The inaugural GAMRIC conference took place in London on 1 to 3 October 2025, bringing together over 300 researchers, funders and innovators to discuss how we tackle the global crisis of antimicrobial resistance (AMR). If you missed it, catch-up on our summary of the presentations below.
This session chaired by Erin Duffy introduced the scale of the of global AMR crisis and discussed what is needed to develop a solution.
Kicking off the first session, David Paterson began by outlining the scale of the global AMR crisis, sharing data from the largest systematic analysis of AMR burden carried out to date1, which shows that the burden in terms of disability-adjusted life years (DALYS) caused by the most problematic resistant Gram-negative bacteria is greater than the individual burdens of tuberculosis, malaria, HIV and dengue.
Paterson argued that antibiotic availability was as much of a challenge as resistance, sharing how outcomes of two patients treated in middle-income and high-income countries were primarily determined by the medicines available.
Turning to drug development, he highlighted how two drugs – the monoclonal antibody pembrolizumab (Keytruda) and the antibiotic sulbactam/durlobactam had the same impact on reducing mortality but a huge difference in sales, which makes it a difficult road for those seeking funding for developing antibiotics.
Main takeaways:
Paterson was joined by Sam Barrell, CEO of Life Arc, Charlotte Watts, Executive Director of Solutions at Wellcome Trust, and Marc Gitzinger, CEO of Bioversys, for a panel discussion.
Main takeaways:
A powerful session on the unaddressed burden of mortality and morbidity caused by neonatal sepsis was chaired by Sumati Nambiar, and featured the following presentations:
Phoebe Williams outlined how neonatal sepsis affects millions of babies globally and is responsible for significant mortality and morbidity. Williams coordinates a global consortium, NeoSEAP, which is gathering data to better understand the burden of neonatal infections within the Southeast Asia and Pacific region.
A key challenge is not only antibiotic access but having the license to use antibiotics in the paediatric setting. This is not just an issue in low- and middle-income (LMIC) settings, Williams said, there is also an increasing need to use off-label antibiotics in high-income countries (HICs) to treat life-threatening neonatal and paediatric infections.
Main takeaways:
A wish-list for treating the rising global burden of neonatal sepsis:
Seun Makinde, of Shionogi provided a real-world perspective on paediatric development plans for new antibiotics, which are challenging because of the rapidly changing development physiology, immature organ function, and central nervous system vulnerability of neonates and additional oversight required.
Paediatric extrapolation is a method that uses evidence from adult drug studies to support safe use in children, ensuring that children participate in trials only when absolutely necessary.
Makinde shared a framework for conducting extrapolation and a case study in which they combined extrapolation with paediatric clinical trials to evaluate safety and pharmacokinetics (PK) and shared what worked.
Main takeaways:
Chris Darlow drilled further into obtaining pharmacokinetics (PK) and pharmacodynamics (PD) data to support neonatal antimicrobial drug development.
The traditional approach to developing antimicrobials in paediatrics is challenging for neonates, namely that PK is much more variable, especially in preterm infants, and disease processes can differ significantly compared with older children and adults.
Physiology-based PK modelling is used to predict compartment-specific (i.e. liver, adipose, brain, kidney) PK and can be useful for simulating PK at different stages of development.
Additional supporting PK data is required to demonstrate brain penetration for neonatal antibiotics and preclinical models mimicking neonatal meningoencephalitis can offer confidence (or not) in whether a treatment warrants further development.
Main takeaways:
Finally, John Bradley provided a clinician’s perspective on the considerable barrier of obtaining informed consent to enrol neonates into clinical trials.
There is never a ‘good time’ to have this discussion, but by becoming part of the care team, and committing to personally tracking the infant’s progress, parents see you as helping their baby get the best possible care.
Bradley uses specific language to reassure parents that the treatment is ‘add-on’ to standard care, has already been approved in older patients and the trial is regulator approved.
Main takeaways:
In this session chaired by Seamus O’Brien, we heard the challenges in treating sexually transmitted diseases in low-resource settings and how these constraints can force innovation.
Sinead Delany-Moretlwe described how 90% of curable STIs are in LMICs and disproportionally impact populations who, because of stigma, are often too fearful to access health services.
The current approaches to STI diagnosis and treatment in LMICs is syndromic management, i.e. recognition of signs and symptoms when patients present to health services. But patients are lost at each stage of management and not appropriately treated.
