Latent tuberculosis (part 1) - The Invisible Plague
In the last month two studies have been published which have focused our attentions on latent TB - not least because their contents have highlighted the potential importance of some of our latest findings.
What follows is an extended discussion on the complexities of latent tuberculosis. It consists of four parts:
Part 1: discusses latent TB, the primary invisible plague, and the current efforts to address it.
Part 2: discusses the secondary doubly invisible plague of latent TB that is drug resistant.
Part 3: discusses the recent data released to us by the North Korean Ministry of Public Health relating to the potential use of moxa as an adjunctive tool for treating latent TB including latent MDR-TB.
Part 4: discusses some further incoherencies which emerge from these new studies and some additional complexities.
So what is ‘latent’ TB?
A latent TB infection (LTBI) is defined by the WHO as being “a state of persistent immune response to M. tuberculosis without any clinical clinically manifested evidence of active TB disease”. Until recently this was confirmable by a simple skin test which came with several shortcomings but has more recently been improved with the introduction of the so-called ‘IGRA’ test. This is a blood test which measures the persistence of release of a specific cytokine that is released by the immune system after exposure to TB.
Thankfully most TB infections never develop into a condition which is actively harmful to the patient, so is of no harm to others nor is potentially lethal. Most TB infections initially occur in the lungs, but the immune system is slow to recognise them largely because the TB mycobacteria replicate much more slowly compared to other infections. Because of this they are generally well established before they are picked up, so the immune system finds it hard to actually destroy the bacteria. The normal response then is to ‘wall off’ the infection in the lungs, evidence of which is measurable in the “persistent immune response” mentioned above. In the vast majority of cases, that’s the end of the story and the invading bacteria never ‘break out’ into re-activated potentially lethal disease.
This ‘persistent immune response is why so-called ‘latency’ might be more correctly described as a ‘sub-clinical’ infection which has effectively contained the disease. It's worth mentioning that when TB was at its height in Europe during the industrial revolution, probably 80% of the population in many cities would have been latently infected. And back then what's now understood to be a ‘re-activation’ of this latency into symptomatic disease was both a constant fear and a common cause of death. Such re-activation from latency is now believed to occur in between 5-10% of these latent cases over the course of a lifetime, and the probability of this happening is largely predicated by several known risk factors, but most of all by the relative state of the infected person’s immune system. (A little later in this first piece we will make our own rudimentary calculation of this risk in any given year).
Some background to this, and a discussion of the current estimated numbers
Back in 1993 the WHO officially announced TB to be a ‘global emergency’, and four years later it began publishing its annual Reports. These have since developed estimates of how much of the disease might be latent or sub-clinical, how much of it might be active or reactivated, and later how much might be drug-resistant.
Over the past twenty years the amount of active disease has been regularly uodated, retrospectively re-estimated (mostly upwards) as more surveillance data has emerged, but despite such retrospective re-estimations it is still nevertheless reckoned that the global peak of drug-susceptible disease was reached in 2005, since when the rates have been declining very slowly. In the most recent WHO Report, for instance, it’s pronounced that 54 million lives have been saved by the WHO-approved intervention since 2000 (though it has to be said that, while this figure is huge, it’s not normally added that probably 40 million lives have been lost to the disease in this same period).
But over this whole period (at least until a key paper was published in 2016) the original estimated percentage of humanity that was latently infected remained unchanged. This percentage was originally set in 1997 at a quite terrifying 32%[i] (i.e. in the late 1990s one in every three human beings alive on the planet was believed to be carrying an infection that could be considered potentially lethal).
Of course, the probability that this percentage remained constant over the next twenty-odd years was frankly unlikely, particularly after 2005 with the active disease reported to be beginning to decline. By 2015, for example, with estimates of the numbers of incident cases of TB being reported to have been very slowly reducing for a decade, it was pretty improbable that the 32% percentage would still stand. So it turned out, with this number officially challenged in 2016 by the publication of a re-estimation of the global latent burden by Rein Houben and Peter Dodd, two UK epidemiologists who used complex mathematical modelling to re-estimate it.[ii] This study calculated that probably 23% of humanity had been carrying the infection during 2014 (their baseline year). Essentially this was an encouraging re-estimation because one-in-every-four human beings were now carrying the infection rather than the previous one-in-three.
This re-estimation has been further roughly confirmed just in the last month by the pre-publication of another re-estimate, this time by a team of Scandinavian experts. In this instance their percentage was calculated by developing a systematic review of all the available data on the subject[iii] and it largely confirmed the UK study from three years before (that one reckoned 23%, this new study reckons a slightly higher 25%).
