Clifton Barry’s career as a tuberculosis researcher began in the 1990s, when the disease began to stage a quiet but dramatic resurgence. At the time, the overall death rate from TB was dropping, but strains of the bacteria were emerging that are resistant to the powerful drugs that had kept it in check for decades.
The World Health Organization estimates that nearly half a million people worldwide contract multi-drug resistant tuberculosis in 2013 alone, accounting for 3.5% of new cases and 20% of recurring cases. Granulomas, or masses of tissue, can keep tuberculosis bacilli sequestered from the immune system for months or years before causing full-blown disease. The bacteria can also continue to hide within granulomas even after antibiotic treatment begins, which is one reason why it usually takes at least six months to cure tuberculosis. When patients don’t stick to the long treatment regimens, the bacteria are more likely to develop resistance to antibiotics. The incidence of infection with strains that can’t be killed with first-line antibiotics is rising in parts of Eastern Europe, Asia and Africa.
Most research into tuberculosis involved searching for new antibiotics, but in 2008, Barry received a call from a friend who’d just seen a talk at a meeting of the National Cancer Institute. “She said, ‘You have to meet this guy,’ ” says Barry, who leads the tuberculosis research section at the National Institute of Allergy and Infectious Diseases (NIAID) in Bethesda Maryland.
But the talk Barry’s colleague had recommended was not about tuberculosis. It was delivered by Rakesh Jain, a tumor researcher at Harvard Medical School. Jain had been studying a drug called bevacizumab, known commercially as Avastin, which sops up a blood vessel growth factor called VEGF, and how the drug affects the vasculature around tumors. A lack of oxygen inside tumors stimulates nearby cells to make lots of VEGF—so much that new blood vessels grow uncontrolled. The end result is a twisty, wild, and leaky vascular network that actually does a poor job of delivering blood, oxygen, and chemotherapy drugs. Originally, anti-angiogenesis drugs like bevacizumab were meant to “starve” tumors of their blood supply by preventing new blood vessel growth.
But Jain noticed a wrinkle in that theory. “If you’re going to starve the tumor, how are you going to deliver chemotherapy?” he says. Jain figured that by tapering doses of anti-angiogenesis drugs, he could restore some normalcy to these blood vessel networks and increase delivery of chemotherapy to tumor cells. In one case, at least, it has worked, extending survival in brain cancer patients involved in a clinical study by about six months.
Barry’s friend noticed striking similarities between tumors and spongy balls of tissue called granulomas that form in the lungs of tuberculosis patients. Granulomas are accumulations of immune and fibrotic cells that gather after tuberculosis takes up residence inside immune cells called macrophages. At their cores, many granulomas are full of dead cells and very low on oxygen, both characteristics that are shared with solid tumors. They also seem to promote new blood vessel growth, again, just like tumors do.
But Jain and Barry wanted to know if Jain’s tapering regimen of anti-angiogenesis drugs could work to straighten abnormal blood vessels in granulomas to improve access to antibiotics, just like it had tumors and chemotherapy. In a study published in the Proceedings of the National Academies of Science in February, Jain and Barry demonstrated by treating infected rabbits with bevacizumab, they were able to straighten out the blood vessels around the granulomas, which made the cells inside more accessible to a dye injected into the rabbits’ bloodstream. Barry’s group is now preparing to test this approach in non-human primates to find out whether it could help clear the infection faster when given with antibiotics. “I have learned more than I ever wanted to know about vasculature,” he says.
Barry and Jain’s collaboration is just one of many that illustrate the lengths to which researchers are going to circumvent antibiotic resistance. By bridging the gaps that divide cancer, immunology, and microbiology, researchers may yet find creative and affordable ways to address the growing problem of drug-resistant tuberculosis.
Aiming for New Targets
The loss of effective options against the disease has left many scrambling for new antibiotic drugs and drug regimens. In the meantime, a growing field of research is pulling drugs already approved by the Food and Drug Administration (FDA) off the pharmacy shelf in hopes of changing the infection’s course. Like bevacizumab, many of these drugs are coming from the cancer field. “We should be looking at everything they’ve done in cancer,” Barry says.
Bevacizumab is an example of a host-directed therapy, a treatment that deals with infection by altering human biology rather than just killing off a pathogen. “Host-directed therapies represent a wedding of cancer fields, immunology, and microbiology,” says Daniel Kalman, an immunologist at Emory University in Atlanta. The earliest example of a host-directed therapy for tuberculosis may be sunlight; scientists now know it enhances production of vitamin D, which has several effects on the immune response to the bacteria that are now being tested in clinical trials.
Host-directed therapies are appealing in the tuberculosis field because of how well the bacteria hide from antibiotics. Normally macrophages are supposed to digest the bacteria, but tuberculosis alters the chemistry inside these cells, converting them into cozy little homes. And if the macrophages are like homes, granulomas are like gated communities, often impenetrable to the rest of the immune system. There is great interest in finding host-directed therapies that either help macrophages do their job of digesting the bacteria or control the inflammation that leads to granulomas and lung damage. In November 2014, the National Institutes of Health (NIH) released a request for grant applications to test FDA-approved host-directed therapies in animal models of tuberculosis or in small proof-of-concept human studies. “I would think everybody and his brother is applying,” Kalman says.
