Drug-resistant strains of tuberculosis are more virulent than experts assumed, study finds

By Mark Shwartz

The emergence of drug-resistant strains of tuberculosis throughout the world is a far greater risk to human health than medical experts had assumed, according to Stanford University scientists.

This finding is based on a Stanford-led study of patients infected with mutant strains of the bacterium that causes tuberculosis. The results of the study, published in the journal Science, challenge a fundamental principle of evolutionary medicine and may lead epidemiologists to rethink their strategy for preventing the global spread of this highly contagious respiratory disease, researchers say.

"Until this study, medical dogma had been that when a bacterium develops resistance to a drug, it becomes weaker as a human pathogen," said Stanford epidemiologist Gary K. Schoolnik, co-author of the June 30 Science study. "According to that very rosy scenario, drug-resistant strains should eventually extinguish themselves in the environment, because they can't compete with the original, drug-susceptible organism. But we found the opposite to be true, and that has very ominous implications for the spread of tuberculosis throughout the world."

Tuberculosis is caused by a species of bacteria called Mycobacterium tuberculosis, which can be transmitted through the air when an infected patient coughs or sneezes. If not controlled, the bacterium may attack and destroy the lungs and other parts of the body.

The World Health Organization (WHO) estimates that 2 billion people are infected with latent M. tuberculosis, which usually remains dormant but may begin actively multiplying, especially if the person's immune system weakens. Approximately 15 million people—primarily in Africa, Asia and Eastern Europe—have active tuberculosis disease, including about 14,000 in the United States.

"Worldwide there are roughly 12 million new active cases annually, and of those about 2 million will die every year," said Schoolnik, professor of medicine and of microbiology and immunology at Stanford. "As a global health threat, tuberculosis assumes a significance that is only equaled by two other infectious diseases—malaria and HIV/AIDS."

Health experts recommend that people with active tuberculosis be treated with at least four antibiotics, such as rifampin, over a six-week period. If the patient does not adhere to this strict multi-drug regimen, then the original M. tuberculosis bacterium may evolve into a mutant strain that is resistant to two or more of the preferred antibiotics. Curing multi-drug resistant tuberculosis isn't easy, however, because patients have to be treated with medications that are more toxic and less effective than rifampin and the other first-line antibiotics.

"In some cases, the organism actually develops a resistance to four or five drugs," Schoolnik explained. "Then there's virtually nothing we can treat it with, so we have to resort to surgery to remove the infected lung."

According to WHO, more than 1 million people may be infected with multi-drug resistant tuberculosis, which they can spread to healthy individuals.

"To cure tuberculosis, the treatment has to be intense and aggressive," said Brendan J. M. Bohannan, associate professor of biological sciences at Stanford and co-author of the Science study. "But it can be very difficult to closely monitor patients, especially in developing countries and low-income populations, so the number of multi-drug resistant cases continues to rise. The global implications of that are pretty scary."

Previous laboratory studies of M. tuberculosis showed that drug-resistant strains tend to die off when forced to compete for food with drug-susceptible bacteria. These experiments, conducted decades ago, led to the widely accepted dogma that acquiring drug resistance actually weakens the overall fitness of the mutant bacteria, making them less able to survive head-to-head competition with their mutation-free ancestors.

But the lead authors of the Science study, former Stanford postdoctoral fellow Sebastien Gagneux and graduate student Clara Davis Long, proposed a different experiment. Instead of relying on lab-raised bacteria, they wanted to see what would happen if drug-resistant strains from tuberculosis patients were forced to compete with the drug-susceptible organisms that caused the initial infection.

To find out, the researchers turned to a long-range study of active tuberculosis cases in California. About 15 years ago, an outbreak of the disease swept through sections of San Francisco. During that epidemic, Schoolnik, Science co-author Peter Small (now with the Bill and Melinda Gates Foundation) and their colleagues began collecting and freezing sputum samples from newly diagnosed patients before, during and after treatment with antibiotics.

"Using DNA fingerprinting, it was possible to identify a small number of patients whose M. tuberculosis strain had mutated from drug sensitive to drug resistant within a few months after treatment started," Schoolnik recalled. "So the organism actually evolved and became drug resistant while living in a human host the entire time."

For the Science experiment, Gagneux and Long obtained bacteria samples from 10 San Francisco patients infected with rifampin-resistant tuberculosis. Each drug-resistant strain was placed in a test tube alongside its drug-susceptible ancestor.

"Our prediction would be, based on dogma, that the drug-resistant strain would grow more poorly in the test tube than the drug-sensitive isolate," Schoolnik said. "But we found that the drug-resistant strain was just as robust, and in some cases slightly more so, than its drug-sensitive parent. This suggests that strains that evolve in human beings treated with antibiotics can also acquire compensating mutations that make up for whatever weakness may have been conferred by the drug-resistant mutations."

"This finding is very bad news for the control of multi-drug resistant tuberculosis, which is already recognized as a global threat to public health," Bohannan noted.

In the United States, the Centers for Disease Control reported 128 new multi-drug resistant cases in 2004, a 13 percent increase over the previous year. The problem is especially serious in the states of the former Soviet Union, where multi-drug resistant infections accounted for more than 10 percent of newly diagnosed cases in Latvia, Estonia and the Russian Federation, according to a 2004 WHO report.

"Unfortunately, in a global economy people from these countries are traveling, and airplanes are fantastic places to transmit tuberculosis," Schoolnik explained. "This is not something I want to scare the public with, but if somebody from Eastern Europe with multi-drug resistant tuberculosis gets on an airplane to New York and starts coughing, a fair percentage of people in that aircraft are going to acquire that multi-resistant strain during that flight."

Patients with active multi-drug resistant tuberculosis should be isolated from the general population and treated until they are no longer contagious, he said. "That involves a fairly significant public health infrastructure, however, and in poor countries it's rather hard to accomplish that," he added.

"This study was the melding of public health, clinical medicine and evolutionary biology," Schoolnik said. "To actually demonstrate that evolutionary process occurring in infected humans was a real achievement. It's a successful example of why we encourage interdisciplinary research."

In the future, scientists may discover ways to disrupt the evolutionary process that allows multi-drug resistant tuberculosis to become so virulent in infected people, Bohannan said.

Lead authors Gagneux and Long are postdoctoral fellows at the Institute for Systems Biology and at the Veterans Affairs Palo Alto Health Care System, respectively. Tran Van, research assistant in the Stanford School of Medicine, also co-authored the study. Research was supported by the National Institutes of Health, Wellcome Trust, Swiss National Science Foundation and Novartis Foundation.

Editor Note:

"The Competitive Cost of Antibiotic Resistance in Mycobacterium tuberculosis," by S. Gagneux et al., appeared in the June 30 issue of Science. For a copy, contact the AAAS Office of Public Programs at (202) 326-6440 or scipak@aaas.org. Photos are available at http://newsphotos.stanford.edu (slug: "tbstudy").

-30-