the forgotten time-table of tuberculosis · confidential: for review only the forgotten time-table...

Confidential: For Review Only

The Forgotten Time-Table of Tuberculosis

Journal: BMJ

Manuscript ID BMJ.2018.043229

Article Type: Analysis

BMJ Journal: BMJ

Date Submitted by the Author: 12-Jan-2018

Complete List of Authors: Behr, Marcel; McGill University Edelstein, Paul; University of Pennsylvania Perelman School of Medicine Ramakrishnan, Lalita; University of Cambridge

Keywords:

https://mc.manuscriptcentral.com/bmj

BMJ

Confidential: For Review OnlyThe Forgotten Time-Table of Tuberculosis

Marcel A. Behr1, Paul H. Edelstein

2,3, Lalita Ramakrishnan

3

1Department of Medicine, McGill University, McGill International TB Centre, Montreal,

H4A 3J1, Canada

2Department of Pathology and Laboratory Medicine, University of Pennsylvania,

Philadelphia, PA, 19104, USA

3Molecular Immunity Unit, Department of Medicine, University of Cambridge, MRC

Laboratory of Molecular Biology, Cambridge CB2 0QH, UK

Correspondence: [email protected] (MB); [email protected] (PE);

[email protected] (LR)

Page 1 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyRequired Information

Copyright The Corresponding Author has the right to grant on behalf of all authors and does

grant on behalf of all authors, an exclusive licence (or non exclusive for government

employees) on a worldwide basis to the BMJ Publishing Group Ltd to permit this article (if

accepted) to be published in BMJ editions and any other BMJPGL products and sublicences

such use and exploit all subsidiary rights, as set out in their licence.

Competing interest statement All authors have completed the Unified Competing Interest

form and declare: no support from any organisation for the submitted work; no financial

relationships with any organisations that might have an interest in the submitted work in the

previous three years, no other relationships or activities that could appear to have influenced

the submitted work.

Contributors and sources The contributors are clinician-scientists with a longstanding

interest in the pathogenesis and epidemiology of tuberculosis. MAB is an expert in

bacterial genomics and molecular epidemiology of TB. PHE is an expert in clinical

microbiology and infectious diseases. LR is an expert in the immunopathogenesis of

mycobacterial infections. All three authors contributed to the literature search, the writing

and the editing of the paper; PHE prepared the figures. The sources of information used

include a contemporary literature search using PubMed and Google Scholar, a search of

older articles from the pre-antibiotic era and a request to many colleagues in the field for

counterfactual reports that challenge the content and/or message of this manuscript.

LR is the nominated guarantor of the article.

Page 2 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyLicence The Corresponding Author has the right to grant on behalf of all authors and

does grant on behalf of all authors, an exclusive licence (or non exclusive for government

employees) on a worldwide basis to the BMJ Publishing Group Ltd ("BMJ"), and its

Licensees to permit this article (if accepted) to be published in The BMJ's editions and

any other BMJ products and to exploit all subsidiary rights, as set out in their licence.

Transparency Declaration The lead author (the manuscript’s guarantor) affirms that the

manuscript is an honest, accurate, and transparent account of the study being reported; that

no important aspects of the study have been omitted; and that any discrepancies from the

study as planned have been explained.

Details of ethical approval Ethical approval not required for this analysis article

Details of funding This work was done in the authors’ own time outside of their funded

projects

Details of the role of the study sponsors There were no study sponsors

Statement of independence of researchers from funders. The funders of the authors’

other research activities had no influence on this article

Page 3 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyArticle

Tuberculosis has a much shorter incubation period than is widely believed and

this has implications for prioritizing research and public health strategies.

One-third of the world’s population is believed to be latently infected with

Mycobacterium tuberculosis

(http://www.who.int/tb/publications/ltbi_document_page/en/). The WHO defines latent

tuberculosis infection (LTBI) as “a state of persistent immune response to stimulation

by Mycobacterium tuberculosis antigens without evidence of clinically manifested active

tuberculosis (TB)”. The WHO guidelines further state: “Individuals with LTBI do not

have active TB disease but may develop it in the near or remote future, a process called

‘TB reactivation’ (Box 1). The lifetime risk of reactivation for a person with documented

LTBI is estimated to be 5–10%, with the majority developing TB disease within the first

five years after initial infection.” Similarly, other authoritative guidelines use terms such

as “dormant” or “alive but inactive” to describe the bacteria in LTBI (Box 1).

The thinking that one-third of the world’s people are walking time bombs for TB

reactivation from their LTBI has engendered a great deal of concern because this

“reactivated” TB is contagious, and therefore a major contributor to the global disease

burden. Therefore, eradicating LTBI is considered a cornerstone of global TB control

(1), and consequently achieving a better understanding of it is deemed a high research

priority (2, 3)(Box 2). The word latent has a biological definition and a medical one. The

biologic concept of latency is that of a resting stage, hidden until circumstances are

suitable for development. The medical definition of latent is simply a pathologic process

Page 4 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlyin which symptoms are not yet manifest. Since TB research on latency mentions

conditions favorable to long-term persistence and the TB clinical community has long

used the apposition of LTBI and reactivation (Box 1), it is clear that the biologic

definition is being applied.

