology, Immunology and Pathology, Colorado State University, Fort Collins, Colorado, USA

Most of our understanding of the initial steps in the establishment of Mtb infection come from studies in which animals are infected with laboratory-grown cultures of Mtb, usually through aerosol exposure or direct, bronchoscopic administration into the lungs. In her presentation, Dr. Diane Ordway stressed that important differences exist between these laboratory infection strategies from “natural infection” strategies, animal models of Mtb infection in which infection is allowed to occur naturally through inhalation of ambient air laden with Mtb derived from either human patients, or from experimental animals, with active TB disease.

Given the importance of studying natural infection to understand events occurring around the time of the establishment of sustained Mtb infection, efforts have been made to advance the development of natural Mtb infection models. Dr. Ordway described one such collaborative effort to establish a human to guinea pig (GP) model of natural Mtb infection through the creation of the Airborne Infections Research Facility (AIR), a facility located in Witbank, Mpumalanga Province, South Africa where air from the rooms of up to six patients with active TB disease is circulated through an adjoining facility in which susceptible guinea pig test animals are housed (discussed by Dr. Nardell earlier in this conference). In this facility, naïve, BCG-vaccinated, and sham-vaccinated GPs were exposed to air circulated from patients with drug sensitive TB, and compared to naïve, BCG-vaccinated and sham-vaccinated GPs exposed to circulated air from patients with multidrug-resistant (MDR) and extensively drug resistant (XDR) TB infections. Experiments to assess possible differences in susceptibility between animals exposed to these different Mtb strains are ongoing [32], [33] and [34].

The results of the natural infection model described by Dr. Ordway suggest that the establishment of sustained Mtb infection is achievable in the human to guinea pig natural infection model, and holds the potential to provide important contributions in efforts to assess TB vaccines for their ability to prevent sustained Mtb infection. Dr. Ordway noted, however, that the full value of natural history experiments only will come to fruition when more sensitive tools to detect initial Mtb infection and reinfection are developed.

Prevention of SustainedMycobacterium tuberculosisInfection: Immunological Considerations. Dr. Joel Ernst, Director of Infectious Disease and Immunology, New York University School of Medicine, New York, NY, USA

Dr. Joel Ernst addressed three critical questions: (1) What are the early events that precede the development of adaptive immunity to Mtb (where are the bacteria during the initial stage of Mtb infection)? (2) Can monocytes be programmed to respond to early Mtb infection and, if so, what are the implications of this strategy? (3) Can human genetic studies contribute to discovering novel vaccine strategies to prevent sustained Mtb infection?

In the early stage of infection, mouse adoptive transfer experiments of CD4+ T-cells that recognize ESAT-6 suggest that Mtb organisms are sequestered where they are not recognized by T-cells [35]. These experiments demonstrate that these adoptively transferred ESAT-6-specific T-cells administered one day before Mtb challenge do not have an effect on Mtb replication rates until after the first week of infection has passed, suggesting Mtb sequestration until that time.

Experiments were conducted to identify sites of cellular localization of Mtb in the first fourteen days after aerosol Mtb infection in a mouse model. High-speed cell sorting was applied, sorting macrophages, lung dendritic cells, recruited macrophages and neutrophils in an attempt to precisely localize the cellular sites of early Mtb sequestration. Three different Mtb infecting strains were used; an H37Rv strain, an H37RvΔRD1 strain, and a Beijing 4334 strain. This investigation suggested that all three strains spread through diverse cell types, with all initially infecting alveolar macrophages during the first six days of infection. By Day 14, both the H37Rv strain and the Beijing strain predominantly infected neutrophils, with dendritic cells and alveolar macrophages the next most frequently infected cell types. The Beijing strain demonstrated both particularly rapid growth from the time of initial infection, and a more efficient capability of spreading beyond alveolar macrophages, mainly to neutrophils but also to dendritic cells. In contrast, the H37RvΔRD1 strain mostly remained within alveolar macrophages at 14 days post-infection. In all cases, recruited macrophages represented the smallest proportion of cells infected with Mtb at 14 days post-infection.

