Shorter TB treatment possible for most using Xpert MTB/RIF® and X-ray

Tuberculosis remains a disease that typically requires months of treatment, and while recent trials suggest many people can be cured with shorter courses, a clear challenge is identifying who needs the standard longer therapy. Daniel Grint and colleagues revisited data from a treatment-shortening study to see whether simple tests at the start of care could predict which patients might safely receive only four months of treatment and which would benefit from the usual longer course. The team used results already available at diagnosis—Xpert MTB/RIF® test measurements and chest X-ray grading—to classify patients into a limited or extensive disease profile. Their goal was practical: find a way, using widely available tools, to separate people into groups with different relapse risks so that shorter regimens could be targeted to those likely to do well, and longer treatment reserved for those at higher risk of relapse. This approach aims to support future trials and real-world changes that tailor treatment length to disease severity rather than applying one duration to everyone.

The researchers analysed data from the RIFASHORT TB treatment-shortening non-inferiority trial and re-examined the trial’s primary outcome using the protocol-defined non-inferiority margin of eight percentage points. They tested combinations of the Xpert MTB/RIF® semiquantitative bacterial burden measure and chest X-ray disease involvement grading to define a TB phenotype that best separated people who relapsed after treatment from those who did not. The strongest discrimination came from combining a high semiquantitative bacterial burden on Xpert MTB/RIF® with extensive TB disease on chest X-ray. This “extensive disease” phenotype represented about one quarter of the trial participants but accounted for more than half of post-treatment relapses (13 of 23 relapses). For participants with a limited TB disease phenotype—defined as

These findings suggest a practical path toward personalised TB treatment durations using tests already in routine use. If Xpert MTB/RIF® semiquantitative results and a chest X-ray at treatment start can reliably classify most people as having limited disease, then a four-month, rifampicin-based regimen could be offered safely to that majority, reducing treatment time for many patients. At the same time, identifying the smaller group with an extensive disease phenotype points to those who may still need longer therapy to avoid relapse. The authors conclude that because three-quarters of participants in the RIFASHORT trial fit the limited profile and did well with the shorter 4-month 1200mg rifampicin-containing regimen, a phase III randomised trial evaluating disease-stratified rifampicin-based treatment is warranted. Wider use of these accessible diagnostic measures could make tailored TB treatment more feasible in practice and guide implementation if larger trials confirm these results.