Despite decades of developments in immunization and drug therapy, tuberculosis remains among the leading causes of human mortality, and no country has successfully eradicated the disease. Reenvisioning tuberculosis from the perspective of population biology, this book examines why the disease is so persistent and what must be done to fight it. Treating tuberculosis and its human hosts as dynamic, interacting populations, Christopher Dye seeks new answers to key questions by drawing on demography, ecology, epidemiology, evolution, and population genetics. Dye uses simple mathematical models to investigate how cases and deaths could be reduced, and how interventions could lead to TB elimination.
Dye's analysis reveals a striking gap between the actual and potential impact of current interventions, especially drug treatment, and he suggests placing more emphasis on early case detection and the treatment of active or incipient tuberculosis. He argues that the response to disappointingly slow rates of disease decline is not to abandon long-established principles of chemotherapy, but to implement them with greater vigor. Summarizing epidemiological insights from population biology, Dye stresses the need to take a more inclusive view of the factors that affect disease, including characteristics of the pathogen, individuals and populations, health care systems, and physical and social environments.
In broadening the horizons of TB research, The Population Biology of Tuberculosis demonstrates what must be done to prevent, control, and defeat this global threat in the twenty-first century.
About the Author
Christopher Dye is the Director of Strategy in the Office of the Director General at the World Health Organization, Geneva. He has been the Gresham Professor of Physic in London and a visiting professor of zoology at the University of Oxford. His work has appeared in many publications including Science, Nature and The Lancet.
Read an Excerpt
The Population Biology of Tuberculosis
By Christopher Dye
PRINCETON UNIVERSITY PRESSCopyright © 2015 Princeton University Press
All rights reserved.
There may be 12 to 25 million infectious cases of tuberculosis in the world. Confronted with such an astronomical computation of distress, it is clear that tuberculosis is very far from being defeated.
—John Crofton (1960)
In 2012, an estimated 8.6 million people developed TB and 1.3 million died from the disease. There were 12 million prevalent cases.
—World Health Organization (2013a)
"Tuberculosis Undefeated" was the title chosen by Crofton for his 1960 Marc Daniels lecture, given in memory of a colleague who made a vital contribution to the development of combination drug therapy against tuberculosis (Crofton 1960). Much has happened in the population biology of TB since 1960, but Crofton's title still reflects the state of TB in the world today, as reflected in WHO statistics (World Health Organization 2013a). Despite the development of highly efficacious drug therapy to prevent and cure TB, 8–10 million people developed some form of the disease in 2012, there were 11–14 million extant (prevalent) cases, and 1.2–1.5 million people died. The number of TB episodes each year has probably now fallen below an all-time high, reached in the decade 2000–2009, but TB remains the largest cause of death from a single, curable infectious agent.
The challenge we face in the twenty-first century is to control—and ultimately eliminate—a pathogen that has inhabited human populations (including the earliest hominids) for tens of thousands and possibly millions of years. To begin to understand how elimination could be achieved, the purpose of this opening chapter is to define the problem. The following sections describe the essential characteristics of the pathogen and the collection of illnesses it causes, its origins and distribution in human populations, the dominant trends through time, and the burden of disease today. This is an explicitly descriptive account, which prefigures the more analytical investigation of TB epidemiology and control in subsequent chapters.
Everything about the biology of Mycobacterium tuberculosis happens slowly. Whereas other bacteria divide in minutes, M. tuberculosis takes hours. Most bacteria can be cultured in hours, but M. tuberculosis takes weeks. Most outbreaks of bacterial infections run over days or weeks, but an epidemic of M. tuberculosis takes years, decades or centuries.
M. tuberculosis is a small (measured in µm), rod-shaped, aerobic, mostly nonmotile, non-spore-forming bacterium. Following convention in bacteriology, mycobacteria are classified according to their reactions to standard stains used in microscopy. M. tuberculosis is bracketed with gram-positive bacteria, although its waxy cell wall, rich in fatty acids, is Gram neutral (Hinson et al. 1981). Nevertheless, once the poorly absorptive cell wall has been successfully impregnated with concentrated dyes (fuchsin), it withstands decolorization by acids—hence the use of an acid-fast stain in procedures such as Ziehl-Neelsen, which renders all mycobacteria bright red on a blue background when viewed under a light microscope. Mycobacterial species are distinguished by their appearance—M. tuberculosis is comparatively "rough" among its congeners—and by their response to biochemical tests (Ait-Khlaed and Enarson 2003).
