Promising Developments in the War Against TB
By Deborah Borfitz
January 18, 2020 | The latest Global Tuberculosis Report, released by the World Health Organization (WHO) in October 2019, contains both good and bad news. On the bright side, thanks largely to improved detection and diagnosis, more people than ever before received life-saving treatment for TB and there were fewer deaths. The number of new cases has also been on a steady decline.
On the other hand, about 10 million people worldwide still contract TB, with the highest burden falling on eight countries—Bangladesh, China, India, Indonesia, Nigeria, Pakistan, the Philippines and South Africa—where prevention and treatment services often aren’t readily available. Drug-resistant TB is also hampering efforts to end the epidemic.
Priority needs include a new vaccine or effective preventive drug treatment, rapid point-of-care diagnostic tests and safer, simpler, and shorter drug treatment regimens, according to the WHO report. Matteo Zignol, unit head of prevention, care and innovation for the WHO Global TB Programme, says a vaccine under development by GlaxoSmithKline (GSK) came as particularly welcome news last year. “We haven’t had very positive news [on this front] for the past five years.”
Among other notable developments in the war against TB is a blood-based triage test for active disease that is getting close to meeting the WHO-recommended performance guidelines of 90% sensitivity and 70% specificity, as reported in Science Translational Medicine. “The approach taken to create this triage test—first identifying multiplexed biomarkers that provide a highly specific signature for a disease and then designing an assay to achieve endpoints that have been established by clinical need—represents a paradigm shift in the way that diagnostic tests are developed,” according to the Wyss Institute Founding Director Donald Ingber.
Meanwhile, University of Pittsburgh researcher Graham Hatfull is in discussions with colleagues in India and South Africa about the possibility of tapping viruses in his crowdsourced phage library to help eradicate the causative organism, Mycobacterium tuberculosis (M. tuberculosis or MTB). “TB is a big problem that warrants multiple strategies for thinking about how to fight back against it,” says the Howard Hughes Medical Institute (HHMI) professor.
The quest for better diagnostics continues. The Foundation for Innovation in New Diagnostics (Switzerland) has 16 projects in its R&D pipeline related to TB, focused on the critical unmet need for a user-friendly, low-cost, non-sputum-based rapid test for diagnosing active TB and rapid drug-resistance tests that enable treatment regimens to be tailored to individuals. Cepheid’s GeneXpert MTB/RIF, the most commonly used rapid molecular diagnostic method, can’t identify all cases of drug-resistant TB or the best drugs to use for treatment so patients remain at risk of being over- or undertreated.
GSK’s vaccine reportedly has overall efficacy of only 50%, but that bests previous attempts and speaks to the complexity of developing a vaccine, says Zignol. Unlike other infectious diseases, natural exposure to TB does not induce enough protective immunity and the body’s response to the M. tuberculosis is poorly understood. Researchers aren’t even sure which animal models best reflect human disease.
Participants in the GSK study were HIV-negative people with latent TB in Africa who were given two shots one month apart. One big unknown is who would benefit most from the post-exposure vaccine, if not those with a small but known risk of developing the disease based on a positive TB test, says Zignol. The next step is a phase III study involving more countries and diverse populations, including people living with HIV who account for 10% of TB cases globally as well as populations at higher risk by virtue of being undernourished or having diabetes.
It’s likely that multiple vaccines will be needed to target different groups, Zignol continues. The only currently licensed vaccine, BCG, is quite effective against the most severe forms of TB (e.g., TB meningitis in children) but much less effective in preventing the more common pulmonary variety in adults.
Researchers at the Vaccine Research Center of the National Institute of Allergy and Infectious Diseases recently discovered that changing the dose and route of administration of the BCG vaccine from intradermal to intravenous (IV) greatly increases its ability to protect rhesus macaques from infection following exposure to M. tuberculosis. The findings support investigation of IV BCG administration in clinical trials to determine whether this route improves its effectiveness in teens and adults.
A TB triage test is sorely needed because the gold standard diagnostic methods—the sputum smear test or the DNA-based GeneXpert test—have limitations, says David R. Walt, professor at the Wyss Institute for Biologically Inspired Engineering at Harvard University and a pathology professor at Brigham and Women’s Hospital and Harvard Medical School.
Many TB-infected individuals, such as HIV-positive patients and children, have a hard time generating a sputum sample for the more common diagnostic method in countries with less developed laboratory capacities, notes Walt. The GeneXpert test is also quite expensive as a first-line test because results are often negative for TB. “Therefore, a rapid, inexpensive triage test is in great demand to identify patients who require further testing.”
