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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
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
PMID: 19970404. Pubmed Central PMCID: 1964987.
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
https://mc.manuscriptcentral.com/bmj
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.
PubMed PMID: 25265362.
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
PMID: 20178619. Pubmed Central PMCID: 2843612.
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
https://mc.manuscriptcentral.com/bmj
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.
PubMed PMID: 11807725.
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960
Confidential: For Review Only
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
https://mc.manuscriptcentral.com/bmj
BMJ
123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960