These results suggest the need for vaccine-induced immunological intervention before natural, adaptive T-cell responses to Mtb develop. A challenge, however, is that the information presented above suggests that early infection and replication appears to be occurring in the alveoli, an immunologically protected compartment, and therefore a safe place for Mtb growth. Non-traditional T-cell and mucosal associated invariant T-cells (MAITs) might have a chance to reach the alveolar compartment, although MAITs are associated with the mucosa and therefore may not be capable of gaining alveolar access. Another opportunity for early, vaccine-induced immune activity is in halting cell-to-cell mycobacterial spread. As this activity entails periods of extracellular transit, vaccine-induced antibodies conceptually could assist in intervening in this spread. As neutrophils appear to play an important role in Mtb spread, the possibility for activating bactericidal mechanisms within these cells via vaccination merits further investigation. Finally, antibody directed cellular cytotoxicity (ADCC) could be important in attempting to kill Mtb within the pulmonary interstitium.

Dr. Ernst also discussed the concept of “training” components of innate immunity, monocytes in particular, to respond more effectively against Mtb infection. Although innate immune responses classically are considered incapable of memory, recent investigations suggest that exposure of monocytes to pathogens, including BCG, generate epigenetic changes in the miRNA transcriptome of monocytes that enhance monocyte function when containing initial infection [36] and [37]. These changes, however, are transient and appear to occur independently of T-cell or B-cell influence, as witnessed by the observation of this phenomenon in plants. Even so, the potential for training this type of innate immune response has been demonstrated in mice [38] and [39]. The possibility of manipulating this innate immune response mechanism via vaccination merits further exploration, particularly when attempting to abrogate the establishment of sustained Mtb infection.

Human genetic studies may provide important information that could contribute to development efforts for vaccines to prevent sustained Mtb infection. One study cited By Dr. Ernst was a genetic investigation of individuals who did not develop TST positivity despite living in an area hyperendemic for TB [25]. In this study, a locus on chromosome 11 was found to be associated with non-conversion of TST in this high-exposure area. An extension of this study to household contacts of TB patients in Paris who remained TST negative found a large locus of 30 million to 40 million bases associated with non-conversion [40]. This large locus contains two smaller loci of interest. One was the tumor necrosis factor-1 (TNF-1) locus, which controls TNF-1 release in whole blood stimulated with BCG. (The locus did not itself include the gene for TNF.) The other small locus contains the WT-1 gene, which encodes a transcription factor reported to contribute to activation of the Vitamin D receptor, and thus might play a role in modifying the response to Mtb infection.

In summary, Dr. Ernst emphasized that the early life cycle of Mtb involves transit through a series of distinct cellular niches. As hilar lymph nodes represent a key passage point in the transit of Mtb-infected, migratory dendritic cells and monocytes, it may be critical to develop vaccines that can exert an immunologic protective effect in the lymph nodes as well as in the lung parenchyma. Monocytes play an important role in the response to Mtb infection and may be capable of being trained to respond early in Mtb infection, although these cells are incapable of immunological memory. To accomplish this, tissue resident memory cells, perhaps generated through aerosol vaccine delivery, may be needed to provide this training. Finally, human genetic studies of persons who apparently remain uninfected with Mtb despite ongoing exposure may point the way toward new vaccine strategies capable of preventing sustained Mtb infection.

3. Session 2: challenges to developing a vaccine with a prevention of sustainedMycobacterium tuberculosis infection indication

Diagnosing NewMycobacterium tuberculosisInfection: Current Challenges. Dr. Marcel Behr, Director, McGill International TB Center, Montreal, Canada.

Accurately diagnosing new infection with Mtb represents one of the major hurdles faced in developing a vaccine for an indication to prevent sustained Mtb infection. Dr. Marcel Behr discussed issues surrounding diagnosis of new Mtb infection with either the TST or the IGRA.

The TST was developed in 1907. Although it is still widely used, Dr. Behr questioned whether this test would be approved by the FDA today. It is a complicated assay that uses a standardized but non-synthetic reagent and displays variability due to both the depth and angle of the needle used to inject the intradermal dose of tuberculin antigen and to human variability in the scoring of the resulting induration. Additionally, the TST cross-reacts with both BCG and other non-tuberculous mycobacteria (NTM), diminishing the specificity of a positive result.

