Transformation of TB diagnostics on horizon: special report

Electronic noses that can “smell” tuberculosis (TB), a desktop machine that can read TB DNA and tell what drugs it is sensitive to within minutes, a DNA amplification-based test for TB that requires no sophisticated equipment at all, or a strip test as simple as a home-pregnancy kit that can detect the hard-to-diagnose forms of TB that affect children or people with HIV in a sample of urine — the TB diagnostics pipeline is burgeoning with new technologies which, if they work out in the field and can be produced at a price affordable to the developing world, could transform how TB is managed within the next decade if not in the next several years.

Data on many of these new diagnostic technologies were presented at the 37th World Union of Lung Health conference in early November and discussions continued later that month at a STOP TB Partnership meeting held in Jakarta, Indonesia, on what sort of framework needs to be set up to evaluate and eventually introduce the more useful new labs tests into TB programmes across the globe.

This simple test for TB exposure, which can be conducted at any level of the healthcare system, involves injecting a small amount of TB protein just beneath the skin. If a reaction (red swelling) of a significant size forms at the injection site within a couple days, it is taken as a sign that the person has been exposed to TB and has either latent or active disease. However, vaccination with BCG (the old TB vaccine) can also lead to a reaction at the TST site (as can repeated TST testing), which limits the test's usefulness in vaccinated children or people repeatedly tested because of high risks of exposure (such as healthcare workers).

Furthermore, people with advanced HIV tend to lose the immune responses that cause a skin reaction to TST, making the test of limited usefulness in this population.

Finally, although people with a TST reaction are more likely to develop active TB one day than those who do not, the test can’t really be used to predict if someone will fall ill with TB in the near future.

The laboratory technology that most TB clinics use today has changed little over the last hundred years since it was first developed by the German Dr Robert Koch who first discovered mycobacteria tuberculosis, the organism that causes TB.

There are essentially three tests.

The gold standard for diagnosis of active TB involves trying to grow (or culture) the organism from a specimen, usually performed in a more sophisticated laboratory. In the classical technique, the specimen is cultured in gelatin-like medium that contain nutrients to support the microbe's growth — a process that can take anywhere from a few weeks to a couple months. When anti-TB medications are added to the culture, the organism’s drug susceptibility can be tested (DST) — this, naturally, takes at least a couple more weeks after the culture is available.

Because it takes so long, culturing is used primarily to confirm smear microscopy (in the industrialised countries), to diagnose those cases that smear microscopy misses, and for DST. But given how long it takes, conventional culture results often come too late for people with HIV, who often have rapidly progressive TB.

These three tests have long been established in widespread use across the globe— with some significant refinements that have been introduced primarily in industrialised countries. For example, smear microscopy has been made more sensitive by better quality control and specimen collection strategies, using fluorescent microscopes (which now may be affordable with the use of new low-cost LED lights), and by improvements to how the sputum specimen is processed (see related pieces on smear microscopy). Meanwhile, culturing has been accelerated by the use of liquid media, in which the mycobacterium grows more rapidly, and automation — some of these improvements also facilitate the more rapid DST.

Many of the refinements, such as liquid culturing, are now being introduced to strengthen the TB laboratory capacity in high-burden, resource-limited countries where even incremental improvements could save thousands of lives. Plus the laboratory strengthening effort could serve as a trial run for the introduction of more new diagnostics that may only be a few years off. According to Dr Vinand Nantulya, of the Foundation for Innovative Diagnostics (FIND) “the diagnostics are right around the corner.”

Several of these new diagnostics are quite novel, not only in the technologies employed but because they could potentially revolutionise how and where TB is managed.

In order to accelerate adoption, implementation and access to these new tools as they become available, the Stop TB Partnership has set up a Task Force on Retooling (as in switching to new laboratory tools). At the recent meeting in Jakarta, the Task Force presented a preliminary work plan on how to move forward — (see the meeting website for a variety of materials presented at the meeting).

