Antibiotics have transformed modern medicine, making once-deadly bacterial infections treatable and allowing procedures such as organ transplants, cancer chemotherapy and complex surgery to be performed more safely.
But as bacteria evolve resistance to existing drugs, these treatments are increasingly under threat. Antimicrobial resistance (AMR) occurs when microorganisms no longer respond to medicines designed to kill them or stop their growth. For bacterial infections, this means antibiotics that once worked reliably may become ineffective, leaving clinicians with fewer treatment options and increasing the risk of complications.
Developing new antibiotics is therefore an important part of the response. Yet scientific progress faces a difficult combination of biological complexity, lengthy development timelines and commercial pressures. Even when researchers identify promising compounds, turning them into safe, effective medicines that reach patients remains a substantial challenge.
The question is no longer simply whether scientists can discover new antibiotics. It is whether research, investment and healthcare systems can deliver new treatments quickly enough, while preserving their effectiveness for the future.
Why antimicrobial resistance is becoming harder to control
Antibiotic resistance develops through natural evolutionary processes. Bacterial populations contain genetic variation, and exposure to antibiotics can favour organisms that possess traits allowing them to survive treatment. These bacteria can multiply and, in some cases, transfer resistance genes to other bacteria.
The overuse and misuse of antibiotics accelerate this selection process. Unnecessary prescriptions, inappropriate treatment and excessive use in human and animal health can all contribute to resistance. However, resistance can also emerge even when medicines are used appropriately, making continued surveillance and the development of new treatments essential.
The consequences extend beyond infections that are already difficult to treat. Resistant bacteria can undermine routine healthcare by making infections following surgery, cancer treatment or other medical procedures harder to manage.
The World Health Organization (WHO) highlighted this challenge in its 2024 Bacterial Priority Pathogens List, which identifies drug-resistant bacteria requiring particular attention from researchers, pharmaceutical developers and public health authorities.
The list includes critical-priority pathogens such as carbapenem-resistant Acinetobacter baumannii and resistant bacteria within the Enterobacterales group, which includes Escherichia coli and Klebsiella pneumoniae. These organisms can cause serious infections and may resist several important classes of antibiotics.
Other priorities include drug-resistant Pseudomonas aeruginosa, methicillin-resistant Staphylococcus aureus and resistant strains of Neisseria gonorrhoeae.
The challenge is that resistance does not develop at a uniform rate across all bacteria or medicines. Researchers must therefore identify where existing treatments are becoming less effective and direct development towards the infections for which new options are most urgently needed.
Why the antibiotic development pipeline remains limited
Despite growing recognition of the threat, the number of new treatments in development remains insufficient to address the scale of the problem.
The WHO’s 2025 assessment of antibacterial medicines found that the clinical development pipeline had fallen from 97 candidates in 2023 to 90 in 2025. Of these, 50 were conventional antibacterial agents, while 40 used non-traditional approaches, including bacteriophages, antibodies and treatments designed to modify the microbiome.
More concerning was the limited degree of innovation. The WHO classified only 15 of the 90 candidates as innovative under its assessment criteria, which include new drug classes, new targets, new mechanisms of action or an absence of known cross-resistance. Only five candidates were reported to be effective against at least one of the bacteria in the WHO’s critical-priority category.
The figures highlight a distinction between developing more antibacterial products and developing treatments that offer genuinely new ways to overcome resistance.
Some new medicines are modifications of existing antibiotic classes. These can still be valuable, particularly when they overcome specific resistance mechanisms or improve treatment options. However, bacteria may already possess resistance mechanisms that reduce the effectiveness of closely related drugs.
Developing antibiotics that act on different biological targets or use unfamiliar mechanisms could help address this problem, although discovering such compounds is scientifically challenging.
The earlier stages of research also show activity. The WHO identified 232 antibacterial candidates in preclinical development in 2025, covering products being investigated before human clinical trials. However, many will not progress to approval, and the transition from laboratory research to a medicine that can be prescribed remains uncertain.
New approaches could expand the search for treatments
Researchers are exploring several approaches that could complement conventional antibiotics and broaden the options available for treating resistant infections.
AI and drug discovery
Artificial intelligence is increasingly being investigated as a tool for identifying potential antibiotic compounds.
Machine-learning systems can analyse large chemical libraries, predict interactions between molecules and biological targets, and help researchers prioritise candidates for laboratory testing. These approaches may reduce the time spent screening compounds that are unlikely to succeed.
In 2020, researchers used deep learning to identify halicin, a compound that showed antibacterial activity in laboratory studies and against certain infections in animal models. The work demonstrated how computational methods could help identify potential antibiotics from chemical spaces that might otherwise be difficult to explore.
However, identifying a promising molecule is only the beginning. Researchers must still establish its safety, determine how it behaves in the body, assess its effectiveness against relevant infections and test whether resistance develops during treatment.
AI can therefore help direct the search, but it cannot replace experimental validation or clinical trials.
Bacteriophages: Using viruses against bacteria
Bacteriophages are viruses that infect bacteria. Because some phages target particular bacterial species or strains, researchers are investigating their potential to treat infections that no longer respond to conventional antibiotics.
Phage therapy could offer a different mechanism for attacking bacteria, including in situations where available drugs have limited effectiveness. Researchers are also exploring combinations of phages and antibiotics, which may provide complementary activity in certain circumstances.
The approach nevertheless presents challenges. Phages can be highly specific, meaning that clinicians may need to identify the bacteria responsible for an infection before selecting an appropriate treatment. Bacteria can also develop resistance to phages, while the immune system may affect how well a treatment works.
Manufacturing, quality control and regulatory requirements add further complexity. More clinical evidence is needed to establish where phage therapy can provide reliable benefits and how it should be incorporated into routine care.
