Credit: Denis Pobytov / DigitalVision Vectors / Getty
Bested only by COVID-19 between 2020 and 2023, tuberculosis (TB) routinely takes the top seat as the world’s leading cause of death by a single infectious disease. Mycobacterium tuberculosis infects about 25% of the global population, with 5–10% converting to symptomatic disease. In 2024, that led to 1.23 million deaths the world over. Associated with, and perpetuated by, poverty, the TB disease burden falls heavily on lower- and middle-income communities, a dynamic that led to decades of underinvestment. Today, a well-coordinated and diverse research and development community that encompasses the public and private sector worldwide, along with a well-defined set of shared goals, is driving the largest TB drug and vaccine pipeline in history.
There are currently 29 candidates in clinical development, including 18 next-generation or new chemical entities (NCEs) and 8 repurposed drugs (Table 1). Three approved drugs feature in trials aimed at shortening treatment of the disease. “All in all, a very promising full pipeline,” says Bern-Thomas Nyang’wa, medical director of the Médecins Sans Frontières Operational Centre Amsterdam. “And of course there are vaccines in phases 2b and 3 that could catalyze success in global TB control.” Of the 16 vaccine candidates, a phase 3 trial of M72/AS01E is leading the pack to introduce the first new vaccine class in 100 years.
Shorter, safer and less toxic
Building on recent advances, drug repurposing and many NCEs, the current TB drug pipeline has the capacity to deliver on three key TB development goals: shortening the duration of TB treatment, reducing its toxicity, and ensuring it remains affordable. “For a disease like TB, where there’s not as big of a commercial draw,” explains Charles Wells, head of therapeutics development at the Bill & Melinda Gates Medical Research Institute (Gates MRI). “It’s rare to have this large of a pipeline.”
Wells explains that until the 2000s, there had not been any new drugs for TB that had not been developed in the 1970s: “Then and now we are using the same standard of care that evolved in that era.” Specifically, the World Health Organization (WHO) recommendation for first-line TB treatment includes a panel of four antibiotics — isoniazid, rifampicin, pyrazinamide and ethambutol (HRZE) — over 6 months. This works well for drug-sensitive TB, still the type of TB for the overwhelming number of cases worldwide, but not for the drug-resistant varieties that have emerged and are typically more lethal.
Unlike other pathogens, TB bacteria grows slowly and hides in multiple biological niches: intracellular components, lung lesions, and hard‑to‑reach cavities. No single drug can reach all these niches. Multidrug regimens are needed to prevent and reduce resistance and to help clear the diverse bacterial populations. But patients struggle to finish long courses of antibiotics. Resistance to isoniazid and rifampicin first emerged in the 1960s, with multidrug-resistant TB (MDR-TB) reported a decade later. As the number of TB deaths continued to rise and approached historic heights, in 1993 the WHO declared TB a public health emergency.
“We develop regimens,” says Mel Spigelman, CEO of the Global Alliance for TB Drug Development (TB Alliance), which leads with the lion’s share of the TB clinical pipeline. The goal now is not just to add new agents to old combinations but to build entirely new regimens by pairing first‑in‑class molecules with improved versions of established drug classes. These combinations are explicitly designed to hit bacterial populations in parallel, allowing treatment courses to shrink from many months to, eventually, just weeks.
According to Nyang’wa, today’s pipeline builds on two new antibiotic classes: Pfizer’s linezolid, introduced in 2000, the first oxazolidinone used for TB; and bedaquiline, a diarylquinoline that became the first new TB drug class in 40 years. Both drug classes remain central to the treatment of drug-resistant TB, but they come with unwelcome toxicities: neuropathy in the case of linezolid, and cardiac complications for bedaquiline. Additionally, in 2019, the TB Alliance updated the regimen for treating MDR-TB when it introduced pretomanib, also an NCE, as part of a combination regimen called BPaL (bedaquiline, pretomanib, linezolid).