The importance accorded to latency is reflected in a major push from researcher

funding agencies to understand LTBI biology, including the state of the bacteria therein

so as to develop specific antimicrobial drugs for LTBI (Box 2). In 2001, the National

Institutes of Health (NIH) issued a specific call to address the problem of LTBI as part of

the response to the Presidential Vaccine Initiative to address LTBI. In 2005, when the

Bill and Melinda Gates Foundation (BMGF) issued their Grand Challenges, a family of

initiatives fostering innovation to solve key global health and development problems, one

of the 43 awarded was entitled “Drugs to treat latent tuberculosis”. Then in 2008 in their

Grand Challenges Explorations call, the BMGF issued a specific call for applications

addressing LTBI and funded 15 of them. The wording of these calls and the summaries of

the funded proposals reflects the central importance accorded to understanding the

biology and epidemiology of LTBI in achieving global TB eradication (Box 2).

In this paper, we review multiple longitudinal epidemiological studies that show

that the vast majority of TB disease manifests soon after infection, with disease following

remote infection being rare. Therefore, the vast burden of global TB disease is from

recently transmitted infection. Only in countries with a low TB burden, where ongoing

transmission is minimal, is TB disease from remotely acquired infection a major

contributor to the active TB burden (4). Importantly, recent studies in low burden areas

find that most such active TB cases resulting from remote infection do not result in major

Page 5 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlydisease outbreaks (5, 6). This appreciation of the natural history of infection and disease

should be useful in strategizing to achieve global eradication of TB. Additionally, this

information can inform the duration of TB vaccine trials; in contrast to 10-20 year BCG

studies conducted in the 20th

century, just two years of follow-up should suffice to detect

most new cases in a high-transmission environment. Finally, the natural history of TB

does not support the many terms currently used to describe the various phases of TB

(Box 1). Indeed, these terms are not only confusing but even misleading. We suggest

replacing them with three simple terms - tuberculous reactivity, primary infection and TB

disease (Box 3).

The incubation period of TB - studies from the preantibiotic era

The relative risk of remote reactivation of LTBI is assumed to be high enough to

make it a substantial contributor to the global TB epidemic. But what is the actual risk of

active TB developing from a remote infection? Three longitudinal studies of TB

acquisition and progression conducted in the pre-antibiotic era in Norway and Sweden (7-

9). Careful monitoring by astute clinicians allowed for a reproducible time-line between

the acquisition of primary infection and the development of active TB. Poulsen, while

working at the tuberculosis station in the Faroe Islands from 1939 to 1947 was able to

trace the time of exposure to TB to within two weeks and often to a single day (8). Thus,

he determined the incubation period of primary infection - new tuberculous reactivity

often accompanied by the clinical features (as described in Box 3) to be within six weeks

(Figure 1A). In this same cohort, the incubation period of active TB was typically within

months and almost always within a year (Figure 1B). Wallgren, working in Stockholm,

Page 6 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlysimilarly found that active pulmonary TB generally developed within 1-2 years of

exposure (9)(Figure 1C). Finally, Gedde-Dahl, who monitored individuals regularly for

TST conversion and then for development of active disease found a similar incubation

period for the development of active TB, usually within months and rarely beyond two

years of newly documented tuberculous reactivity (7)(Figure 1D).

TB incubation period - additional insights from studies from the post antibiotic era

Once isoniazid became available in 1952, there was great interest in using it to

both treat TB disease and to use it as a chemoprophylaxis agent, i.e. to prevent the

development of TB disease once infection was diagnosed. In 1970, Shirley Ferebee

published a review of the controlled isoniazid chemoprophylaxis trials in the US

conducted between 1956 and 1966 (10); in Figure 2, we present her tabular data in a

graphical format. Examination of the placebo recipients reveals that, as in the older

studies, the likelihood of developing TB disease after infection dropped precipitously

after the first year, leaving a tail of what might be considered reactivation TB. However,

examination of the INH recipients provides additional insights into this tail. INH was

given for 12 months after infection, and its efficacy in preventing TB disease is reflected

in the five-fold reduction seen at year one, the end of the treatment period. After year

one, there was no substantial difference in the rates of TB disease between placebo and

INH, suggesting that newly acquired infections, rather than reactivation of the original

infection, were substantial contributors to this tail. These findings are mirrored in the

recent INH prevention trial in South African gold miners, where a transient decrease in

cases during the intervention period was followed by a convergence of the study groups

Page 7 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only(11). Thus, by extension, the true incidence of TB occurring remotely after infection (>2

years) may be lower than was surmised from the tail in the studies from the pre-antibiotic

era (7-9)(Figure 1B-D).

