Science1 publisher2 min readPublished
Bedaquiline resistance threatens the WHO's six-month TB regimen, Nature warns
Nature warns that resistance is already emerging to the WHO's six-month, bedaquiline-based TB regimen, in a disease that killed 1.23 million people in 2024. Nobody can yet say how common that resistance is, because the surveillance data needed to count it are scarce.
The Scientist · Science desk

What happened
- The regimen the 2022 course replaced took up to 20 months and relied on injected medicines whose side effects included hearing loss.
- According to the Nature piece, researchers still have only a limited understanding of how and why resistance to the new treatment arises.
- The piece argues a regimen that works at population level can still fail if nobody knows which drug combinations, bacterial mutations and treatment conditions let resistance emerge.
- It asks that TB research budgets cover surveillance, genomics, diagnostics, patient-outcome evidence and implementation research alongside drug discovery.
Compiled by The ScientistSomething wrong?How this is made
Why it matters
- constraint Health programmes cannot yet tell how fast bedaquiline resistance is spreading, so any decision to change treatment course has to be made without trend data.
- cost If resistance forces a fallback, patients face courses up to 14 months longer with injected drugs, and health systems pay for more complex, more expensive care.
- exposure Under trial-and-error prescribing, a patient whose bacteria already evade the chosen drugs can pass on resistant strains before anyone sees that the treatment has failed.
- decision TB funders now have to decide whether surveillance and sequencing in high-burden countries get the same backing as the next new drug.
The warning rests on an observation with no denominator attached. Resistance "is being seen," the piece says [4]. By its own account, though, surveillance and drug-susceptibility testing are too sparse to measure how common drug-resistant TB is, and that includes resistance to bedaquiline-based regimens [7]. Without those data, the piece says, trends and patterns of resistance are hard to spot [7].
The biology is a reason to take the signal seriously anyway. The TB bacterium can evolve quickly to evade a new drug it is exposed to [13]. The piece describes the result as a clock: each new medicine "starts a ticking clock in the race against drug-resistant strains" [16].
Part of the problem is in the clinic. The piece calls TB treatment "a stab in the dark" [14]. Clinicians try a drug combination and wait to see whether it works for that patient, and resistance can spread undetected before they realize the treatment has failed [10]. The fix it proposes is genome sequencing that links the bacterium's genetic make-up to its response to specific drugs and to patient outcomes, with the technology made accessible in the countries where TB burden is highest [9].
What is at risk is concrete. The 2022 course is six months of pills [1], up to 14 months shorter than the injectable regimen it replaced [1]. The piece also calls it more effective, cheaper and faster to implement [2]. If resistance spreads, treatment moves back towards longer, more complex and more expensive regimens [4].
The policy request is about how research money gets divided. "Research into how existing drugs are used, monitored, optimized and preserved deserves as much support as drug development does," the piece says [15]. It wants that evidence to come "as part of the innovation process, not after" [12]. In my view the missing prevalence figure strengthens the case, since counting resistant infections is the first job of the surveillance the piece is asking for [7]. The piece does not say what that surveillance would cost or who would pay for it.
What to watch
- Published prevalence figures for bedaquiline resistance from national surveillance or drug-susceptibility testing programmes.
- Genomic studies tying specific bacterial mutations to failure of bedaquiline-based treatment in patients.
- Whether funders commit money to genome sequencing capacity in high-burden countries as part of TB drug programmes.