Lung scans predict best meds to treat TB

The image on the left shows "hot spots" of infection in a patient’s lungs before treatment. The one on the right shows disease improvement after six months of taking the drug linezolid. (Credit: University of Pittsburgh)

Sophisticated lung scans can correctly predict which experimental drug will be the most successful in treating people with tuberculosis.

In 2012, an estimated 8.6 million people in the world contracted TB. The first-line treatment is to take four different drugs for six to eight months to get a durable cure.


Patients who aren’t cured of the infection—about 500,000 annually—can develop drug-resistant TB, and then have to take as many as six drugs for two years, says senior investigator JoAnne L. Flynn, professor of microbiology and molecular genetics at the University of Pittsburgh.

“Some of those people don’t get cured, either, and develop what we call extensively drug-resistant, or XDR, TB, which has a very poor prognosis,” she says.

“Our challenge is to find more effective treatments that work in a shorter time period, but the standard preclinical models for testing new drugs have occasionally led to contradictory results when they are evaluated in human trials.”

In previous research, Flynn’s colleagues at the National Institutes of Health found that the drug linezolid effectively treated XDR-TB patients who had not improved with conventional treatment, even though mouse studies suggested it would have no impact on the disease.

TB hot spots

To further examine the effects of linezolid and another drug of the same class, researchers performed PET/CT scans in TB-infected humans and macaques, which also get lesions known as granulomas in the lungs.

In a PET scan, a tiny amount of a radioactive probe is injected into the blood that gets picked up by metabolically active cells, leaving a “hot spot” on the image.

Humans and macaques had very similar disease profiles, and both groups had hot spots of TB in the lungs that in most cases improved after drug treatment.

CT scans, which show anatomical detail of the lungs, also indicated post-treatment improvement. One patient had a hot spot that got worse, and further testing revealed his TB strain was resistant to linezolid.

The findings show that a macaque model and PET scanning can better predict which drugs are likely to be effective in clinical trials. That could help get new treatments to patients faster, Flynn says. The scans also could be useful as a way of confirming drug resistance, but aren’t likely to be implemented routinely.

“We plan to use this PET scanning strategy to determine why some lesions don’t respond to certain drugs, and to test candidate anti-TB agents,” she says. “This might give us a way of tailoring treatment to individuals.”

Funding for this study was provided by the National Institute of Allergy and Infectious Diseases and the National Cancer Institute; the Ministry of Health and Welfare, Republic of Korea; and the Bill and Melinda Gates Foundation.

Source: University of Pittsburgh