The inference that TB occurring remotely after a primary infection is

overwhelmingly due to newly acquired infection rather than reactivation of the remote

infection is supported by several other studies. In individuals who died of causes other

than tuberculosis in the pre-antibiotic era, the majority of visible tuberculous lung lesions

were sterile, as determined by guinea pig inoculation, an exquisitely sensitive assay

where even one viable TB bacillus causes mortal disease (12, 13). In particular, the

pathological lesions of primary TB are invariably calcified by five years of infection, and

85% of these calcified lesions in particular do not harbor any live bacteria - a finding that

is consistent with the findings in the epidemiological studies. These findings are

reinforced by two subsequent studies that performed detailed histopathological analyses

of lungs of individuals dying from TB to ask if they died from reactivation of their old

foci or newly acquired infection (14, 15). Both were also done in the pre-antibiotic era -

1932-1950, Buffalo (15), and 1940-1944, Paris (14) - allowing for the study of the role of

“reactivation” TB in the context of the natural history of disease. Terplan (15) found that

of those over 40 years of age, 87% died of new exogenous infection with the old focus of

infection remaining uninvolved. Similarly, Canetti (14) could rule out endogenous

reactivation in 74% of the cases; in the remaining 26% he could not definitively rule it

out because a small calcified primary focus could easily be missed in a lung riddled with

TB. Canetti also took an additional tack that took advantage of the fact that antibiotics

had entered the picture and resistance to them had already developed, by the time he

Page 8 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlysubmitted the paper in 1971. He reasoned that if reactivation of remote TB was

responsible for most active TB, then older individuals should have a much lower

frequency of drug-resistant TB than the younger population of TB patients. To test this

he examined the proportion of isolates with drug resistance in sequential age cohorts of

TB sufferers, ranging from 15 to ≥60 years. While he did find that resistance decreased

with age, the difference was minimal; 9.2% of all patients had resistance to isoniazid,

streptomycin or paramino-salicylic acid, compared to 7.6% of those aged 60 years and

older, which supports reinfection rather than reactivation in most of these patients.

In sum, both histopathological and epidemiological approaches suggest a far

greater role for exogenous infection than reactivation of old primary TB.

The incubation period of TB remains unchanged in the 21st century

Could the natural history of TB could have changed in the intervening half

century since the earlier studies were performed, so that the median incubation period is

now longer? Three studies (two from the Netherlands and one from Canada) show that

the incubation period of TB remains unchanged in the 21st century (16-18). Sloot et al.

identified recent active TB cases in Amsterdam, and monitored their TST-positive

contacts who did not take isoniazid prophylaxis, for ten years (2002-2011)(18). They

found that 75% of cases occurred within a year of conversion and 97% within two years

(Fig 3A). The study confirmed that children and adolescents were at greater risk of

developing active TB, but in all age groups the time-line of developing TB was the same

- overwhelmingly with the first year (Figure 3B).

Page 9 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyThe other two studies combined epidemiologic and molecular fingerprinting

methods for an increased accuracy of the incubation period for each individual.

Borgdorff et al. identified secondary TB cases among contacts of index cases, and

confirmed strain identity between index and secondary case using molecular

fingerprinting methods (16). They found that the median incubation period was 1.3 years

(95% CI 1.1-1.4 years, range 0 -12.8 years). The probabilities of developing disease

within one, two and five years were 45%, 62% and 83% respectively. A Canadian study

tracked transmission of infection during an outbreak using a clever methodology that

used a time-labelled genome phylogeny of the M. tuberculosis strains to estimate the time

of infection for each of the secondary cases (17). The majority of the 50 secondary cases

resulting from this outbreak presented with TB disease within two years of infection

(Figure 4). Because both studies used molecular methods to track transmission, and were

conducted in an otherwise low-incidence setting, these results unambiguously confirm

the previously described time-line. In summary, the typical incubation period of TB

disease is unchanged and remains a few months to two years. The importance of recent

infection as a risk factor for active TB was emphasized by Houben and colleagues in a

modelling paper who noted that while 25% of the world’s population has evidence of

LTBI, only 0.7 to 0.9% has recent TB infection (defined as acquired within 2 years) (19).

Is there a late spike of TB disease?

The aforementioned epidemiological studies were carried out up to 10 years so

the possibility that the reactivation of TB after that period might be missed. If this were

the case, then there would be an increase of TB incidence decades after TB infection.

Page 10 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlySuch a second increase in TB incidence would have been missed by these studies. The

antibiotic resistance data from the Canetti study suggests that this was not the case.

However, it was important to examine this question rigorously given the widely held

belief that reactivation TB most frequently occurs later in life when immunity wanes or

intercurrent illness occurs. Wiker and colleagues did just this in 2010 by testing the

hypothesis that TB incidence increases with age (20). These authors analyzed the

incidence of TB in Norway in 10-year birth cohorts starting 1879-1888 and up to 1959-

1968. They analyzed the incidence of TB in these cohorts over 20-30 years depending on

the cohort. If the rate of reactivation TB increases with age, there would have been an

increased incidence. Instead, in all cohorts, they saw a decreased incidence with time.