The slow growth of mycobacteria in culture is somewhat less slow in liquid (days) than on solid media (weeks on the traditional Löwenstein-Jensen slopes). In diagnostic laboratories that rely on culture, the slow growth can cause significant diagnostic delays, with clinical and epidemiological consequences, including the loss of patients and continuing transmission before the start of treatment. In partial compensation, the variable growth rates of mycobacteria species are an asset in differential diagnosis.
Mycobacteria are widespread in the environment and commonly found in soil and water. Not all are obligate parasites. The M. tuberculosis complex (MTBC) of species contains the major mammalian pathogens, including M. tuberculosis sensu stricto and M. africanum. It also includes the mainly animal pathogens M. bovis (cattle), M. caprae (goats), M. microti (voles), and M. pinnipedii (seals, sea lions; Smith et al. 2006). As a cause of human TB, M. bovis is more commonly associated with extrapulmonary than pulmonary disease (Durr et al. 2013). With improved management of infection in cattle, plus the pasteurization of milk, there are nowadays few human cases due to M. bovis (median < 3% worldwide). However, in localities where these basic control methods are not used, the percentage of human cases due to M. bovis can be much higher (up to 30%–40% in surveys; Muller, Durr et al. 2013). Two other important human mycobacteria in the same genus are the agents of leprosy (M. leprae) and Buruli ulcer (M. ulcerans). The other 100-plus known species—more surely await discovery—are called environmental mycobacteria or, less satisfactorily, nontuberculous mycobacteria (NTM, which can, in fact, cause lung tubercles), atypical mycobacteria, or mycobacteria other than tuberculosis (MOTT bacilli). Despite uneven sampling effort, it is clear that NTM are distributed worldwide and that species differ among continents and among countries within each continent (Hoefsloot et al. 2013).
M. tuberculosis and M. africanum are the principal causes of human lung disease (pulmonary TB), in addition to diseases of other organs and tissues (extrapulmonary TB), notably lymph glands, bone, brain and skin, but all organs are vulnerable. Disseminated disease, caused by the spread of bacteria away from the lung to other organs, is most likely in individuals with compromised or immature immune systems. Among these are people with HIV infection, diabetes, malignancies, infants, and people on therapies, including tumor necrosis factor inhibitors and corticosteroids.
Compared with other bacteria, M. tuberculosis is renowned for its lack of genetic variability, though the view that M. tuberculosis is essentially a clone has been overturned by ever more refined explorations of the genome (Borrell and Gagneux 2011; Gagneux 2013; Coll et al. 2014; Anderson and May 1991). Furthermore, M. tuberculosis has a remarkable capacity to generate drug-resistant mutations, which will appear in almost any patient that is treated with suboptimal doses of drugs (Chapter 5). Even with the limited genetic variability known so far, a suite of markers is proving to be immensely informative about M. tuberculosis biology. The toolbox for investigating genetic variation includes restriction fragment length polymorphisms (RFLP) of the M. tuberculosis insertion sequence IS6110, spoligotyping (based on unique, nonrepetitive DNA spacer sequences), variable number tandem repeats (VNTR), single nucleotide polymorphisms (SNPs), multilocus sequences (MLS) and, increasingly, whole genome sequencing (Tortoli 2011; Ford et al. 2012; Walker et al. 2013; Bryant et al. 2013; Merker et al. 2013; Kohl et al. 2014; Anderson 2014; Niemann and Supply 2014). Genetic markers are being used to study, among other things, who acquires infection from whom, whether recurrent episodes of TB are caused by reinfection or relapse, the population dynamics of benign, virulent, and drug-resistant strains, and the history and geography of M. tuberculosis dispersal.
HUMANS AS HOSTS
M. tuberculosis is a parasite of the human immune system. It attacks the host at the heart of its defenses, inducing transmission by leading the host to its own self-destruction.
The bacilli survive and replicate in phagocytic cells, including macrophages, neutrophils, monoctyes, and dendritic cells. Macrophages are the very cells that normally protect the host by engulfing foreign bodies, including microbes. A proportion of invading M. tuberculosis are killed immediately by an innate immune response. The survivors must establish an infection that will eventually allow transmission to other hosts by stimulating cells of the immune system to produce an adaptive response that is neither too weak nor too strong. Too little immunity allows the multiplication and dissemination of infection away from the lungs, causing widespread tissue and organ damage, eventually killing the host and limiting the chances for transmission. Too much immunity, innate or acquired, prevents bacilli from replicating or kills them prematurely before transmission is possible (van Crevel et al. 2002; Verrall et al. 2014; Ernst 2012; Cooper and Torrado 2012).