IGRAs, in contrast, are “cleaner”, use recombinant reagents, result in a quantitative laboratory based result, and do not cross react with BCG, although cross reactivity has been seen with some other NTMs. IGRAs eliminate background noise by focusing on the detection of a single cytokine signal generated in response to 2–3 antigens. Dr. Behr suggested, however, that this simplicity of the IGRA could also represent a problem, particularly if the test-retest variability often seen with serial IGRA testing results from biologic factors, such as the possibility that infected individuals change their antigen-specific T-cell responses to Mtb infection frequently, perhaps even on a daily basis. Recognizing the unexplained variability of IGRA assays, the Canadian recommendations for LTBI advise that a TST should be used in all situations where the test is planned to be repeated to detect new infection, for example, in the serial testing of healthcare workers [41].

Dr. Behr raised the concern that IGRA variability in clinical trials for prevention of sustained Mtb infection vaccines could generate sufficient statistical noise to bias the results of an efficacious vaccine toward the null. He did note, however, that such statistical noise occurring in randomized clinical trials theoretically should be randomly assigned between the comparator arms. Ultimately, it will be important to consider whether the statistical noise inherent in the use of IGRAs would affect the power of this and future studies, and actually bias results of an otherwise efficacious vaccine to the null.

Potential Alternative Biomarkers Useful for the Detection of Mtb Infection. Dr. Morten Ruhwald, Statens Serum Institute, Copenhagen, Denmark.

Dr. Morten Ruhwald described progress in the development of the next generation of IGRAs, which promise to reduce assay variability while also making the assays simpler and cheaper to use. In the currently licensed IGRAs, the patient blood sample containing T-cells and antigen presenting cells is combined with Mtb-specific antigens: peptides derived from the ESAT-6, CFP-10 and TB7.7 proteins. The tube is incubated to allow the Mtb antigens to drive the immune response and the T-cell generated IFN-γ is then quantitated using a detection assay. A diagnostic algorithm is applied to determine the cutoff point between a positive and negative result.

Dr. Ruhwald discussed opportunities to improve the IGRA assay by identifying additional Mtb specific antigens and alternative immunodiagnostic markers to IFN-γ. The proinflammatory chemokine IP-10 (IFN-γ inducible protein 10, also known as CXCL10) is secreted by monocytes in response to a number of T-cell derived signals including IFN-γ [42]. In contrast to cytokines like IFN-γ, which are secreted at very low concentrations to direct immunological responses from cells in the immediate area of the cytokine release, chemokines like IP-10 are intended to have an effect over larger distances, and therefore are expressed at much higher levels. IP-10 is expressed at 100 fold higher levels than IFN-γ. This level of expression dramatically reduces analytical variability in assays incorporating IP-10 as a diagnostic indicator. Importantly, this higher expression level also permits the use of simpler diagnostic techniques that may be more accessible in low resource settings, such as utilizing dried blood spots [43], a rapid lateral flow test[44] and RT-qPCR assays [45]. More than 29 diagnostic accuracy studies have demonstrated that IP-10 and IFN-γ responses are at least comparable in specificity and sensitivity (reviewed in [46] and [47]).

The identification of new Mtb antigens to drive IFN-γ and IP-10 release has the potential to improve the sensitivity of IGRAs [48] and may also identify new antigens for vaccine development. Development of IGRAs utilizing different Mtb antigens also is necessary to test new vaccines that include ESAT-6, CFP10 or TB7.7 antigens. Such vaccines cannot use currently available IGRAs to determine Mtb infection because the antigens contained in the vaccines themselves, not Mtb exposure, would prime antigen specific T-cells and create false positive results.

The current effort to assess the H56:IC31 TB vaccine represents an example of this challenge. The H56:IC31 vaccine is a fusion protein of three Mtb antigens, ESAT-6, Ag85b and Rv2660c, combined with the adjuvant IC31. Because of the presence of the ESAT-6 antigen, studies to assess the efficacy of the vaccine to prevent Mtb infection in humans will require development of an IGRA that does not contain this antigen. Dr. Ruhwald described the process of creating an ESAT-6-free IGRA assay, which began with the single antigen screening of 10 potential Mtb antigens not found in BCG in three geographically diverse sites: Egypt, Greenland and Denmark. Three of the most potent antigens for IFN-γ release, QCT 6, 7 and 13, were combined with CFP10 to produce an ESAT-6-free IGRA antigen cocktail that now has been validated in studies in Egypt/Denmark, Tanzania and South Africa. In all cases, this ESAT-6-free IGRA had comparable accuracy to the conventional IGRA assay using either IFN-γ or IP-10 as the immunologic diagnostic readout.