The rate of decline was greater in the younger cohorts consistent with the sharp decline in

TB incidence after the first and second years seen in all the earlier studies (7-9, 18).

Interestingly, there continued to be a decline in the older cohorts, consistent with the idea

that infected individuals who do not progress during the first two years either contain or

clear the infection (Figure 5).

In sum, contrary to the prevailing view, there is no bi-modal distribution

separating ‘primary progressive disease’ from ‘reactivation disease’. Rather, the low

rate of TB disease many years after infection continues to dwindle with age.

What does the presence of TB immunoreactivity really mean?

In light of the findings that the risk of TB drops precipitously after the first one to

two years and continues to drop further, we have revisited the assertion that one-third of

the world’s population is believed to be latently infected with Mycobacterium

Page 11 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlytuberculosis (http://www.who.int/tb/publications/ltbi_document_page/en/). This statistic

is derived from the finding that ~ one-third of the world’s population exhibits

immunoreactivity to TB, as manifested by a positive TST or IGRA test (Box 3). The

basis of adaptive immunity of course is that the inciting pathogen need not remain present

for such a memory response. Therefore, tuberculous reactivity must encompass both

current and past infection. But are there data that directly suggest that TB reactivity can

persist after M. tuberculosis is cleared? Indeed findings from several reports shed light

on this question. Atuk and Hunt (21)examined TST persistence at the end of one year of

INH treatment of asymptomatic TST-positive hospital employees. For recent converters

(< one year), only 5/20 individuals remained TST-positive at the same level, while the

remainder became TST-negative or decreased their TST magnitude. In contrast, all 17

individuals who had been asymptomatically TST-positive for more than a year remained

so after the year of INH treatment. A study done in naval officers had virtually identical

findings (22). Almost all individuals who had been TST-positive for only a few weeks at

the start of INH therapy reverted quickly within three months; in contrast all individuals

who had been positive for > one year remained positive at the end of a year of treatment.

These findings are consistent with the idea that TB immunoreactivity can be retained well

after infection is cleared. The more stable immunoreactivity of long-term TST-positive

individuals is consistent with immunological memory being more robust and long-term if

the infection lingered for longer before being cleared by the host. This conclusion is

corroborated by a study that looked for TST reversion after treatment of active TB which

found that TST-positivity was retained in all 38 patients even though they had completed

a treatment regimen associated with <3% five-year recurrence (23). Together, these

Page 12 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlystudies suggest that the estimate of the burden of “LTBI” is an overestimate as it reflects

immunoreactivity to past or present infection. Finally, it is important to note that not only

are these individuals not at higher risk for TB but multiple studies have suggested that

they may actually be protected against TB, upon new exposure (24-26)(Figure 6).

Summary and implications of our analyses

It is true that asymptomatic Mycobacterium tuberculosis infection can

occasionally result in disease decades later. Perhaps the most dramatic example is

genome-sequence confirmed father-to-son transmission in Denmark with a 33 year

interval between infection and disease (27). As clinicians caring for patients in low

transmission countries, we predominantly see cases of “reactivation TB”: elderly people

who contracted infection when they were very young at a time when TB was rampant and

individuals immigrating to low transmission countries from high transmission who

present with the strains prevalent in their country of birth (5, 6, 28). But it is important to

recognize that the respective contribution of recent versus remote infection differs in high

and low transmission settings (4). In high endemicity areas, the vast majority of the

burden is accounted for by newly acquired disease whereas in low endemicity areas, the

bulk of active TB is due to remote infection(19). This sort of shift in relative

contributions to risk is appreciated in other diseases as well. When polio vaccination

campaigns succeed, the relative proportion of polio due to the live attenuated vaccine

increases, because circulation of wild-type poliovirus has decreased. Similarly, when

there is reduced HIV transmission in the men who have sex with men (MSM) population,

the proportion of new infections in women increases. Even when TB rates decline and a

Page 13 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlygreater proportion of disease is due to a remote infection, the highest yield activity

remains identifying active TB cases and interrupting transmission.

Every single longitudinal study on the time-line of TB infection and disease

supports a median incubation period of a few months to two years, with only a small

proportion of outliers who get disease farther out. Importantly, the curve is not bimodal

as has been widely believed, but rather unimodal. Therefore, there is no epidemiological

basis to differentiate “primary progressive TB” from “reactivation TB”, just as we do not

distinguish “early AIDS” from “reactivation AIDS”. Furthermore, there is little

epidemiologic support for a special bacterial state (e.g. dormancy) during the

asymptomatic state. Whole genome sequencing on the bacteria that the Danish patient

had harbored for 33 years and other isolates that were cultured after decades-long

incubation periods were found to have the same mutation frequency (0.2-0.3 SNPs per

genome per year) as seen in outbreak strains (29).