Transmission takes place when individuals with disease of the lung produce airborne droplets while coughing, sneezing, spitting, or just talking. The droplets typically stay airborne for a few hours, during which time they can be inhaled by another person.
The course of M. tuberculosis infection begins when inhaled droplets carrying M. tuberculosis lodge in the lung tissue, where they are captured by scavenging macrophages. Bacilli replicate slowly within each macrophage, bypassing the host cell's usual destructive mechanisms. Replication leads eventually to breakout and the colonization of more macrophages as they are recruited to the site of infection. In this initial phase, infected macrophages spread through the lymphatic system to the draining lymph nodes, but infection is otherwise contained in the lung.
With the stimulation of macrophages and other immune cells, the human host builds a cell-mediated immune response that stops bacterial replication after a period of several weeks. At this point, the presence of infection and the ensuing cell-mediated immune response are detectable by antigen stimulation tests. The Pirquet, Mantoux, or tuberculin skin test evokes an immune response in the skin via intradermal inoculation of M. tuberculosis purified protein derivative, causing erythema, edema, and induration on the skin surface. Alternatively, macrophages and T cells stimulated with antigens specific to M. tuberculosis will signal the presence of infection via the production of interferon gamma (the basis of interferon gamma release assays, IGRAs), among other cytokines (Grosset 2003; Pai et al. 2014).
Populations of activated macrophages and lymphocytes that congregate at the site of infection form granulomas, which appear as nodules or tubercles (Ehlers and Schaible 2012). Within the granulomas, the continuing effort to control the multiplication of bacilli causes the human host to destroy its own lung tissue. Cell death within the granulomas creates Emmentaler-like2 lung cavities (caseous necrosis), typically in the middle and lower lung, a hallmark of pulmonary TB.
If the granuloma breaks down, the bacilli can continue to multiply outside as well as inside macrophages (Grosset 2003; Shakak et al. 2013). If the infection is able to proceed, replication leads to the creation of more cavities, and the lung is progressively destroyed in a process that may take months or years. The lung damage causes coughing, with the expulsion of bacilli to the outside air and the expectoration of sputum, with blood in the advanced stages of disease. For TB, the extent and severity of pulmonary disease is associated with infectiousness, and this alignment is used in mathematical modeling of M. tuberculosis transmission (Chapter 2).
The more effective the immune response, the slower the rate at which individuals progress to active disease, having acquired infection. Varying degrees of suppression by immune responses generate a spectrum of latent or subclinical states, ranging from the early containment of infection by an innate or an acquired immune response, through the containment of live but nonreplicating bacilli and bacterial replication at low levels partially contained by immunity (Barry et al. 2009; Young et al. 2009; Colangeli et al. 2014). Tuberculin skin testing to detect cell-mediated immune responses suggests that the human population carries a vast reservoir of live bacilli. The conventional wisdom is that, in long-term infections, the bacilli are in a quiescent or latent state, which can reactivate in any individual after years or decades of dormancy. However, as latency comes under closer scrutiny, the idea that this is a single state (Barry et al. 2009) or even that latency exists at all (Shakak et al. 2013) is under challenge (see also Chapter 2). A more contemporary view is that latent tuberculosis—infection in the absence of clinical symptoms—is a collection of conditions that form part of a heterogeneous response to M. tuberculosis invasion. Infection leads to the formation of distinct granulomatous lesions, with differential ability to support or suppress bacterial replication (Barry et al. 2009; Lin et al. 2013). Whatever the true nature of persistent infection, some people will develop TB months or years after first exposure. But the great majority of healthy people who have signs of exposure to infection and who presumably carry viable bacilli do not live long enough to develop TB disease.
M. TUBERCULOSIS AFFECTING THE DISTRIBUTION OF OTHER DISEASES
M. tuberculosis is one of a set of pathogens interacting competitively or facultatively via the human immune system. The adaptive immune response described previously generates two antagonistic subclasses of T helper cells—Th1 and Th2—each with its own set of cytokine mediators. Microbial infections have the potential to influence the balance between Th1 and Th2 responses by altering cytokine profiles, with positive or negative consequences for the progression of disease. Bacterial infections probably have such a role in atopy, an allergic state producing mucosal inflammation characteristic of asthma and characterized by overreactive Th2 responses.
In simple terms, the Th1 response to M. tuberculosis infection is associated with granuloma formation and protection, whereas the Th2 response results in tissue-killing hypersensitivity and the progression of disease. The processes that determine the balance of the two responses also affect and are affected by the interaction between M. tuberculosis and other infectious agents, from viruses to helminths.