Close to 30% of TB cases go undiagnosed each year, largely because people lack access to the GeneXpert test, says Rushdy Ahmad, former research scientist at the Broad Institute of Harvard and MIT who is now president of True North Bio, a medical device manufacturing and distribution company. “We want to help reduce the number of missing cases to zero, if possible.” When untreated, one person with active TB can infect five to 15 additional people through close contact within the course of a year, perpetuating the epidemic.
Remote, resource-limited regions of the world also lack access to genetic experts and basic healthcare infrastructure. The triage test, which Ahmad and Walt are helping to develop, is designed to make a rapid decision about which patients with TB-like symptoms need further diagnostic confirmation and which ones can safely be sent home.
The Road to Now
When Ahmad began his research at the Broad Institute in 2006, the goal was to come up with molecular signatures of disease to turn into low-cost screening tests in resource-poor settings. The central difficulty was identifying biomarkers of interest from the approximately 20,000 proteins in circulation when only a few hundred of them had been identified by mass spectrometry—a figure that has since mushroomed to roughly 4,000, he notes.
Proteomic scientists eventually figured out how to deplete the abundant but less interesting proteins from the blood, such as albumin, to reduce the signal-to-noise ratio when analyzing samples, Ahmad adds. Researchers could then begin detecting potential biomarker candidates at levels needed to achieve the sensitivity and specificity required by clinicians.
It took about five years to discover a blood-based signature of TB, says Ahmad, which is when he started actively working on the triage test. The rule-out TB test is currently at 86% sensitivity and 69% specificity, percentage points away from the WHO goal, and results are not complicated by HIV infection status.
As Walt explains it, a preselected inflammatory protein panel was run on the Myriad RBM platform and the concentration levels of 47 proteins were measured in 387 blood samples taken from people with and without active TB in Tanzania and the Philippines. The protein-level data were then analyzed using a machine learning algorithm that down-selected to the most informative biomarkers for the triage test—interleukin (IL)-6, IL-8, IL-18 and vascular endothelial growth factor.
Walt’s lab had previously developed an ultrasensitive immunoassay called Simoa (licensed to Quanterix, which Walt founded) that enables the triage test to detect very low concentrations of the four proteins in as little as a drop of blood from a fingerprick, says Walt. The algorithm was independently validated on a different set of 317 samples taken from patients in Vietnam, South Africa, and Peru. At the same time, antibody to the TB antigen Ag85B was added to the biomarker panel to get the test to its current diagnostic performance.
It may take a few additional or alternative markers for the triage test to meet the 90%/70% WHO criteria for sensitivity and specificity, Walt says. Knowing patients’ HIV status might also help improve the model, he adds.
The test’s specificity, and the fact it can be completed in under an hour, gets it closer to the goal of being a point-of-care test in rural areas of developing countries that can’t support a well-outfitted clinic, where power may not be reliable and refrigeration is unavailable for reagents, Walt says. The Wyss Institute continues to work on making the test faster and cheaper ($2 versus $10 for the setup used in the study) by developing a simpler, battery-operated device for single molecule digital counting with a sample processing speed of tens rather than hundreds per afternoon.
But even in its present form, a triage test has the potential to address a significant percentage of the population in economically poor regions of the world where TB is endemic, notes Walt. Limited resources can be reserved for those identified as needing diagnostic follow-up using more conventional approaches.
The test covers the half dozen strains of TB but has not yet been studied in India and China, which account for more than half of the overall disease burden, Ahmad says. Talks are now underway with representatives in both countries.
Getting the necessary permissions to conduct a trial in India is difficult and China has in-country requirements making trials more complicated to initiate, says Ahmad. But if he and Walt are successful in bringing a TB triage test to the field, it could potentially save hundreds of thousands of lives every year.
Hatfull has been working with bacteriophages (aka phages)—viruses that infect bacteria—since he came to Pittsburgh in 1988, but the rationale until recently was simply to better understand their biology and evolution, he says. The focus has always been on phages that infect the mycobacterium, the germ species that causes TB.
In 2008, Hatfull helped create the HHMI’s Science Education Alliance-Phage Hunters Advancing Genomics and Evolutionary Science (SEA-PHAGES) program that has been the chief contributor to an ever-enlarging phage library at the University of Pittsburgh. Students enrolled in the two-semester, discovery-based undergraduate research course scout for new phages in nature and then isolate, purify, name, and characterize them, including their genetic properties, before they get added to the collection.
