(This article appeared on Nature’s news site on 1 September 2010).

Infection with tuberculosis can be diagnosed easily and accurately in less than two hours.

A new test can accurately diagnose tuberculosis (TB) in people in 90 minutes, compared with the six weeks needed for the current standard test.

The Xpert MTB/RIF test, described today in the New England Journal of Medicine¹, identifies TB in 98% of active cases — an improvement of more than 45% on one of the most commonly used current techniques. It also detects whether the TB-causing bacteria are resistant to rifampicin, a first-line drug for TB, in nearly 98% of cases.

“It has the potential to be revolutionary,” says Richard Chaisson, director of the Johns Hopkins Center for Tuberculosis Research in Baltimore, Maryland, who was not involved with the work.

A single test can also diagnose TB in 72% of those infected with both HIV and TB, which the current standard smear cannot do at all. In some parts of Africa, up to 60% of people with HIV also have TB. A second and third round of testing bring the Xpert MTB/RIF’s sensitivity up to 90%.

“It’s huge,” says Giorgio Roscigno, chief executive officer of the Foundation for Innovative New Diagnostics (FIND), a non-profit organization based in Geneva, Switzerland, which conducted the study. “This test has the potential to make the biggest impact ever in TB control.”

The past few years have seen the development of several drugs for TB treatment. But the tools for diagnosis remain antiquated. The most common method worldwide, called sputum microscopy or the smear test, has changed little since it was invented more than 125 years ago, and misses more than half of all cases. Furthermore, the technique neither identifies drug resistance, nor detects TB in those also infected with HIV.

Culturing the TB bacterium (Mycobacterium tuberculosis) allows a much more sensitive test, but it takes weeks of work by trained lab staff to get enough of the slow-growing bacteria.

‘Espresso’ ease

Xpert MTB/RIF, by contrast, is as easy to operate as an espresso machine, says Peter Small, senior programme officer for TB at the Bill & Melinda Gates Foundation in Seattle, Washington, which co-funded development of the test. “And it is blindingly fast compared with other technologies,” he adds.

TB kills roughly 1.8 million people each year, according to the World Health Organization (WHO). In people who have both HIV and TB, the TB bacteria are often resistant to multiple drugs and the infection is quickly fatal.

“It is not unusual for a patient to die before he knows the results of the culture test,” says Roscigno.

Xpert MTB/RIF relies on real-time polymerase chain reaction (PCR) technology, which generates thousands to millions of copies of a piece of DNA, to amplify the DNA of the TB bacterium. The process is mostly automated, reducing contamination of the sample, and does not need much equipment, giving it great potential for use in resource-poor clinics in the developing world. It can be carried out by technicians with minimal training. The test is based on a platform developed by the US government to detect the anthrax bacterium (Bacillus anthracis); FIND began developing it for TB in 2005.

Researchers have been trying to develop TB tests using PCR for more than a decade, but this is the first one that is easy and sensitive enough to have a broad impact, says Roscigno.

Over the next two weeks, the WHO is expected to review evidence on the test’s effectiveness before it decides whether to recommend it to health ministries.

Cautious optimism

But there are reasons for caution. The test has so far been used only in relatively well-equipped labs. It has not yet been trialled in places with a lot of dust or an unreliable power supply, both of which could cause problems with the PCR equipment. Such conditions are common in many of the places worst affected by TB and HIV.

Another key issue is price, which is still being negotiated by FIND. Even with a high-volume discount, the cost per test is likely to be more than the roughly US$8 for multiple conventional tests — smear plus culture, or multiple rounds of smear.

But Chaisson says that the price of the Xpert MTB/RIF test will be comparable to that of viral load tests for HIV, widely used to decide who to put on therapy.

Chaisson says that ever since he first heard about the test at a conference a year ago, he has been eagerly awaiting its arrival. He is trying to get funding from the US National Institutes of Health to use the test to identify TB infections in a study in Malawi on HIV treatment, and has already purchased an Xpert MTB/RIF machine for use with the Tibetan exile community in Dharamsala, India, which he says has “about the lowest-tech clinic I’ve ever worked in”.

“Smear is terrible — we’re going to have to have a better technology and it is going to cost more,” says Chaisson. “We can say that smear is cheap, but given that it misses most cases, who cares how cheap it is?”

References

• 1. Boehme, CC et al. N. Engl. J. Med. doi:10.1056/NEJMoa0907847 (2010).