These percentage reductions are naturally welcome. Nevertheless, given that TB was initially declared a Global Emergency by the WHO back in 1993, a proportional reduction of only 7% or 9% (take your pick) over a quarter of a century is hardly anything to write home about. In fact, both re-estimations suggest that TB could be with us for a very long time indeed unless things change dramatically.
But it’s the actual numbers that can be computed from these percentages that are worrying as well. In 1995 the global population was estimated to be 5.6 billion and so a thirty-two percentage of this number amounted to 1.8 billion latent infections. In 2014, 23% of that year's global population spat out 1.7 billion. By 2018, however, the global population was estimated to have risen to 7.7 million, and 25% of this total spits out pretty much the same number as the original 1.8 billion.
In fact it’s a hundred million higher...
In other words, twenty-five years of concerted effort has apparently not even made a dent in the numerical number of inhabitants of this pool of latent TB. Perhaps it’s not so much of a surprise then that Houben and Dodd, the authors of that 2016 paper, had warned that, even if disease transmission were to have miraculously stopped in 2014 (i.e. from that moment on there had been no more infections from any active infectious cases pouring more cases into the latent pool), the World Health Assembly’s target of ‘ending TB’ by 2035 was effectively out of reach. This was allowing for the expected proportion of the existing prevalent pool of latent cases to work themselves out into active disease in the following twenty years. (The reason the authors of this paper had used data from 2014 as their baseline, incidentally, was because that year was the most recent year from which they had available numbers to work from).
This must have been sobering food for thought for anyone who scrutinised this study back then, particularly bearing in mind that stopping transmission back in 2014 was an impossibility, and that the reality was (and still is) that actually there are at least 10 million new reactivations each year stoking the cycle on infections and therefore the numbers in the latent pool.
Notwithstanding this, given that these two key numbers (1.8 billion and 10 million) seem to have remained so relatively stable, they do help us with one thing: they allow us to take a primitive stab at your annual risk of developing active TB if you harbour a latent infection. If 10 million cases are emerging year on year from a pool of 1.8 billion latencies it suggests that you have roughly a one-in-two-hundred chance of developing active TB in any given year as long as survive. (Of course, this is simplistic because many factors impact on this probability, but nevertheless this is an interesting calculation to be able to make. In fact the risk is believed to be very high in the first year at around 10%).
But the reality is that a latent case of TB can be treated after it’s been diagnosed with a 60-90% chance of curing the sub-clinical infection.
What is being done about this
Given the challenges these numbers clearly pose to meeting the current targets to end TB by 2035, it may seem like they haven't been that seriously considered. The importance of doing something about them has been far from ignored, though. Last year’s High-Level Meeting (HLM) on TB at the UN, for instance, set a target for addressing exactly this aspect of the disease – demanding that at least 30 million latent cases to be found and treated in the five years running up to 2022, with 4 million of them to be under five years old.
Of course, 30 million is a spit in an ocean considering that this ocean consists of 1.8 billion latent cases (less than 2% in fact), but this isn’t the end of it. While these 30 million may only be a tiny percentage, the ones who are being particularly targeted are the ones considered to be at most significant risk of developing active TB.
So the current target from the UN’s HLM also demands that all nations focus on exactly these – the most vulnerable – and the WHO has broken them down into four target groups in order to do so:
people living with HIV (all countries);
children aged under 5 years who are household contacts of bacteriologically conﬁrmed pulmonary TB cases (all countries);
people aged ﬁve years or more who are household contacts of pulmonary TB cases (in upper middle-income and high-income countries with a low incidence of TB);
and an additional recommendation “to consider testing and treatment” for people aged 5 years or more who are household contacts of bacteriologically conﬁrmed pulmonary TB cases in countries with a high incidence of TB.
The WHO is careful to note that there are “at least” 30 million of these vulnerable cases available (because of course there must be many more out there who are also at high risk because of other risk factors). In line with this, in its Political Declaration the UN added that its target of finding and treating 30 million must be a threshold for a floor and not a ceiling.
There’s a mountain to climb here, however. Numbers of LTBI cases put on treatment every year are not yet even properly being reported so it's difficult to really judge how high the mountain is still. But the impending annual Global TB Report for 2019 (due in October) must now report, not just on each country’s numbers of latent cases treated, but also how they break down into these target groups so that each country can be properly monitored. If it doesn’t do so, then the Global Report will be out of step and incoherent with one of the main targets set by the UN a year into the target programme.