However, many of the diagnostic being evaluated are very complicated and perhaps not entirely appropriate for the operating conditions in resource-limited settings where they are meant to be used.

“This is very good that we are getting new tools for diagnosis of TB, but… a lot of people now are not talking about whether the new tools are coming, they talking about whether they will be useable in high burden countries,” said one audience member at the STOP TB Working Group on New Diagnostics/FIND meeting in Paris. “It’s becoming synonymous that high burden countries means no modernisation, no change, so that they have to rely on smear microscopy developed in 1882.”

One problem according to Martine Usdin of Médicins Sans Frontiers (MSF), is that among many of the companies working on new TB diagnostics “there seems to be an insufficient knowledge about what those local conditions are, about the human resources that are available, about the number of samples that need to be processed in a day in a different setting, and… how those tests are going to be used? Who's going to be performing the test, how will the tests be analyzed and how will that information be used to inform medical decisions?”

“We've also seen repeatedly how high tech technologies - that are sort of 'forced fit' into a low-tech setting - just doesn't seem to work,” she continued. “We have innumerable examples of CD4 machines that are 'parachuted' into a setting that break down, that can't get serviced in which cold chain can't be assured, reagents aren't available and so the diagnostic process is interrupted.”

But requiring ill patients to hike on dirt roads half the day to get to the nearest peripheral laboratory isn’t acceptable either.

There are essentially two possible approaches to “retooling” laboratories (though both should involve collecting the patient’s biological specimen wherever they present for care — with as few repeat visits or referrals as possible). Point-of-care testing, where the entire test can be performed wherever the patient is at, would be the best. However, for tests that fill an unmet need, a more centralized approach — where the healthcare worker collects a sample for transport to where it can be analyzed with the new test — could be acceptable until a point-of-care test can be developed, provided that the process of sample collection, storage and transport is simple, safe and stable.

The new technologies are at hand, but they are completely different from what has gone before — and which may prove stubbornly entrenched. They will be more expensive — at least to introduce — than simply purchasing microscopes and going on with the inadequate status quo. But they should make up for the investment over time by improving TB control.

The lab tests (or testing approaches) described in the following articles are merely a selection of the developments in the TB diagnostics pipeline. Some of the tests that are described in these linked articles could be revolutionary because they are being specifically designed for ease of use in the most peripheral facilities in high burden settings. Other test are not, but still fill unmet needs — which in some cases could be important enough that it may be necessary to set up specimen transport systems. The final example is of an already commercially available technology that could be years of research away from fulfilling its true potential, but which clearly merits more basic science and clinical research.

This test for active TB involves using a microscope to look for evidence of TB microbes in a stained specimen (usually coughed-up sputum) smeared onto a slide. A basic lab (set up with a microscope and accessories, a regular supply of chemicals and other consumables) as well as at least one trained laboratory technician is needed to be able to conduct this test. This means that smear microscopy tends to be performed at mid-level peripheral healthcare facilities (as opposed to primary healthcare clinics). Thus someone with TB usually has to be referred to a peripheral laboratory for diagnosis (and so the diagnosis comes after someone has had active disease for some time).

Although diagnosis by smear microscopy is the foundation for TB control throughout the world, it doesn’t detect TB in everyone with active disease. In fact, it can miss up to half the cases or more in children or people with HIV. The sensitivity of the test is limited by biology (specimens from TB patients don’t always contain a high enough concentration of the TB organism to detect) as well as the conditions under which the test is performed. As one might imagine, operating conditions can be quite poor in resource-limited settings where the microscopes may be old or poorly maintained and staff may be short or overworked.

MSF’s recent report on the Diagnostics Pipeline

The FIND website: www.finddiagnostics.org

FIND and TDR’s recent report on the Global Market for TB Diagnostics

The Task Force on Retooling’s Guide to New Technologies for TB Control