The ability to accurately assess new Mtb infection is critical to potential regulatory approval of Mtb vaccines for a prevention of infection indication. Accordingly, the development of better, more sensitive assays, including improved IGRAs, to detect Mtb infection must be included in any comprehensive plan to develop Mtb vaccines for this indication.

Regulatory Implications of Seeking a Prevention of SustainedMycobacterium tuberculosisInfection Indication. Dr. Roshan Ramanthan, Medical Officer, Division of Vaccines and Related Product Applications, Office of Vaccine Research and Review, Center for Biologics Evaluation and Research, US Food and Drug Administration, Silver Spring, Maryland, USA

Dr. Roshan Ramanathan discussed the laws and regulatory pathways applicable to the clinical development of an Mtb vaccine. The United States FDA licenses vaccines that have been demonstrated to be safe and effective, and that are manufactured in a consistent manner [49]. All indications must be supported by substantial evidence of effectiveness with the expectation that this evidence is derived from adequate and well-controlled clinical studies [50]. Dr. Ramanathan described two pathways applicable to licensure of vaccines: traditional approval and accelerated approval [51]. For traditional approval, evidence of effectiveness is based on protection against clinical disease or, in certain cases, an immunologic response. An immunologic response marker may support traditional approval of a vaccine if it predicts protection from disease and can be reliably measured in a validated assay. Use of an immunologic marker is facilitated by an understanding of the disease pathogenesis and the mechanism by which the vaccine prevents disease.

The accelerated approval pathway permits approval of vaccines to prevent serious or life-threatening diseases or conditions and that provide a meaningful benefit over existing therapies. Approval under this pathway is based on the effect of the vaccine upon a surrogate endpoint that is reasonably likely to predict clinical benefit or on the basis of an effect on a clinical endpoint other than survival or irreversible morbidity. A surrogate endpoint is a marker, such as a laboratory measurement, radiographic image or other measure thought to predict clinical benefit but is not itself a measure of clinical benefit. Adequate data supporting the use of the chosen surrogate marker are required. Licensure of a vaccine via this pathway requires that adequate and well-controlled studies confirm clinical benefit. Confirmatory studies must be conducted with due diligence and are often underway at the time the applicant submits the marketing application to the FDA.

Regardless of the pathway pursued, a critical issue with respect to the development of a vaccine to protect against sustained and/or latent Mtb infection is that only 10% of those with sustained or latent Mtb infection will develop active TB disease. The applicant would need to provide a compelling argument that this indication is clinically meaningful, and that the vaccine is likely to prevent disease in the targeted population.

Dr. Ramanathan emphasized that regulatory pathways to licensure for TB vaccines are available. Given the complexities of regulatory issues concerning the development of a vaccine for prevention of sustained Mtb infection, Dr. Ramanathan encouraged developers to engage with the FDA early in vaccine development so that the Agency may provide feedback on the clinical development plans for specific products. Engagement of regulatory authorities, manufacturers and the scientific community will be necessary in order to develop a safe and effective TB vaccine.

Design Challenges for Prevention of SustainedMycobacterium tuberculosisInfection Trials. Dr. Ruth Ellis, Director of Clinical Development, Aeras, Rockville, MD, USA

Dr. Ruth Ellis described the rationale for and design of the first trial of a TB vaccine with a primary prevention of Mtb infection (PoI) endpoint, a phase 2 proof-of-concept clinical trial now underway in a South African adolescent population at high risk for Mtb infection (ClinicalTrials.gov Identifier: NCT 02075203). In her presentation, Dr. Ellis also highlighted the challenges in designing studies attempting to assess vaccine efficacy in preventing de novo Mtb infection.

