New tuberculosis biomarker may help forecast length of treatment

Written by Campbell Brooks

Researchers from Germany and Romania have identified an RNA transcriptome based on 22 genes, the presence of which in patient blood samples may help predict the length of treatment for tuberculosis (TB) patients. They suggest utilization of these biomarkers may reduce treatment times for TB patients, especially those with multidrug resistant TB.

One of the most challenging aspects of TB treatment is ascertaining when to end the course of antibiotics as the absence of bacteria in sputum samples may not reliably signal remission from the disease. This can lead to patients having to undergo very long courses of treatment, which can necessitate taking four antibiotics every day for over 18 months. In addition, the length of treatment can vary greatly depending on the patient’s individual immune system, the virulence of the pathogen and the drug’s availability.

“We urgently need a biomarker that enables the implementation of an individualized treatment duration,”

Stated Christoph Lange, Clinical Director at the Research Center Borstel (Germany) and director of the study, which was conducted at the German Center for Infection Research (DZIF; Braunschweig, Germany). Host RNA has previously been identified as a potential TB biomarker due to rapid changes in patients’ expression profiles during treatment for TB.

You may also be interested in:

To this end, the authors of a recent article, soon to be published in the European Respiratory Journal, identified an RNA signature based on studying five cohorts of patients with confirmed TB in Germany and Romania. By employing a machine-learning algorithm that analyzed the RNA data, they were able to hypothesize end-of-treatment timepoints for both drug-susceptible and multidrug-resistant TB. These timepoints were then validated by comparing them to clinical and mycobacterial data such as patient radiology, chest X-rays and cell cultures.

From these data, the authors estimate that some patients with multidrug-resistant TB may be able to cease treatment hundreds of days earlier than current Tuberculosis Network European Trials Group (TBNET; Borstel, Germany) guidelines suggest. However, this will depend on a patient’s individual RNA signature and the prognosis will be determined on a case-by-case basis. Lange stated:

“The individualization of the treatment duration is an important milestone on the road to precision medicine for tuberculosis,”

To confirm that monitoring of these biomarkers can safely justify an earlier end to treatment, the next stage of the research will involve a prospective study at the DZIF. A control group of patients with TB will continue their treatment in accordance with TBNET guidelines, while a study group will cease treatment according to an assessment of their RNA data.

The authors are confident that patients in the study group will be able to end their treatment before those in the control group. Lange remarked:

“…hopefully, it will then be possible for patients with multidrug resistant tuberculosis to save about one-third of treatment on average.”

Sources: Heyckendorf J, Marwitz S, Reimann R et al. Prediction of anti-tuberculosis treatment duration based on a 22-gene transcriptomic model. Eur. Respir. J., doi:10.1183/13993003.03492-2020 (2021) (Epub ahead of print);