While the biology of latency is tantalizing, the importance of this phenomenon for

TB control, and for prioritizing TB research is less convincing. While it is well-cited that

approximately 10 million new cases of TB disease are diagnosed each year, we have been

unable to find a published estimate of the number of individuals newly infected each

year, even though these are at highest risk of developing disease. Assuming that each

contagious case infects 3-6 contacts, based on revised estimates of the Styblo rule (30), if

there are more than 10.4 million prevalent cases of TB worldwide

(http://www.who.int/tb/publications/global_report/en/), there could be up to ~ 60 million

new TB infections per annum – how can we prevent these infections, and failing that,

how can we prevent these newly-infected individuals from developing disease? How can

Page 14 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Onlywe detect the contagious cases sooner, either through process or product innovations?

What can we learn through genetic epidemiology of the differing outcomes in their

contacts, between pathogen elimination, infection and disease? With a vaccine that

prevents ‘only’ 50% of infections, there could be 30 million fewer new infections per

year, and in theory, the trial length to show such an effect would be short, requiring only

the monitoring of TST or IGRA conversions. Moreover, a vaccine that reduces disease by

a similar degree could be shown to achieve this efficacy in trials with just 18-24 months

of follow-up.

Finally, the WHO has pledged to eliminate TB by 2035 through its “End TB

strategy” (http://www.who.int/tb/strategy/end-tb/en/). Its staged implementation plan

includes “new tools: a vaccine, new drugs & treatment regimens for treatment of active

TB disease and latent TB infection” which again reflects the

the concern that the large reservoir of people with LTBI may stymie efforts to achieve

this goal. We hope that our analysis of the numerous studies showing that most TB

outbreaks have shown that the majority of TB cases occur within 18-24 months of

infection, and that the few cases of active TB from remote infection in low burden areas

are self-contained and do not generally lead to major disease outbreaks (5, 6) will lead to

a reconsideration of the current strategy. If focused attention was applied to those with

active TB disease and their newly infected contacts, might it be possible that TB

elimination can be achieved sooner than projected?

Page 15 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


REFERENCES

1. Dye C, Glaziou P, Floyd K, Raviglione M. Prospects for tuberculosis elimination.

Annual review of public health. 2013;34:271-86. PubMed PMID: 23244049.

2. Bayer R, Castro KG. Tuberculosis Elimination in the United States - The Need

for Renewed Action. The New England journal of medicine. 2017 Sep 21;377(12):1109-

11. PubMed PMID: 28930506.

3. Esmail H, Barry CE, 3rd, Young DB, Wilkinson RJ. The ongoing challenge of

latent tuberculosis. Philosophical transactions of the Royal Society of London Series B,

Biological sciences. 2014;369(1645):20130437. PubMed PMID: 24821923. Pubmed

Central PMCID: 4024230.

4. Cardona PJ. Reactivation or reinfection in adult tuberculosis: Is that the question?

International journal of mycobacteriology. 2016 Dec;5(4):400-7. PubMed PMID:

27931680.

5. Walker TM, Lalor MK, Broda A, Ortega LS, Morgan M, Parker L, et al.

Assessment of Mycobacterium tuberculosis transmission in Oxfordshire, UK, 2007-12,

with whole pathogen genome sequences: an observational study. The Lancet Respiratory

medicine. 2014 Apr;2(4):285-92. PubMed PMID: 24717625. Pubmed Central PMCID:

4571080.

6. Yuen CM, Kammerer JS, Marks K, Navin TR, France AM. Recent Transmission

of Tuberculosis - United States, 2011-2014. PloS one. 2016;11(4):e0153728. PubMed

PMID: 27082644. Pubmed Central PMCID: 4833321.

Page 16 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only7. Gedde-Dahl T. Tuberculous infection in the light of tuberculin matriculation.

American journal of hygiene. 1952 Sep;56(2):139-214. PubMed PMID: 12985565.

8. Poulsen A. Some clinical features of tuberculosis. 1. Incubation period. Acta

tuberculosea Scandinavica. 1950;24(3-4):311-46. PubMed PMID: 14783027.

9. Wallgren A. The time-table of tuberculosis. Tubercle. 1948 Nov;29(11):245-51.

PubMed PMID: 18101320.

10. Ferebee SH. Controlled chemoprophylaxis trials in tuberculosis. A general

review. Bibliotheca tuberculosea. 1970;26:28-106. PubMed PMID: 4903501.

11. Churchyard GJ, Fielding KL, Grant AD. A trial of mass isoniazid preventive

therapy for tuberculosis control. The New England journal of medicine. 2014 Apr

24;370(17):1662-3. PubMed PMID: 24758626.

12. Opie EL, and J. D. Aronson. Tubercle bacilli in latent tuberculous lesions and in

lung tissue without tuberculous lesions. Arch Pathol Lab Med 1927;4:1-21.

13. Feldman WH, Baggenstoss AH. The residual infectivity of the primary complex

of tuberculosis. The American journal of pathology. 1938 Jul;14(4):473-90 3. PubMed


14. Canetti G, Sutherland I, Svandova E. Endogenous reactivation and exogenous

reinfection: their relative importance with regard to the development of non-primary

tuberculosis. Bulletin of the International Union against Tuberculosis. 1972 Feb;47:116-

34. PubMed PMID: 5077111.

15. Terplan K. [Pathogenesis of postprimary tuberculosis, in relation to chronic

pulmonary tuberculosis (phtisis)]. Bibliotheca tuberculosea. 1951;5:186-219. PubMed

PMID: 13018097.