A vaccine that prevents Mtb infection may be able to interrupt the cycle of disease and transmission, potentially leading to a large reduction in TB incidence at the population level. Additionally, a clinical trial with an endpoint of Mtb infection, in contrast to an endpoint of TB disease, will inherently require fewer subjects and a shorter time frame to produce statistically significant results. Accordingly, a phase 2 PoI trial represents an important strategic step in the development pathway of a TB vaccine candidate, as it offers the potential of identifying a biological signal of potential efficacy, one that can provide the confidence needed by vaccine developers to commit the extensive resources required to conduct late phase 2 and phase 3 licensure studies. Obtaining a biological signal of potential efficacy is particularly important in the effort to develop a TB vaccine given the absence of an immune correlate of protection to guide these efforts.

A major challenge in designing and conducting a PoI study is to select an assay with sufficient reliability to accurately identify new Mtb infections. In the trial described by Dr. Ellis, the QuantiFERON^®-TB Gold In-Tube (QFT) IGRA was selected to identify new Mtb infections among participants. A challenge in using the QFT IGRA is the numerous sources of variability in this assay [52]. Additionally, a valid conversion cutoff point continues to be unclear. Also unclear is whether reversion from a positive to a negative result reflects a true biologic event or simply assay variability. The TST could not be used to identify new Mtb infection in this trial because of its lack of specificity in BCG vaccinated populations.

Another major challenge in designing and conducting a PoI trial is to identify a population with a high enough incidence of Mtb infection to make the trial feasible to conduct in a stringent budgetary environment. The PoI trial described by Dr. Ellis is being conducted in a population of 990 healthy, HIV uninfected, remotely BCG vaccinated and QFT negative adolescents aged 12–17 in the Western Cape area of South Africa, where annual rates of conversion to a positive QFT test are 14% [53], and QFT positivity and conversion is associated with an eight-fold higher risk of progression to TB disease within 2 years when compared to nonconverters [54]. It is this extraordinarily high incidence of Mtb infection, in combination with relaxed confidence intervals around a targeted vaccine efficacy of 50%, appropriate for a phase 2 proof-of-concept study, that permits a sample size of fewer than 1000 to be enrolled (330 per group).

The study is partially blinded and subjects are randomized 1:1:1 to receive vaccination with H4:IC31 (Aeras 404: a fusion peptide vaccine developed by the Statens Serum Institute, Sanofi Pasteur and Aeras consisting of Mtb antigens 85B and TB 10.4 and adjuvanted with the proprietary adjuvant IC31, developed by Valneva), BCG revaccination, or placebo on Day 0. QFT assays performed at 3 months (Day 84) allow a washout point for QFT positive subjects who were infected around the time of the initial screening but were not detected because measurable cellular responses of Mtb infection were not yet evident. QFT assays are repeated at 6, 12, 18, and 24 months. Individuals who become QFT positive are followed up at 3 and 6 months after the conversion event to determine if reversion to a QFT-negative state occurs. The primary endpoint of the trial is to evaluate prevention of Mtb infection by rates of initial QFT conversion. The secondary endpoint is to measure rates of sustained QFT conversion. In both cases, the effect of Aeras 404 and BCG revaccination will be compared to placebo. Additional exploratory objectives of the trial include the effect of applying alternative conversion cut-off points to rates of QFT conversion and reversion.

Aeras plans to implement additional PoI trials in an effort to efficiently obtain biological signals of potential efficacy for other TB vaccine candidates that will increase the likelihood of vaccine success in later stage trials. If vaccine licensure is sought for a prevention of Mtb infection indication, it will be important to improve the sensitivity of the IGRA assay, or develop a new, more sensitive and less variable assay to reliably detect new Mtb infection, to provide regulatory authorities the degree of confidence required to merit licensure.

Selecting Vaccine Candidate Attributes Optimal for Preventing Sustained Mtb Infection. Dr. Barry Walker, Vice President, Preclinical Development, Aeras, Rockville, Maryland, USA.

Dr. Barry Walker noted that either clearance of Mtb or blocking further replication of the organism may be functionally equivalent to preventing sustained infection. A vaccine being developed for a prevention of sustained Mtb infection indication has the opportunity to interdict establishment of infection at a number of points in the earliest stages of infection, from the initial encounter with the organism at the mucosa, through the ultimate formation of granulomas. This implies that a polyfunctional vaccine strategy, targeting more than one point of potential susceptibility of Mtb to immunological activity, may have the greatest opportunity to achieve this goal.