Six months of BPaL plus moxifloxacine (BPaLM), a repurposed antibiotic, is now the WHO-recommended treatment for MDR-TB, while HRZE remains the standard of care for drug-sensitive TB. A half-year of treatment with BPaLM is a vast improvement over the highly toxic 18-month regimen that preceded it. However, the TB drug-development community is collectively pursuing 3-month treatments for both drug-sensitive TB and drug-resistant TB — an ambitious goal that the growing number of NCEs in the pipeline may finally make realistic.
Keeping TB drugs and vaccines affordable and accessible relies on shared licensing, open collaboration and public–private partnerships that are ubiquitous in TB drug development today. Pretomanib now costs less than US $1 per day, a 50% drop from its original price in 2019, enabled by the TB Alliance’s licensing production to multiple manufacturers, and by the United Nations making it widely available to low- and middle-income countries through its Stop TB Partnership Global Drug Facility.
Shorter regimens also keep overall costs down. The next‑generation oxazolidinones (such as sutezolid and delpazolid) and improved diarylquinolines (such as sorfequiline) have a key role in reducing treatment times. They aim to preserve the potency of linezolid and bedaquiline, respectively, while reducing their major toxicities. Nyang’wa adds “We also have candidates from new drug classes, like quabodepistat, which could be considered for novel drug regimens.” Quabodepistat has a completely novel mechanism of action and has high potency at low doses.
At the 2025 Union World Conference on Lung Health, Rod Dawson, managing director of the Cape Town University Lung Institute, presented early results from the PAN‑TB program (Projects to Accelerate New Treatments for Tuberculosis). Dawson revealed that combinations built around bedaquiline, quabodepistat and sutezolid, paired with either pretomanid or delamanid, were active but not yet potent enough to support 3 months or less of treatment. Dawson says, “Both investigational arms were clearing TB but not at the speed required to be considered for a 3-month regimen.”
Also at the conference, Maria Beumont, chief medical officer at the TB Alliance compared sorfequiline to 10-year-old bedaquiline as “10-fold more potent, retains activity against most frequent form of resistance against bedaquiline, and safer.” In a phase 2 trial comparing different doses of sorfequiline–pretomanid–linezolid (SPaL) with existing BPaL‑based regimens for drug-resistant TB and HZRE for drug-sensitive TB, the highest dose (100 mg) of sorfequiline tested showed stronger activity earlier, and preserved potency against bedaquiline‑resistant strains. If these findings hold, SPaL or related combinations could form the first universal regimen: one suitable for both drug‑sensitive TB and drug‑resistant TB, which would streamline treatment protocols worldwide.
Shifts in delivery
Another frontier in TB treatment is long‑acting injectables, which have already transformed treatment for human immunodeficiency virus with monthly or even twice-yearly injections after a 2-day lead-in of oral drugs. Injectables could address one of TB’s biggest barriers: adherence. Missed drug doses during lengthy regimens drive resistance and treatment failure. A long‑acting formulation of key drugs—administered monthly or even less often— could change that equation for both latent TB and active TB.
The first such candidate, long‑acting bedaquiline, entered phase 1 testing in 2024, with results expected in late 2026. Bedaquiline is normally an oral drug taken repeatedly over 6 months, so a safe, injectable version could ease delivery and reduce pill burden. “This is super exciting,” says Wells. “With long-acting injectable formulations, you could treat somebody at the very beginning and maybe only need to give them one more round of injections 2 months later.”
The TB Alliance and partners are also advancing long‑acting versions of sorfequiline, pretomanid and other candidates. Their goal is to deliver future ‘1×1’ strategies—1 day of treatment for latent TB, and injections and 1 month of treatment for active disease—if potency and pharmacokinetics align. Spigelmann says this goal is reachable “by combining the best drugs with better technology.”
Finally, several groups are testing immunomodulators, repurposed from other diseases, to undermine TB’s inherent ability to resist immune defenses. Because TB pathology is driven partly by immune‑mediated lung damage, these agents could reduce tissue destruction, shorten treatment, or help patients with difficult TB recover more fully. Early candidates are in phase 2, while the preclinical pipeline—accelerated by artificial intelligence‑guided target identification—is beginning to diversify.