Page 17 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only16. Borgdorff MW, Sebek M, Geskus RB, Kremer K, Kalisvaart N, van Soolingen D.

The incubation period distribution of tuberculosis estimated with a molecular

epidemiological approach. International journal of epidemiology. 2011 Aug;40(4):964-

70. PubMed PMID: 21441552.

17. Hatherell HA, Didelot X, Pollock SL, Tang P, Crisan A, Johnston JC, et al.

Declaring a tuberculosis outbreak over with genomic epidemiology. Microbial genomics.

2016 May;2(5):e000060. PubMed PMID: 28348853. Pubmed Central PMCID: 5320671.

18. Sloot R, Schim van der Loeff MF, Kouw PM, Borgdorff MW. Risk of

tuberculosis after recent exposure. A 10-year follow-up study of contacts in Amsterdam.

American journal of respiratory and critical care medicine. 2014 Nov 1;190(9):1044-52.


19. Houben RM, Dodd PJ. The Global Burden of Latent Tuberculosis Infection: A

Re-estimation Using Mathematical Modelling. PLoS medicine. 2016

Oct;13(10):e1002152. PubMed PMID: 27780211. Pubmed Central PMCID: 5079585.

20. Wiker HG, Mustafa T, Bjune GA, Harboe M. Evidence for waning of latency in a

cohort study of tuberculosis. BMC infectious diseases. 2010 Feb 23;10:37. PubMed


21. Atuk NO, Hunt EH. Serial tuberculin testing and isoniazid therapy in general

hospital employees. Jama. 1971 Dec 20;218(12):1795-8. PubMed PMID: 5171441.

22. Houk VN, Kent DC, Sorensen K, Baker JH. The eradication of tuberculosis

infection by isoniazid chemoprophylaxis. Archives of environmental health. 1968

Jan;16(1):46-50. PubMed PMID: 5638224.

Page 18 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only23. Sepulveda RL, Araya D, Ferrer X, Sorensen RU. Repeated tuberculin testing in

patients with active pulmonary tuberculosis. Chest. 1993 Feb;103(2):359-63. PubMed

PMID: 8432119.

24. Andrews JR, Noubary F, Walensky RP, Cerda R, Losina E, Horsburgh CR. Risk

of progression to active tuberculosis following reinfection with Mycobacterium

tuberculosis. Clinical infectious diseases : an official publication of the Infectious

Diseases Society of America. 2012 Mar;54(6):784-91. PubMed PMID: 22267721.

Pubmed Central PMCID: 3284215.

25. Heimbeck J. Immunity to tuberculosis. Archives Internal Medicine.

1928;41(3):336 -42.

26. Stead WW. Tuberculosis among elderly persons: an outbreak in a nursing home.

Annals of internal medicine. 1981 May;94(5):606-10. PubMed PMID: 7235393.

27. Lillebaek T, Dirksen A, Baess I, Strunge B, Thomsen VO, Andersen AB.

Molecular evidence of endogenous reactivation of Mycobacterium tuberculosis after 33

years of latent infection. The Journal of infectious diseases. 2002 Feb 1;185(3):401-4.


28. Reed MB, Pichler VK, McIntosh F, Mattia A, Fallow A, Masala S, et al. Major

Mycobacterium tuberculosis lineages associate with patient country of origin. Journal of

clinical microbiology. 2009 Apr;47(4):1119-28. PubMed PMID: 19213699. Pubmed

Central PMCID: 2668307.

29. Lillebaek T, Norman A, Rasmussen EM, Marvig RL, Folkvardsen DB, Andersen

AB, et al. Substantial molecular evolution and mutation rates in prolonged latent

Page 19 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyMycobacterium tuberculosis infection in humans. International journal of medical

microbiology : IJMM. 2016 Nov;306(7):580-5. PubMed PMID: 27296510.

30. van Leth F, van der Werf MJ, Borgdorff MW. Prevalence of tuberculous infection

and incidence of tuberculosis: a re-assessment of the Styblo rule. Bulletin of the World

Health Organization. 2008 Jan;86(1):20-6. PubMed PMID: 18235886. Pubmed Central

PMCID: 2647347.

Page 20 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyBOX 1 Definitions/Statements From Authoritative Bodies and Online Medical

Resources

World Health Organization (WHO)

(http://www.who.int/tb/areas-of-work/preventive-care/ltbi_faqs/en/)

Latent TB infection (LTBI) is a condition in which TB bacteria (M. tuberculosis) survive

in the body in a dormant state. Persons with LTBI do not have active tuberculosis but may

develop it in the near or remote future. The overall lifetime risk of developing active TB is about 10%.