Dr. Walker listed five potential points leading to the establishment of sustained Mtb infection where a vaccine could have an effect: (1) at or around the initial invasion through the lung mucosa; (2) during phagocytosis into alveolar macrophages; (3) during replication within alveolar macrophages; (4) at the establishment of the primary Ghon focus and (5) during the period of regional dissemination and establishment of secondary foci, mainly within pulmonary granulomas. For a vaccine regimen to successfully interdict at the pulmonary mucosa or at the time of phagocytosis, some combination of vaccine-mediated physiological modulation of the mucosa, particularly the development of mucosal antibodies, in combination with systemic antibody production, an ADCC response, innate immunity stimulation, and Mtb-specific cell mediated immunity (CMI) development would seem optimal. The potential for priming tissue effector memory which could provide a rapid, direct, parenchymal-based response to Mtb infection and also potentially signal local innate responses to function at maximum efficiency is an area of particular research interest. The possibility of priming intra-luminal macrophages in addition to interstitial macrophages to respond to Mtb infection also should be explored.

A vaccine regimen targeting the downstream steps leading to establishment of sustained Mtb infection, including the development of the primary Ghon focus and the process of stimulating granuloma formation, would most likely rely on a robust CMI response either alone or in combination with innate immunity stimulation, prompted either by the vaccine itself or by novel adjuvants administered along with the vaccine. Targeted pathophysiological mechanisms may include macrophage functions, T-cell homing, accentuated macrophage killing of phagocytosed bacteria, or direct killing of infected macrophages along with the internalized mycobacteria.

When developing a vaccine aimed at prevention of sustained mycobacterial infection, Dr. Walker cautioned that the target produce profile (TPP) should be carefully spelled out prior to proceeding with product development. Components of the TPP that should be clear as development proceeds include the target population (e.g., adolescents/adults, infants, HIV-infected or–uninfected); the specific endpoint being assessed as this will likely become the indication for which marketing approval will be sought; the mechanism of vaccine action and the nature of vaccine administration (e.g., injection, aerosol).

Ultimately, a combination of vaccines using different platforms expressing the same set of genes, or different vaccines altogether (e.g., a vectored subunit vaccine in combination with a whole mycobacteria cell vaccine), perhaps administered in different ways, e.g., one via aerosol and one via injection, may be required to prevent sustained Mtb infection. Clearly a robust means to assess the potential efficacy of these various vaccine strategies in the preclinical space will be critical in identifying the most promising vaccines and vaccination regimens to take to the clinic.

4. Session 3: panel discussion: the path forward to a vaccine indicated for the prevention of sustained Mycobacterium tuberculosis infection

Panel Chair: Dr. Dan Hoft, Department of Molecular Microbiology and Immunology, St. Louis University.

Panelists

Dr. Marcel Behr

Dr. Mark Hatherill, Director, South African Tuberculosis Vaccine Initiative, University of Cape Town, South Africa

Dr. Helen McShane, Professor of Vaccinology, Nuffield Department of Medicine, Oxford University, UK

Dr. Larry Schlesinger, Chair, Department of Microbial Infection and Immunity, Wexner Medical Center, Ohio State University, Columbus, Ohio, USA8

Dr. Christine Sizemore, Chief, TB, Leprosy and Other Mycobacterial Diseases Section, Division of Microbiology and Infectious Diseases, NIAID, NIH, Bethesda Maryland, USA

Dr. Tom Evans, CEO, Aeras, Rockville, Maryland, USA.

Addressing the challenges inherent in interpreting IGRA results, particularly in cases of IGRA reversion, was highlighted as a major need by the panel. Currently, it cannot be determined with confidence whether a person who demonstrates a transiently positive IGRA result was infected with Mtb and then cleared it, or whether such a result represents vagaries in the assay. Clarifying this situation would represent an important contribution to efforts attempting to assess TB vaccines for their ability to prevent sustained Mtb infection.

