PET-CTs, AI-interpreted chest X-rays can predict TB

Research led by scientists at the Centre for Infectious Diseases Research in Africa (CIDRI-Africa) at the University of Cape Town (UCT) revealed that advance imaging technologies, like the positron emission tomography (PET) and computed tomography (CT), can detect evidence of asymptomatic tuberculosis (TB) years before symptoms appear or routine tests return positive.

The study, titled: “PET-CT benchmark detection and 5-year progression of asymptomatic tuberculosis: A longitudinal, prospective cohort study”, was published in the Lancet Respiratory Medicine – a leading peer-reviewed medical journal. The study is the largest of its kind to follow high-risk individuals with sensitive imaging over a five-year period.

According to global research, up to one quarter of the world’s population has been infected by the Mycobacterium TB bacteria. However, this figure is based mainly on inference from tests that demonstrate an immune response to the bacteria, and not direct evidence of infection. Interestingly, most people with a positive immune response never develop the disease. And scientists say successfully predicting the number of patients who will progress to TB remains one of the largest gaps in disease prevention.

“Identifying those most likely to develop TB is crucial if we want to prevent transmission and intervene earlier,” said the study’s lead author, Professor Hanif Esmail.

Spotlighting the study

Over the five-year period, researchers followed 250 HIV-negative, asymptomatic individuals from Khayelitsha. Each participant was a household contact of a drug-resistant TB case and underwent an 18-fluorodeoxyglucose PET‑CT – a dual imaging test designed to detect diseases by measuring metabolic activity, particularly glucose usage, as well as an artificial intelligence (AI)-read digital chest x-ray.

“Identifying those most likely to develop TB is crucial if we want to prevent transmission and intervene earlier.”

During the follow-up period, 18 participants were diagnosed and treated for TB. Further, six patients were identified early through enhanced screening at the start of the study, five of whom would have been missed by routine rapid molecular testing. The remaining 12 participants were diagnosed with TB after a median of three years.

Many of the research participants were asymptomatic at the time bacteria was detected in their sputum. This highlights that transmission can occur before routine systems detect disease.

The most sensitive imaging tool

The PET-CT is said to be the most sensitive imaging tool for research and revealed a wide spectrum of lung abnormalities on participants. Those participants whose scans indicated a specific set of abnormalities associated with TB at the start of the study were 28 times more likely to be diagnosed with TB during a follow-up, compared to participants whose scans appeared normal.

“These findings position PET-CT as a powerful research tool for understanding how TB progresses in the body.”

While the research findings indicated that 205 of the 250 participants showed an immune response to the bacterium, those with PET-CT lung abnormalities were at a higher risk of being diagnosed with TB.

“These findings position PET-CT as a powerful research tool for understanding how TB progresses in the body. While the method is too costly for wide-scale public health use, its precision offers valuable insights for clinical research studies to develop improved diagnostics and therapeutics,” said co-author, Associate Professor Anna Coussens.

Promise for mass screening

Finally, scientists also described the performance of AI-interpreted chest X-rays as “more immediately impactful”.

Professor Robert Wilkinson, one of the study’s co-authors, said although AI-interpreted chest X-rays are less sensitive than PET-CTs, the AI readings demonstrated “good alignment” with PET-CT predictions. This, Professor Wilkinson added, suggests significant promise for mass screening.

“AI-read chest X-rays could play a vital role in strengthening TB control strategies through mass-screening efforts,” he said.