Canadian TB Standards

(http://www.phac-aspc.gc.ca/tbpc-latb/pubs/tb-canada-7/assets/pdf/tb-standards-tb-

normes-appa-eng.pdf)

Latent tuberculosis infection (LTBI) The presence of latent or dormant infection with

Mycobacterium tuberculosis. Patients with LTBI have no evidence of clinically active

disease, meaning that they have no symptoms, no evidence of radiographic changes that

suggest active disease and negative microbiologic tests; they are non-infectious.

Primary tuberculosis This includes primary respiratory tuberculosis and tuberculous

pleurisy in primary progressive tuberculosis.

Reactivation tuberculosis The development of active disease after a period of latent

tuberculosis infection. In Canada, the term "reactivation" tuberculosis was previously

used to refer to a recurrence.

Post-primary tuberculosis older term – see reactivation tuberculosis.

Centers for Disease Control and Prevention (CDC)

(https://www.cdc.gov/drugoverdose/prescribing/guideline.html) Latent TB infection a condition in which TB bacteria are alive, but inactive in the body.

People with latent TB infection have no symptoms, don't feel sick, can't spread TB to others,

and usually have a positive TB skin test or positive TB blood test reaction. But they may

develop TB disease if they do not receive treatment for latent TB infection.

TB disease an illness in which TB bacteria are multiplying and attacking a part of the body,

usually the lungs.

National Institute for Health and Care Excellence (NICE)

(https://www.nice.org.uk/guidance)

Latent TB Infection with mycobacteria of the M. tuberculosis complex in which the bacteria

are alive but not currently causing active disease.

Post-primary TB The stage following primary tuberculosis. This is when infection with the

bacteria has advanced to disease, possibly symptomatic, with bacterial growth

demonstrable by culture.

Primary TB the initial stage of disease following infection with TB bacteria in individuals

with no prior antimycobacterial immunity. It is often asymptomatic, but can be detected by

tuberculin tests or interferon-gamma release assays.

Page 21 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyReactivation The advancement of old latent TB (whether previously detected or not) into

active TB.

UpToDate (online medical resource from Wolters Kluwer)

(https://www.uptodate.com/home)

Latent TB — During this latent stage of TB, also called TB infection, the person is well and

cannot spread the infection to others. If the person is treated at this stage, active TB can

usually be prevented.

Active TB — Active TB may develop if latent infection is not fully treated. This is called

reactivation TB, and it occurs in 5 to 10 percent of people with latent infection at a later

time in their lives.

Reactivation TB may occur if the individual's immune system becomes weakened and is no

longer able to contain the latent bacteria. The bacteria then become active; they overwhelm

the immune process and make the person sick with TB. This also is called TB disease.

Page 22 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyBOX 2 Funding for TB latency

2001 NIH RFA

Title “Response to the Presidential Vaccine Initiative - Overcoming the Tuberculosis

Latency Challenge” (https://grants.nih.gov/grants/guide/rfa-files/RFA-AI-01-009.html)

Wording of call “The objective of this Request for Applications (RFA) is to stimulate

investigator-initiated research to elucidate the mechanisms underlying persistent,

asymptomatic infection (also referred to as “latency”) with Mycobacterium tuberculosis.

… A better understanding of persistent infection is an essential step toward developing

improved intervention strategies to eliminate tuberculosis as a public health problem.”

2005 BMGF Grand Challenge in Global Health, a family of initiatives fostering

innovation to solve key global health and development problems

(https://gcgh.grandchallenges.org/)

One of the 43 funded grants was entitled “Drugs for treatment of latent tuberculosis”

with the following abstract: An estimated 2 billion individuals—a third of the world's

population—have been exposed to Mycobacterium tuberculosis (MTB) and carry the

infection in its latent form, retaining a lifelong risk of developing TB disease. Programs

to control tuberculosis now focus on childhood vaccination and treatment for people with

active disease. Reversing TB's spread, however, requires an intervention that will prevent

disease in those who are already infected. The lack of knowledge about the biology of

latent TB infection stands in the way of the development of such an intervention. [An]

international team of researchers from the U.K., U.S., Singapore, Korea, and Mexico is

attempting to further elucidate the fundamental biology of latency and use this knowledge

to develop drugs against latent TB.

2008 BMGF Grand Challenges Explorations

(https://gcgh.grandchallenges.org/challenge/explore-basis-latency-tuberculosis-round-1

and https://gcgh.grandchallenges.org/challenge/explore-basis-latency-tuberculosis-round-

2).

Call for LTBI proposals: “ ROADBLOCK: Most humans exposed to Mycobacterium

tuberculosis contain the infection in an asymptomatic latent form. One third of the

world’s population is estimated to have latent TB, representing a vast reservoir from

which active disease and subsequent transmission propagates. Interventions that identify

and eliminate latent infection might break the cycle of disease transmission and reverse

the TB epidemic.”