The SEA-PHAGES program started with 12 participating institutions, which has since mushroomed to 150 schools involving 5,000 undergraduates annually, says Hatfull. At last count, a total of 17,630 phages (many isolated on a non-pathogenic strain of M. smegmatis) were being stored in a freezer, and they’re all available for exploration in the Actinobacteriophage Database (PhagesDB.org). Genome sequences are available for 3,166 of the phages.
The idea of using bacteriophages for TB control is an old idea, says Hatfull, noting descriptions of experiments using phages to treat infected mice first appeared in the Russian literature decades ago. One of two approaches continue to be explored, the first to use phages prophylactically to interfere with transmission and acquisition of the disease from someone who is already infected.
The appeal of the preventive approach is that exposed individuals could theoretically inhale many phages with good odds of knocking out the causative organism among the relatively small number of bacteria typically involved in TB transmission, says Hatfull. He collaborated on a recent paper (DOI: 10.1128/AAC.00871-19) in Experimental Therapeutics that attempted to test the viability of that idea, and pretreatment was found to significantly decrease M. tuberculosis burden in mouse lungs.
Data analysis proved tricky because it involved quantifying phage particles delivered to mice and bacteria administered through an aerosolized system, Hatfull notes. “But it was an encouraging first step.”
The second tactic is to use phages directly to treat patients with TB, the chief obstacle being getting phages to bacteria that are inside human lung cells and encased in granuloma-like structures. That would seem to make phage therapy a somewhat unlikely, if untested, solution, says Hatfull. But there are other ways it might prove effective—for example, in combination with antibiotics to help reduce the incidence of resistance to new TB drugs emerging on the market.
A paper (DOI:10.1038/s41591-019-0437-z) Hatfull co-authored last May in Nature Medicine reported on the successful phage treatment of a 15-year-old patient named Isabelle infected with M. abscessus, a TB-related organism, who had exhausted all other therapeutic options. It was the first such demonstration of phages being used therapeutically for a mycobacterial infection, he says, and “raises the possibility that perhaps bacteriophages could be useful for treating TB.”
It took Hatfull and his team three months to find three phages in the collection at the University of Pittsburgh that could kill the mycobacteria isolated from Isabelle's wounds and sputum. Two of the phages had to be genetically altered to activate their lethality, he says, and the third was a natural phage named Muddy that was found in decomposing vegetable matter by a student attending one of his educational workshops in South Africa. After six weeks of being infused with the phage cocktail twice daily, Isabelle’s infection practically disappeared.
Compassionate use under the FDA’s emergency investigational new drug (eIND) application process could similarly be used to explore if phages are useful for TB, Hatfull continues. The alternative is to conduct a clinical trial and treat cohorts of patients with phages and see who gets better. “We’re interested in both of those strategies, but they are extremely different.”
Window of Opportunity
The compassionate use route would require finding TB patients who a have similar kind of desperate condition and then convincing regulatory authorities that they have no other options left, Hatfull says. Identifying those patients can be difficult—although partnering with researchers in India may help locate those in that narrow clinical spectrum.
Proving that all available drugs have been not only prescribed but ingested would be another hurdle, continues Hatfull, as would ensuring would-be participants are stable enough to have a chance at recovery at all. “We’ve certainly been in discussions with colleagues as to whether we can identify such patients in India, or anywhere else in the world, who we might be able to help with phage intervention.”
The idea of using phages in a clinical trial is “quite attractive” because it’s likely that only a relatively small number of phages would be needed in a treatment cocktail since TB isolates tend to be like one another, says Hatfull. With most other infections, including M. abscessus, the strains of bacterial pathogens almost always vary slightly person to person—including whether they are vulnerable to attack by certain phages.
Hatfull says he plans to exploit this window of opportunity in hopes of identifying phages that could be used to broadly treat patients without prior knowledge of whether their TB strain is affected by the phages. “We think we know which phages to use in a clinical trial… and chances are good that clinical isolates [of TB patents] would be affected by one or more of them used in combination.”
His group is now in serious discussions with collaborators in South Africa about how to move that forward, possibly sometime later this year, despite the scarcity of animal data. Admittedly, it’s a “relatively high risk” proposition that will require significant financial backing to become a reality, but Hatfull says he’s eager to get started.