One of the challenges is missed diagnoses, and although there is evidence to support acceptability of point-of-care tests and modelling suggests opportunistic screening could dramatically increase the number of correctly treated infections, the costs of doing so were prohibitive.
Main takeaways:
Alison Luckey from GARD P provided an update on the status of antibiotic development for gonorrhoea, multidrug resistant gonorrhoea, which has acquired resistance to all first-line antimicrobials.
The development pipeline for novel gonorrhoea antimicrobials is sparse and phase 3 trials for two new drugs failed to show non-inferiority to the standard-of-care (SOC).
There is an exceptionally challenging target product profile for new gonorrhoea treatments, and a clinical trial design that has not advanced in years.
Barriers include oral formulation and single dosing requirements, globally variable SOCs and operational challenges in areas with underdeveloped healthcare resources.
Main takeaways:
A wish-list for treating gonorrhoea includes:
Susanne Hodgson finished on an optimistic note that a global gonorrhoea vaccine could be feasible. As a human-only pathogen with high antigenic variability, Neisseria gonorrhoeae is a tricky pathogen to develop vaccines against, but mechanistic insights from a meningococcal group B vaccine could help.
Main takeaways:
Removing jargon, balancing the urgency with positive solutions and tailoring the narrative to government priorities were the take home messages from this session chaired by Lotte Steuten exploring how we communicate the urgency of tackling AMR.
Economist Jorge Mestre-Fernandez outlined the challenges of convincing policymakers and payers that investing in AMR solutions is a good economic choice.
Existing treatments are cheap and the superiority of new products is hard to prove, and the current business model focuses on needing antibiotics but not wanting to use them, bringing uncertainty and financial risk for both payer and developer. Existing health technology assessments don’t consider the future social value of new antibiotics and only focus on patient impact today.
Main takeaways:
Robert Horne talked about how to get AMR on to the political agenda in an environment not currently prioritising healthcare.
Although AMR fits many criteria that typically prompt the US Congress to allocate funding, it is not presently viewed as a pressing concern. An alternative approach is to relate AMR to an issue that is a current and future priority – national security.
Main takeaways:
Adding an EU perspective, Marc Gitzinger emphasized how important it was for the global AMR community to speak as once voice on the urgency of the AMR pandemic. Policymakers need to hear that getting new antibiotics to patients saves money – in terms of healthcare costs, mortality and DALYs.
Main takeaways:
Eva Krockow argued that language plays a vital role in landing messages about the urgency of AMR and maintaining political momentum. Of 40 key illness terms, AMR-related terms were among the least memorable, highlighting the need for less jargon to ensure messages are understood.
Presenting the same information in different ways can also affect outcomes, with more positive framing (e.g. lives saved versus number of deaths) resulting in greater engagement.
Another challenge is the illusion of distance, where a problem seems geographically or temporally far away, affecting distant others, or feels invisible.
Main takeaways:
The current AMR discourse focuses on alarmist doomsday predictions based on future forecasts or stressing burdens in certain areas which can cause panic and lead to disengagement. The solution is to:
This session, chaired by Valeria Gigante, highlighted challenges in treating hospital- or ventilator-acquired pneumonia, including predicting if a colonising pathogen will cause infection and how to choose the most appropriate antibiotic without rapid diagnostics.
Stephan Harbath shared clinical cases illustrating the real-world challenges of diagnosing and treating hospital-acquired and ventilator-acquired pneumonia (HAP and VAP, respectively). These cases highlighted the heterogeneity of patients affected and the spectrum of outcomes that can follow.
One of the difficulties is that widely used HAP/VAP definitions are based on signs and symptoms that are neither sensitive nor specific and subject to high levels of inter-observer variability in HAP/VAP diagnosis.
Main takeaways:
Mo Yin shared data from a large case-based AMR surveillance study across 41 hospitals in 19 Asian countries highlighting that VAP is one of the main culprits of excess mortality. This showed how many potential deaths that could be averted if we could make resistant organisms susceptible to antibiotics, with carbapenem-resistant infections having the greatest potential impact.
Yin discussed potential solutions including prophylactic antibiotic use and reducing the duration of antibiotics.
Main takeaways:
Virve Enne presented results of the INHALE research programme, a real-world study trialling a PCR-based diagnostic for HAP/VAP, FilmArray Torch, and an associated prescribing algorithm, within 14 intensive care units in England3.
