A blood test for TB? Well, maybe this time
A blood test for TB? Well, maybe this time
New crop of wannabes hold promise
Mention the words "serologic TB test," and you’ll see plenty of TB folk roll their eyes, as to say, "Yeah, right," says Mark Perkins, MD, manager of diagnostics, research, and development for in the tropical disease research program at the World Health Organization (WHO) in New York City.
That’s because ever since a French physician named Sarturnin Arloing offered the first such test in 1911, hundreds of the would-be assays have tried — but failed miserably — to take advantage of the fact that people with active TB have detectable antibodies in their bloodstream.
The contenders are as seductive-sounding as you can imagine, offering inexpensive, fast ways to see whether someone has TB or not. But so far, as many as 500 of the tests have failed, Perkins say. Where they generally fall down is specificity — meaning that though the assays successfully pick out those with active disease, they tend to deliver too many false-positive verdicts to make them of any use.
Thanks to recent technological advances, however, that long losing streak may be about change. In the last two decades, major advances have been made in the way reagents are designed, Perkins says. Recombinant antigen packages used by the new crop of tests now are "quite clean and pure," compared with those of predecessors, he adds. "My hunch is that even though of the new crop will be as good as we hope, we might — just might — find one this time that’s still pretty good."
Developed with high-tech knowledge and tools, these tests are designed for use in low-tech environments. Such inexpensive tests would be a blessing to poor countries; and even in rich countries, it could be a boon in certain arenas, such as pediatric TB, where making a definitive diagnosis is still fraught with difficulty. Plans now call for Perkins’ division at WHO to evaluate the current crop of a dozen or so contenders, a process that will be accomplished by letting the assays test their mettle against some of the 15,000-plus isolates now sitting ready in WHO’s new specimen bank.
American Bionostica Inc., a biotech company based in Wenonah, NJ, has developed one of the new tests — dubbed, appropriately, the Rapid TB Test. Like virtually all the others vying for a share of the U.S. diagnostics market, this design is a lateral-flow, immunochromatographic assay — or simply put, a test strip that shows a positive result by the appearance of a colored band.
So far, so good, say data
The American Bionostica test features what the company literature calls "a rationally designed combination of at least three antigens," which is intended to bind with TB-specific antibodies generated by the immune system of someone with active TB. The test takes just 15 minutes to perform, requires 10 µ of whole blood or serum, and will likely be priced at between $2 and $4 per test, says Rick Thompson, president of ABI. Like the other candidates, Thompson’s assay discriminates between latent and active TB (since latently infected people don’t produce antibodies). The company also says it can discriminate between active TB and BCG-vaccinated persons, and between those infected with M. TB and one of the "atypical" mycobacterium not found in the M.TB complex.
So far, American Bionostica’s assay has been tested on more than 500 TB patients, principally in Africa, Eastern Europe, and India, reports Thompson. Data from preliminary trials show sensitivity and specificity of better than 80%, he adds. Forthcoming results of an ongoing trial on 400 Botswana TB patients, many of whom are HIV-infected, should provide more data on how well the test works in HIV-infected patients.
Whipping up bucket of E. coli
These days, what typically happens when recombinants are made for tests similar to that one is that TB genes are snipped out, then grown in a rapidly dividing organism, such as E. coli, says Perkins. The E. coli is spun out of the slurry, leaving, one hopes, only TB antigens left behind; the final step is to get rid of cross-reacting E. coli antigens in the brew. "If you purify it well enough, you should wind up with a single antigen specific to TB," he adds.
Along with false-positive problems, candidates so far have tended to do poorly with TB patients who also have severe HIV disease, since such patients tend to make fewer antibodies. That’s a big limitation, notes Perkins, since it would mean the test wouldn’t work well wherever there’s much HIV coinfection.
More generally, the tests are simply hard to evaluate properly, he says. Since diagnostic devices aren’t as closely regulated as vaccine and drug candidates, companies often never do manage to generate adequate amounts of data. And even if the tests scheduled to happen soon at WHO turn up a truly viable candidate, TB controllers still will have to decide whether they’re willing to invest in a tool that won’t ever function as a stand-alone assay.
"With these serologic tests, you’re always trying to find a balance between sensitivity and specificity," Perkins points out. "Frustratingly, the more specific they are, the less sensitive they tend to be, and vice versa. Everyone’s hoping for a technological solution to the problem; but in fact, we’re probably going to have to find some ground to accommodate the imperfections of this particular technology."
In other words, Perkins (or someone else) may eventually find a test that works fairly well under some conditions, but not others; but that limitation doesn’t mean the test won’t be useful — it just means it won’t be perfect. A test that’s extremely sensitive but not very specific, for example, could be used to tell whether a patient is negative and doesn’t have the disease; positive results, on the other hand, would have to be confirmed with a backup test. Conversely, a test that’s extremely specific could be used to triage TB suspects; if such a test reads positive, it’s clear that patent needs to be put into isolation.
Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.