New targets and alternative treatments
Other research is examining antimicrobial peptides, antibodies, anti-virulence compounds and therapies that influence the relationship between bacteria and their hosts.
Rather than always killing bacteria directly, some approaches aim to block the mechanisms that allow pathogens to cause disease. Others seek to help the immune system respond more effectively or restore microbial communities disrupted by infection and antibiotic treatment.
These strategies could eventually complement conventional antibiotics, particularly where resistance has narrowed treatment options. Their suitability will vary by pathogen, infection type and clinical circumstances, and they should not be treated as interchangeable with established antibiotic therapies.
The broader goal is to diversify the available tools for treating bacterial infections, rather than relying exclusively on successive generations of similar drugs.
The commercial problem behind antibiotic innovation
Scientific advances alone will not solve the antibiotic shortage. The economics of drug development create another significant barrier.
Developing a new antibiotic requires substantial investment in discovery, preclinical research, clinical trials, manufacturing and regulatory approval. The process can take many years, and candidates may fail at any stage.
Unlike medicines for some chronic conditions, antibiotics are generally prescribed for relatively short courses. New antibiotics intended for resistant infections may also be deliberately reserved for patients who need them most, limiting sales volumes.
This creates a difficult commercial equation. Developers need to recover their investment, but public health authorities have good reasons to avoid unnecessary use of new antibiotics, since excessive exposure can accelerate resistance.
Smaller biotechnology companies are particularly exposed to these pressures. They may be responsible for developing promising candidates but lack the resources to finance expensive later-stage trials or commercial manufacturing. Even successful products can struggle to generate sufficient revenue to sustain the business.
The result is a market in which the medical need for new antibiotics does not necessarily translate into a predictable financial return.
Can new funding models make antibiotic development viable?
Governments and healthcare systems are experimenting with ways to reward antibiotic innovation without relying solely on the volume of medicines sold.
One approach is a subscription-style payment model. Instead of paying primarily for each prescription, a healthcare system provides a developer with a defined payment for access to an antibiotic, subject to agreed conditions. This can help separate revenue from the number of doses used, reducing the commercial incentive to maximise sales.
The UK has been developing this approach through NHS England’s antimicrobial subscription models, which pay participating companies for access to selected antibiotics based on their value to the healthcare system rather than simply the volume prescribed.
The UK Government’s 2024–2029 national action plan on antimicrobial resistance also commits to addressing market barriers and supporting research into new antimicrobials, diagnostics, vaccines and alternative therapies.
Other possible measures include public funding for early-stage research, grants for clinical development, partnerships between universities and biotechnology companies, and international arrangements to share the cost of developing treatments for resistant infections.
No single funding model will resolve every problem. Policymakers must determine how to reward genuinely useful innovation, maintain reliable supplies and ensure that medicines are available to patients who need them.
Why faster diagnosis matters as much as new drugs
Developing new antibiotics is only one part of tackling antimicrobial resistance. Clinicians also need to know which bacteria are causing an infection and which treatments are likely to work.
When the cause of an infection is uncertain, antibiotics may be prescribed before laboratory results are available. In serious infections, this can be necessary to avoid dangerous delays. However, it can also mean patients receive broad-spectrum treatments that may not be required.
Rapid diagnostic technologies could help narrow treatment choices sooner. Molecular tests can detect particular pathogens or resistance-associated genetic markers, while other methods assess whether bacteria are susceptible to specific antibiotics.
The effectiveness of these tools depends on the infection, the organisms being tested for, and the resistance mechanisms involved. Not every test can identify every relevant form of resistance, and results may need to be interpreted alongside conventional laboratory investigations.
Nevertheless, faster and more accurate diagnosis can support more targeted treatment, reduce unnecessary antibiotic exposure and help preserve the effectiveness of existing medicines.
The WHO’s 2025 assessments of antibacterial development and diagnostic technologies highlight the need to improve both sides of the response: developing effective treatments and strengthening the ability to identify infections and guide treatment decisions.
Preserving the antibiotics already available
New treatments will be essential, but they cannot replace measures designed to slow the development and spread of resistance.
Antimicrobial stewardship aims to ensure that antibiotics are prescribed and used appropriately. This includes selecting suitable medicines, using the correct dose and duration, and avoiding antibiotics for infections that do not respond to them, such as most viral illnesses.
Infection prevention is equally important. Vaccination, effective hygiene, hospital infection-control procedures and improved surveillance can reduce the number of infections requiring treatment in the first place.
The problem also extends beyond human healthcare. Resistant microorganisms and resistance genes can circulate between people, animals and the environment. This is why the UK and international health authorities use a One Health approach, recognising the links between human health, animal health, food systems and environmental conditions.
These measures cannot eliminate resistance, which is a natural evolutionary process. They can, however, reduce avoidable selection pressure and help protect the effectiveness of treatments for longer.
Can innovation keep pace with antimicrobial resistance?
The search for new antibiotics is progressing, with researchers exploring AI-assisted discovery, new drug targets, bacteriophages and other approaches that could expand the treatment options available to clinicians.
Yet the latest WHO pipeline assessment shows that the number of candidates in clinical development has declined, while relatively few meet its criteria for innovation. The challenge is particularly acute for resistant bacteria that already withstand important classes of antibiotics.
Progress will depend on sustained investment in research, better support for promising candidates as they move towards clinical trials, and commercial models that reward useful treatments without encouraging unnecessary prescribing.
It will also require a broader strategy. Rapid diagnostics, infection prevention, responsible antibiotic use and international surveillance are all necessary to reduce the pressure that drives resistance.
There is no single breakthrough that can end antimicrobial resistance. Instead, the response will depend on a continuing supply of effective treatments, supported by scientific innovation and healthcare systems capable of using them appropriately.
