Latent TB (Part 3) – moxa therapy, viability in the invisibility?

So far we’ve looked at some of the complexities and uncertainties relating to these so-called ‘latent’ aspects of a pandemic plague of sub-clinical infection. We’ve wondered how the outflow from this immense 1.8 billion-strong ocean of potential infection into the active sea of re-activated infectious TB might be stifled with effective targeted treatment. Doing so could help starve the disease of what it essentially needs - but yet we seem to be as far away from this goal as ever, particularly in respect of the proportion of latency which is drug-resistant.

Ultimately this is a numbers game – and we can say two sure things about them. Firstly, we can say that all the numbers are very uncertain; but secondly that the odds within the numbers still appear to be leaning in TB’s favour, not just because it’s impossible to tell if a case is drug-resistant when it’s latent, but because there now appears to be a disquieting amount of drug-resistance (either mono-drug resistance or multi-drug resistance) in the ocean of infection with nothing to treat it.

We’ve seen that all four approved treatments for treating latent TB are highly vulnerable to drug-resistance (period..), but also that their potential replacements (if they prove to be safe and effective to use on the high-priority target groups like under-5’s and HIV co-infections) are inevitably going to be expensive to roll out.

Meanwhile a set of targets for finding and treating these high-risk groups has been set by the UN and they're intended to be actively monitored - but those setting these targets seem to have too casually glossed over these complexities regarding MDR-TB in the heat of the hasty follow-up activities after last year’s High Level Meeting.

So what may happen as a result?

It’s difficult to be specific, but several things are possible.

1. A significant number of these treatments are going to fail.

2. In doing so existing drug-resistances may get stoked,

3. If this happens it will actually add to the challenges of ending TB by 2035 rather than reduce them.

4. Many latent TB cases will be inevitably treated unnecessarily with strong antibiotics, potentially leaving patients constitutionally more compromised than they were before treatment.

5. If this happens, then the probability that their latent infection will ‘break out’ into re-activated infectious disease is potentially more likely than before they had treatment

6. And (if it does) it is also probable that their disease will be drug-resistant.

7. At the same time a substantial cohort of people will die unnecessarily, paradoxically because they were initially targeted for treatment on account of their vulnerability to this disease.

8. And so the proportion of MDR-TB ( as well as the numbers) generally will increase.

Without question, some of these points are debatable but all are possible. The point is that the very cases who are being targeted for treatment are already identified as being highly vulnerable and so they really don’t need unnecessary futile treatment. (Does anyone relish three-to-nine months of strong antibiotics with a risk of ending up more vulnerable?).

(If you think we're wrong about this, by the way PLEASE get in touch to correct us. We welcome engagement.)

A recognised side-effect of rifapentine, for instance, is neutropenia (a crash of white blood cells). That adds up to a compromised immune system (hardly helpful for any latent infection). Rifampycin, meanwhile, is well known to cause gastro-intestinal problems and loss-of appetite (hardly a recipe for maintaining a healthy constitution). Isoniazid (the least problematic of all three drugs) is still known to sometimes cause gastro-intestinal problems. These side-effects may well be deemed acceptable if the latent infection ends up being reliably sterilised, but there’s an increasing likelihood that in many cases (in exactly the patient groups being targeted) it won’t be.

The fact that isoniazid is the least problematic certainly explains its higher use, of course (and also makes it the drug of choice probably for children), but it might equally go some way to explain the apparent higher rates of mono-resistance to it (isoniazid resistance found in over 10% of children tested, let's not forget). And in this respect it’s worth recalling that the Lancet paper particularly identified that children under 15 were at nearly three times greater risk of having latent MDR-TB and were therefore at significantly higher risk of developing reactivated MDR-TB.

So what’s to be done?

It’s a good question - not least because the WHO doesn’t generally recommend using second-line drugs for treating latent TB in MDR-TB contacts, even though their infection is likely to be resistant to any approved treatment for it. So it looks like things are in a bit of a bind.

Well we want to make a cautious suggestion: maybe moxa may help.

(By this we mean a daily administration of small cone moxa smouldered down to the skin on the leg until the patient feels the heat). We make this suggestion based on two hypotheses, and on a slowly growing body of evidence that desperately needs further testing:

1. It may help the host immune system keep the infection in a latent state.

2. It may help any struggling pharmaceutical intervention sterilise the infection by provoking a sympathetic immune response even if the infection is drug-resistant.

Regarding the first suggestion, we’re not the first to have made it. It was actually made in 1929 by Dr Shimetaro Hara in Japan in Fukuoka University’s Medical Journal when he published his results of the use of moxa for treating TB. But we now have an even better reason for making this suggestion because we have recently been given the results of a study that’s been conducted looking at exactly this in North Korea by its Ministry of Public Health (and they've given us permission to share their data).

This study looked at the use of this therapy for close contacts of known infectious TB cases who were already known to be latently infected, monitoring them for re-activated disease. They studied a total of 294 cases, 152 of whom used a daily prophylactic dose of small cone moxa at St36 (an acupoint on the leg with a longstanding reputation for promoting a host immune response) for a 3 month period, with the other 142 using isoniazid (INH) medication (10mg/kg per day) for a period of 6 months. (

The results in terms of incidence of TB disease were compared after a 12-month period.