But there’s an obvious catch to this because of the timing of the demand and what's already in the pipeline. The target numbers required for reporting for 2018 (i.e. for the year that will be covered in this current year’s report) were actually only set at the end of 2018 soon after the HLM. As such, it’s pretty much impossible that this coming report is going to be very comprehensive in this respect and the prospects of reaching this overall target by 2022 is therefore going to get off to a shaky start.
Sadly, there’s some very recent evidence that this will be the case unfortunately. We received a bulletin from the Stop TB partnership dated 6th August 2019 (i.e. just this week) which soberly notes that “several high-TB burden countries have already started to use these indicative targets to plan their actions towards ending TB”.
"Already started to plan" in August 2019 for a programme that should have started in 2018?!!
Given that we’re now more than half way through the second year of a five year target programme and that “several countries” have only started using these “indicative” targets in order to “plan their actions”, things are not boding very well at all for this important target, and there’s plainly already significant catching up to do. Perhaps there's some cracking of whips needed - but if so calling the targets "indicative" isn't going to help much.
In fact, in terms of the numbers that were retrospectively set for all countries for 2018 (and which they are apparently only already planning around) these post-HLM national targets appear to require many countries to at least quadruple their annual treatment numbers through to 2022. Quite a few of them, it seems, must still be developing a preventative programme pretty much from scratch.
Pakistan, for instance, failed to report any data at all for children in 2017, but is now required to increase its commitment to treating LTBI by a factor of 7 between what was set for it for 2018 (and which it may not have even begun implementing) and 2022.
Indonesia (which similarly reported no data regarding children and has a terrible record of support for HIV coinfections) needs to ramp up its act by a factor of 10. It will be interesting to see how these countries report in the impending Global Report. What’s more, it’s far from clear on what grounds some of these national targets have been set.[iv]
Immediately below is a graphic from last year's WHO Global Report identifying the percentage provision of preventative treatment for children under five that was reckoned to be out there in 2017. This percentage is a proportion of the estimated total of children who are estimated to be eligible (which overall was reckoned to be just 23% of an estimated 1.3 million eligible children). From this map it can be seen that most high burden countries were hopelessly underreporting. Just look at Nigeria, India, China and Indonesia, the four countries with the highest numbers of TB cases - with only one of them providing any data at all, and the one that did only providing less than 25% coverage).
This next graphic (immediately below) from the same 2018 Report tells a similar story relating to the deficiency of preventative TB therapy for those newly enrolled in HIV care in a selected few countries.
But the real picture is sadly actually worse than shat's represented because, according to the last Global TB Report, “of the 30 high TB/HIV burden countries, [only] 15 reported providing TB preventive treatment to people newly enrolled in HIV care in 2017.” (We've added the word 'only' for emphasis because we feel this sentence needed it, by the way). It’s certainly worth wondering what was happening when all of these high burden countries merrily signed up to the Political Declaration made at the end of the UN HLM, because surely they must have realsied that this would involve some reporting.
It should be further noted that this chart above is a little disingenuous anyway in that it basically illustrates the proportion of those started on preventative treatment (in blue) and those diagnosed with TB and HIV who were started on treatment (in yellow) compared to those who were estimated in 2017 to be coinfected with both HIV and TB (in green). It only shows one category of high risk latent cases (HIV coinfections). Apart from children who are close contacts, wat about miners (who have very high risks indeed) or diabetics (etc., etc.). If the chart were to show the overall number of latent infections started on treatment against the estimated number of high risk latent infections in each of these countries it would unquestionably show something altogether more worrying still.
But there's something more, which relates to how much of this host of latent infections are already drug-resistant and who will therefore be much more difficult to successfully treat, and this aspect will be covered in Part 2.
[i] Dye C, Scheele S, Dolin P, Pathania V, Raviglione MC. Consensus statement. Global burden of tuberculosis: estimated incidence, prevalence, and mortality by country. WHO Global Surveillance and Monitoring Project. JAMA: the journal of the American Medical Association. 1999;282(7):677–86
[ii] Houben RMGJ, Dodd PJ. The global burden of latent tuberculosis infection: a re-estimation using mathematical modelling. PLoS Med 2016; 13: e1002152.
[iii] Cohen A, Mathiasen VD, Schoen T, Wejse C. The global prevalence of tuberculosis: a systematic review and meta-analysis. European Respiratory Journal. https://doi.org/10.1183/13993003.00655-2019
[iv] http://www.stoptb.org/assets/documents/global/advocacy/unhlm/UNHLM%20on%20TB%20-%20Preventive%20Therapy%20Targets.pdf for a list of country targets.