Of the 1313 awards made, 15 were to study TB latency

(https://gcgh.grandchallenges.org/grants?f[0]=field_challenge%253Afield_initiative%3A

37073&f[1]=field_challenge%253Afield_short_title%3ATuberculosis%20Latency)

Page 23 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review Only BOX 3 Suggested Simplified Terms

Tuberculous Reactivity Indirect evidence of present or past infection with

Mycobacterium tuberculosis as inferred by a detectable adaptive immune response to M.

tuberculosis antigens (tuberculin skin test (TST) or interferon gamma release assay

(IGRA)) in an asymptomatic subject.

Primary infection Evidence of new tuberculous infection, obtained via a TST conversion

or a new positive IGRA test, which may be asymptomatic or accompanied by transient

fever, erythema nodosum, elevated ESR and/or characteristic roentgenographic

abnormalities.

Active tuberculosis (TB) Evidence of progressive disease of the lung and/or other

organs generally accompanied by a positive culture for M. tuberculosis and/or

roentgenographic findings and/or histopathology consistent with tuberculosis.

Page 24 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

Confidential: For Review OnlyBOX 4 KEY MESSAGES

- There is no evidence for a bi-modal distribution that distinguishes primary progressive

TB from reactivation TB. Rather, the incubation period of TB is typically several months

to two years, and after that, disease is relatively infrequent.

- A positive TST or IGRA response, indicative of immunoreactivity to TB, does not

necessarily indicate an ongoing infection, as tuberculous reactivity can persist after

infection has been cleared

- Classifying 2 billion persons with evidence of tuberculous reactivity as having latent TB

infection (LTBI), and hence being at risk of active TB, may divert fundamental research

and public health interventions away from transmissible active TB disease and newly

infected individuals at highest risk of progression to disease.

Page 25 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 1 The incubation period of TB. A. The cumulative percentage of individuals

developing primary infection over time after exposure to a patient with active pulmonary

TB. B. The cumulative percentage of these same individuals developing active TB in any

organ over time. Data for A and B from text descriptions in Poulsen, 1950 (8). C. The

cumulative percentage of individuals developing active pulmonary tuberculosis from the

time of onset of primary infection. Redrawn from figure in Wallgren, 1948 (9). D. The

cumulative percentage of individuals developing active TB from the time of onset of

primary infection. Data drawn from tables in Gedde-Dahl, 1952 (7).

1st Qtr

2nd Qtr

3rd Qtr

1st Yr

2nd Yr

3rd - 7th Yr

0

25

50

75

100Cumulative %9% 3rd-7th yr

10 20 30 40 500

25

50

75

100

Days

Cumulative %

Primary

90 180

270

360

450

540

630

720

0

25

50

75

100

Days

Cumulative %

0 1 2 3 4 5 60

25

50

75

100

Years

Cumulative %

A B

C D

Figure 1

Page 26 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 2 Rates and incubation period of active tuberculosis in tuberculin-positive

(induration ≥ 5 mm) household contacts of patients with recently diagnosed tuberculosis.

Half of the contacts were treated with isoniazid, and half with placebo, for one year, then

observed for another ten years. Total number of subjects for both groups was ~ 13,200.

Error bars show 95% confidence intervals. Data taken from tables in Ferebee, 1970 (10)

1 3 5 7 90

200

400

600

800

1000

Study year

rate/100,000

INH

Placebo

Figure 2

Page 27 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 3 Time to development of active tuberculosis in contacts of patients with active

tuberculosis. A. Total population of 739 subjects. B. Fraction of subjects developing

tuberculosis over time by age group. The incubation time is the time since the

tuberculosis diagnosis of the index case. Redrawn from data in Sloot, et. al, 2014 (18).

0 1 2 3 4 5 6 7 8 9 100

25

50

75

100

Years after exposure

Pulm

onary Tb

(Cumulative %

)

0 1 2 3 4 5 6 7 8 9 100

10

20

30

Years after exposure

<5 yrs (N=36)

10-14 yrs (N=84)

≥15 yrs (N=619)

Fraction of exposed people

developing TB (%)

A

B

Figure 3

Page 28 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 4 Incubation time to onset of active tuberculosis in a large tuberculosis outbreak

inferred by whole genome sequencing of isolates and mutation rates of isolates. Data

redrawn from Hatherell, et. al, 2016 (17).

0.5

1.0

1.5

2.0

2.5

3.0

3.5

4.0

4.5

5.0

0

5

10

15

20

25

Years

Relative frequency (%)

Figure 4

Page 29 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 5 Decreasing rates of tuberculosis in different age cohorts of men. Redrawn from

original figure in Wiker, et al, 2010 (20).

30 40 50 60 7020

40

60

80

Age Cohort (yrs)

% Decline in 10 yr periods

Figure 5

Page 30 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960


Figure 6 Rates of tuberculosis acquisition in nursing students by pre-exposure tuberculin

skin test result. Data from Heimbeck, 1928 (25).

TST pos TST neg0

10

20

30TB Acquisition (%)

200 210N

Figure 6

Page 31 of 31


BMJ

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

the forgotten time-table of tuberculosis · confidential: for review only the forgotten time-table...

Documents