Kalman has been working with a drug used for chronic myelogenous leukemia, called imatinib, which is marketed as Gleevec. His lab found that imatinib blocks a human enzyme that tuberculosis bacilli hijack to prevent macrophages from digesting them. As it turns out, a lot of pathogens use this same enzyme, called ABL; Kalman says imatinib has varying degrees of success treating cells infected with E. coli, Ebola, pox viruses, and malaria. In a 2011 study, his group showed that the drug helped clear tuberculosis from the lungs and spleens of infected mice even when given without antibiotics.
Kalman is also exploring a potentially useful side effect of imatinib, which causes the bone marrow of treated mice to make extra immune cells that make their way to the lungs and help fight infection. Kalman says this “emergency response” may normally be suppressed by tuberculosis, and in this way, imatinib helps the immune system get back on its feet.
This second effect of imatinib falls into the category of host-directed therapies that boost the immune system to fight tuberculosis. Not everyone agrees that this is the right direction, at least not for every patient. “It seems like inflammation gets in the way of the [antibiotic] drugs working—at least the drugs that we have now,” says Robert Wallis, an infectious disease specialist and chief scientific officer at the Aurum Institute in Johannesburg, South Africa.
Ten years ago, Wallis was involved in human studies that tested anti-inflammatory drugs called corticosteroids in patients that had both tuberculosis and HIV. This common combination can be very deadly because tuberculosis activates HIV-infected immune cells and causes them to spread more of the virus. Wallis and his colleagues thought corticosteroids would help keep HIV under control during tuberculosis treatment. To their surprise, the steroids helped patients clear the bacteria faster. Unfortunately, the treatment also made their HIV worse and caused side effects too severe to justify continuing the studies.
Still, Wallis is convinced something can be learned from this work. “I’ve been thinking about this now for a decade.” He is now looking for clues in autoimmune diseases like rheumatoid arthritis, psoriasis, and Crohn’s disease, which are often treated with drugs that block a specific inflammatory protein called TNF. He thinks that blocking TNF could be a safe alternative to steroids, and in at least two cases, he’s found that it successfully controlled tuberculosis infection in the lungs or the brain. In 2012, a group at John’s Hopkins reported that blocking TNF greatly reduced the amount of bacteria in the lungs of infected mice, providing further support for the idea.
Paradoxically, autoimmune patients on anti-TNF therapy are at high risk for contracting tuberculosis since TNF mobilizes the immune response to the bacteria. But this is exactly why it might be able to help patients who already have the infection. Wallis says the immune system drives tuberculosis into a sort of hibernation state in which it turns off many pathways that antibiotics target, like DNA replication. Dampening the immune response by blocking TNF might trigger the bacteria to come out of that state make them more vulnerable to antibiotics.
Wallis says he’s interested in the anti-inflammatory route because it has the potential to make antibiotics more effective and reduce lung damage that occurs during the course of the months-long infection. But, he says turning the immune system down can also be dangerous and that some sort of pre-screening would be necessary to identify the patients who could be safely treated with anti-inflammatory drugs.
“Tuberculosis is a variegated disease,” Kalman says. “The challenge really is how to deal with the patient that walks in the door. What stage are they in? What state is their immune system in?” The tuberculosis field, he says, is still in the throes of a debate about what kinds of markers and tests should be used to determine the usefulness of any given host-directed therapy. “The field will have to proceed slowly and carefully in terms of trying to match the drugs to the state of the disease where they feel like they have the best chance,” says Kalman. Otherwise, he says, a drug could fail for all the wrong reasons, “and no one will ever touch it again because it failed.”
“Host-directed therapies are complex. But those of us that work on the immune system and tuberculosis think it’s worth a go,” says Alan Sher, head of immunology in the Laboratory of Parasitic Diseases at the NIAID. Sher points to the cancer field as a place where pre-treatment screening is already routinely done. “The concept is already there,” he says. Last year, Sher’s lab reported in Nature that an asthma drug called zileuton (also known as Zyflo) increased survival rates in tuberculosis-infected mice by rebalancing levels of inflammatory proteins called type I interferons. In this case, patients with high levels of type I interferons may gain the most from treatment with zileuton.
Sher admits that cost is a major hurdle to getting new screening techniques into place in the areas most hard-hit by tuberculosis, which tend to be developing countries. But it’s not impossible, he says, since diagnostics have to be done before antibiotic treatment begins anyway. The real hurdle may be the expense of using cancer or arthritis drugs in these countries. Generic versions of many of these therapies, including bevacizumab, already exist, but even the costs of generics could be prohibitive and several of them must be administered through an IV, adding to the time and expense of treatment.
“TB is largely a disease of poor people,” Wallis says. “To have the largest impact, you’d want to produce things that are readily available, inexpensive, and with little risk.” But he says that approach has not been very successful; in the last 50 years, only one new antibiotic, bedaquiline, has been developed for tuberculosis. To move forward, Wallis says, it must be recognized that every new drug has to start somewhere, even if in its current form it may not be widely available or easy to use. “You do [a study] because it can prove a concept that can stimulate additional research to come up with other ways to achieve the same result,” he says.
That is pretty much where host-directed therapy research for tuberculosis is at the moment. Wallis expects several human studies will test these therapies in the next year or two, but more basic mechanisms of tuberculosis infection are still being identified. It may be a slow start, but the good news is that the list of existing drugs that could target those mechanisms has a lot of room to grow.