The results shared with us so far have revealed no difference in incident numbers of new disease between the two groups (2 new active cases emerging in each group making for a 1.31% and 1.41% incidence rate in each group respectively). In other words, moxa appeared to be neither worse nor better in preventing re-activation of disease from latency than isoniazid.

But let’s remember that isoniazid is currently approved and is the most frequently used treatment for latent infectious TB – but also let's not forget that it will be sub-optimal or ineffective for treating MDR-TB or an infection that is isoniazid mono-resistant.

So here’s our first question which relates to isoniazid-resistant TB (including MDR-TB):

Could moxa therapy adjunctive to isoniazid tilt the balance away from TB’s favour by reducing the growing incidence of active MDR-TB disease?

Let’s not forget that isoniazid resistance may be quite common now in many TB endemic countries – at 12% or more now, and that latent MDR-TB is occurring in a rate of at least 1.2% in the latent pool – quite possibly more like 5% at the shaper end of infections entering into it).

To contemplate this question further, let’s quickly remind ourselves of the current state of apparent equilibrium between the two states of drug-susceptible disease. Each year we are still seeing reported estimates of around 10 million new infections (reducing at about 1.6%) and this number is both feeding and being fed by a latent pool of 1.8 billion cases. (We saw in Part 1 of this blog that this number basically hasn't changed in 22 years, so each year the prevalent infectious pandemic must be topping up the latent one with pretty much the same number of new infections).

Here’s a simplistic representation of this in terms of the current relatively stable pipeline of cyclical drug-susceptible disease that is so hard to reduce:-

Seeing this graphically represented makes it obvious that tightening any of the faucets at any point in this cycle could reduce numbers and save lives – and we now think that moxa could help tighten down those faucets. Here's why:-

The data from the earlier Ugandan moxa-TB RCT suggest that moxa could help tighten the upper faucet (please see the 'Results' page for more detailed information on all our research). This is the pipe that feeds the pandemic of re-activated infectious TB. We think it could do so by helping reduce the periods of infection and potentially improve recovery rates which would then restrict the outflow of disease dribbling down the pipe that feeds the pool of latency.

Meanwhile the data from the DPRK suggests it could help tighten the middle faucet by helping reduce new infections emerging from the patent reservoir in the same way that IPT does. It might even tighten it further by helping the isoniazid preventative therapy work better (but we definitely need further investigation to test this). But more importantly, given that the moxa’s effect on disease prevention may be comparable to IPT but works differently, it might be able to do the job when isoniazid can’t.

So could adjunctive moxa therapy alongside isoniazid help?

Our answer to this is very simple: we don’t know. But our answer comes with a no-brainer qualification: “… if there’s a chance that it might do, then we’d be morally derelict if we didn’t do all we can to find out!”

To help explain our answer, here’s another pipeline representation of what we suspect may be happening but is being largely under-reported in terms of drug-resistance. This time our pipeline covers 2014 (using the numbers estimated for that year in the Lancet study and from WHO Reports), but extrapolates them into 2015. It recognises that (unlike the relationship between latent and reactivated drug-susceptible TB which is stable) the same relationship relation to MDR-TB is dynamic and developing, increasing somewhere along the line annually by a factor of 7% (which is what we calculate was roughly suggested by the Lancet paper).

We can see in this image that all aspects of MDR-TB should be assumed to be on the rise (i.e. the number of latent cases, the numbers of re-activated cases, and also the proportion and number of deaths).

It’s worth remind ourselves of the possibilities that we identified earlier about the consequences of continuing treating latent TB the way we are doing knowing full well that there is a rising tide of undiagnosable drug-resistance within it:

We identified that some worrying things are possible.

* A significant number of these treatments are going to fail.

• Moxa might help this number reduce if it promotes an improved immune response to the latent infection, in turn reducing the ongoing cycle of infection.

* Drug-resistances may get stoked by clinically inappropriate and futile treatment.

• Moxa might help reduce this risk, particularly if rates of successful treatment in latency are increased and rates of consequent infectious DR disease are thus reduced.

* Latent TB cases will be treated unnecessarily with strong antibiotics with negative effects on patients

• Moxa use in the DPRK suggest that latent TB cases using moxa had trends of elevated lymphocytes suggesting a strengthened immune system.

* If redundant treatment leaves vulnerable patients more vulnerable, then the probability that their latent infection will ‘break out’ into re-activated infectious disease is more likely than this would have been prior to preventative therapy, and (if it does) this disease will be drug-resistant.

• Moxa could help reduce this probability by bolstering host immunity in the treatment cases reducing incidence of DR disease.

* A substantial cohort of people will die unnecessarily, paradoxically because they were initially targeted for treatment because they were recognised for their vulnerability to this disease.

• Moxa might help possible reduce the impact of this devastating paradox and could do so within the post HLM target period (running up to 2022).

What this all adds up to is this: we need to see these North Korean data put to the sword of science

We’re not frightened of doing this, by the way, but the fact is that we don’t have the resource or expertise to do what’s necessary, and we need help.

We’ve seen enough evidence of efficacy in drug-susceptible TB, MDR-TB and now latent TB to be confident in calling on others to take this work on now, though – either to work with us (which we'll happily do) or to take the task from us. This is no longer an option for those concerned with global public health: we think it's an obligation.