Because mycobacteria elicit strong Th1 responses, thereby shifting the Th1/ Th2 balance away from Th2, M. tuberculosis infection could protect against asthma. One study of Japanese children found that strong tuberculin responses, probably attributable to M. tuberculosis exposure, were associated with less asthma, rhinoconjunctivitis, and eczema in later childhood (Shirakawa et al. 1997). In positive tuberculin responders the rate of current atopic symptoms was one-third the rate in negatives, and asthmatic symptoms were one-half to one-third as likely. On top of this, remission of atopy in children aged 7–12 years was six to nine times as likely in positive tuberculin responders. A study of South African children found an inverse association between M. tuberculosis infection and atopic rhinitis (Obihara et al. 2005). Other comparisons among countries have found that asthma tends to be more common where TB is not (von Mutius et al. 2000; Shirtcliffe 2002). The debatable implication of these results is that TB has been inhibiting the spread of asthma and other atopic disorders worldwide. Taking this one step further, a mycobacterial vaccine might be constructed to prevent atopy and asthma. BCG could already serve that purpose, though the evidence is ambiguous (Hopkin 2000). M. tuberculosis infection may also protect against leprosy, as does BCG (Karonga Prevention Trial Group 1996) and natural TB transmission could have contributed to the decline of leprosy in Europe (Lietman et al. 1997).
As the evidence for an immunological link between TB and asthma becomes more compelling, interactions between other infections have come under investigation. For instance, vigorous Th2 responses are seen in protective immune reactions to helminth infections, and helminths could modulate atopic disease while compromising the immune response to BCG and M. tuberculosis (Hopkin 2000; Obihara et al. 2006; Ferreira et al. 2002; Franke et al. 2013).
The complexity of interactions between M. tuberculosis and individual immunity presents a set of fascinating and important problems in within-host population biology. Just one of the many relevant questions is how mycobacteria can survive for years within hosts in small but viable populations, neither being eliminated by the host nor being allowed to break out of granulomas to cause destructive and progressive disease. One theory proposes that an equilibrium can be maintained by a combination of intra-and extracellular populations of bacilli (Blaser and Kirschner 2007). As the ecology of bacilli, macrophages, T cells, and other cell species are further explored, a key question is whether the systems biology of within-host dynamics can offer insights into the between-host dynamics of TB in populations (Chapter 8; Young et al. 2008).
The synergistic and antagonistic interactions between bacterial, viral, and parasitic infections, mediated by immunity, are complex and unresolved. Nonetheless, the preceding examples at least raise the possibility that mycobacteria influence and are influenced by other infections to a far greater extent than hitherto appreciated, with potentially important effects at population level. Conceivably but speculatively, the manipulation of Th1 and Th2 responses has potential for TB immunotherapy.
Excerpted from The Population Biology of Tuberculosis by Christopher Dye. Copyright © 2015 Princeton University Press. Excerpted by permission of PRINCETON UNIVERSITY PRESS.
All rights reserved. No part of this excerpt may be reproduced or reprinted without permission in writing from the publisher.
Excerpts are provided by Dial-A-Book Inc. solely for the personal use of visitors to this web site.
Table of Contents
Chapter 1 Tuberculosis Undefeated 1
Chapter 2 Concepts and Models 26
Chapter 3 Risk and Variation 64
Chapter 4 Interventions and Control 100
Chapter 5 Strains and Drug Resistance 138
Chapter 6 TB and HIV/AIDS 162
Chapter 7 Elimination and Eradication 190
Chapter 8 Populations and Social Diseases 207
Appendix 1 Derivation of the Basic Case Reproduction Number and Epidemic Doubling Time 219
Appendix 2 Formal Description of the Standard Model 222
What People are Saying About This
"At once informative and captivating, this book is without question the most comprehensive text on the ecology and evolution of tuberculosis, and represents a landmark contribution to the field by one of its most authoritative figures. Painting the most up-to-date picture of tuberculosis from diverse perspectives, Dye lays bare the key intellectual concepts and, in a wonderfully elegant and compelling manner, examines their conclusions. A pleasure to read."Pejman Rohani, University of Michigan
"This book is a brilliant rethink of tuberculosis within the context of a rapidly evolving global environment. Dye examines the multifaceted factors of the disease and lays the foundation for a novel approach to tackling it. He challenges the global health community to address TB as a social disease as well as a medical one, and makes the case for a comprehensive and holistic response."Ariel Pablos-Méndez, U.S. Agency for International Development