The panel summarized the perspectives derived from the conference speakers concerning the type of immunological activity that a vaccine likely would need to stimulate to prevent sustained Mtb infection. A successful vaccine likely would need to induce regional tissue immunity, particularly in the lung; lead to a rapid response to repeated low-level challenge from inhaled environmental Mtb organisms; induce both antibodies and conventional T-cells; and induce unconventional T-cell responses.

The panel noted that the information presented during this conference suggested that developing vaccines for a prevention of sustained Mtb infection indication would have both a public health and economic value, and potentially would be feasible from a regulatory standpoint. Additionally, developing vaccines for this indication appeared to be biologically plausible. Based on the information presented in this conference, it also seemed clear that the TB vaccine development field was asking the right questions about potential pathways for vaccine development. Given the importance of pulmonary-based immune responses in attempting to prevent sustained Mtb infection, it was recommended that every TB vaccine development pathway include studies in which the pulmonary immune response to the vaccine under investigation be assessed with specimens obtained via bronchioalveolar lavage. Secondary factors, such as diabetes or underlying lung pathology caused by environmental lung disease and occupational exposure to smoke, also must be considered when assessing vaccine efficacy to prevent sustained Mtb infection. Additionally, it was noted that twenty percent of TB disease is extrapulmonary, resulting in a need to develop better assays for detecting extrapulmonary TB. The need to develop more sensitive means of identifying Mtb infection also was stressed, to assess the efficacy of vaccines to prevent both sustained, de novo Mtb infection and Mtb reinfection among those who have been cured of initial TB disease.

The panel emphasized the need to utilize advanced immunological assessment technologies, such as transcriptomics, in both the lung and periphery of non-human primates and humans, to more accurately bridge assessments of vaccine activity. This approach would serve to help de-risk the resource commitment required for later-stage development of TB vaccine candidates.

While there was agreement that developing vaccines to prevent active TB disease would remain the major focus of the TB vaccine development field, the panel concluded that there did not appear to be biological, public health, economic, or absolute regulatory constraints against developing vaccines to prevent sustained Mtb infection, particularly if improvements were made in our ability to diagnose Mtb infection in a more sensitive and reliable fashion.

Acknowledgements

The organizers of the conference express their gratitude to Dr. Stephanie Cascio for her major contribution to the drafting of this manuscript and to Daniel Yeboah-Kordieh for his assistance in preparing this manuscript. Sincere thanks also to Dr. Joseph Chiu and the NIAID for organizational support of this conference and to the Bill and Melinda Gates Foundation, without whose support this conference would not have been possible.

References

[1]
E.A. Nardell, R.S. Wallis
Here today – gone tomorrow: the case for transient acute tuberculosis infection
Am J Respir Crit Care Med, 174 (2006), pp. 734–735
View Record in Scopus

Site is Back

Are three drugs for malaria better than two?

Public health Interventions and epidemic intensity during the 1918 influenza pandemic

Chloroquine and hydroxychloroquine as available weapons to fight COVID-19

Using models to shape measles control and elimination strategies in low- and middle-income countries ...

DEVELOPING VACCINES TO PREVENT SUSTAINED INFECTION WITH MYCOBACTERIUM TUBERCULOSIS: CONFERENCE PROCEEDINGS: NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASES, ROCKVILLE, MARYLAND USA, NOVEMBER 7, 2014

This article is one of four on the subject appearing in the June 2015 issue of Vaccine, which those interested in TB vaccine development may wish to consult.

DEVELOPING VACCINES TO PREVENT SUSTAINED INFECTION WITH MYCOBACTERIUM TUBERCULOSIS: CONFERENCE PROCEEDINGS: NATIONAL INSTITUTE OF ALLERGY AND INFECTIOUS DISEASES, ROCKVILLE, MARYLAND USA, NOVEMBER 7, 2014

Abstract

Abbreviations

Keywords

1. Introduction

2. Session 1: prevention of sustained infection with Mycobacterium tuberculosis: definitions, impact and feasibility

3. Session 2: challenges to developing a vaccine with a prevention of sustainedMycobacterium tuberculosis infection indication

4. Session 3: panel discussion: the path forward to a vaccine indicated for the prevention of sustained Mycobacterium tuberculosis infection

Acknowledgements

References

41253558