The PCR-based test led to a greater proportion of patients on an active and appropriate antibiotic at 24h, but there was no significant difference in the primary outcome of clinical cure from pneumonia.
The explanation for this remains unclear, but behavioural studies suggest prescribers did not always believe the PCR test results or trust the prescribing algorithms.
Main takeaway:
Phuc Phan Huu talked about burden of lower respiratory tract infections in children in low-and middle-income countries, and how inappropriate antibiotic use driven by patient demands, easy over-the-counter access, poorly regulated healthcare providers, and pressures on health workers is driving resistance.
Greater uptake of pneumococcal conjugate vaccine (PCV), influenza vaccines and the respiratory syncytial virus (RSV) vaccine and therapeutic antibody have potential to avert millions of antibiotic-resistant infections, but coverage is currently suboptimal because of barriers including cost, distribution constraints, and missed outreach opportunities.
Main takeaways:
Mycobacterium tuberculosis (TB) is one of the most drug-resistant pathogens and treating it requires a radical shift from traditional drug R&D approaches. This session, chaired by Kelly Chibale, covered:
Steve Berthel explained how the TB Drug Accelerator (TBDA) model brings together pharmaceutical and biotechnology companies, universities, research institutes and funders organisations to share compound libraries and data and work together to feed the TB drug pipeline.
The TBDA’s goals are to generate 1-2 mechanistically distinct TB drug candidates per year, sufficient to advance at least one universal drug regimen (that is, with no patient stratification or regiment adjustments) to a one-month clinical proof of concept. There are two main milestones: entry into portfolio and preclinical development candidate.
Main takeaways:
Lessons from TBDA that apply to other global health challenges included:
Dirk Schappinger provided an overview of how genetic approaches can predict novel TB targets. By conditionally deleting or degrading M. tuberculosis gene products in vitro and in vivo and studying the consequences they’ve identified important metabolic targets in TB.
Genome-wide ‘vulnerability analysis’ is then used to determine how much a drug needs to affect a target to impact TB infection. A target that needs to be inhibited by 99%, for example, sets too high a bar for a drug to achieve. The vulnerability concept can help identify targets where the drug effect doesn’t need to be as potent.
Main takeaways:
Bree Aldridge explained how, with 28 TB drugs to choose from, testing the potential 20,000 combinations would be impossible to test preclinically. An alternative is to use machine learning to optimize drug combinations for experimental testing.
DiaMOND is based on a chequerboard assay for measuring drug synergy under different growth conditions – such as dormancy or acidic growth. By adding more preclinical data in and classifying outcomes to be better or worse than SOC, the machine learning model can find predictive signals in the data to predict outcomes of different drug combinations.
Main takeaways:
Jansy Sarathy explained that studying drug pharmacology within TB lesions, rather than in conventional models, more closely matches what’s seen in patients.
By capturing granuloma tissue and developing animal models using this tissue it becomes possible to assess PD against non-replicating persistent M. tuberculosis.
This reveals which compounds have efficacy against different TB lesions, allowing development of a new generation of drugs that can reach and kill all populations of M. tuberculosis in a patient.
Main takeaways:
In this session, chaired by John Rex, experts explained why antifungal R&D is so difficult and shared how lessons from developing antibacterials could help.
Carole Sable began with a timeline of antifungal drug development, showing that the last approval was in 2001. There are still only four classes of antifungals, and many unmet needs.
Some of the challenges include the lack of fungal-specific targets, requirement for long-term treatment in patients who often have complex underlying disease, and the rarity of invasive fungal infections meaning that trials recruit slowly. There is also a lack of standard trial comparators, and diagnostics and SOC antifungals are not always widely available.
Despite looking promising in nonclinical models, it has taken 18 years for new first-in-class fosmanogepix to reach phase 3 trials for invasive mold infections. This was achieved through an expanded access program.
Main takeaway:
Shampa Das argued for more extensive pharmacology studies of antifungal drugs to supplement clinical data and discussed the preclinical models available.
Existing models fail to differentiate between PK/PD and efficacy, are often immunocompromised and time of exposure can shift between different strains of the same species.
There is also less use of PK/PD models to drive clinical dose selection for antifungals compared to antibacterials, where PK/PD data is being used to support approved indications.
Main takeaways:
Sumati Nambiar shared insights on the clinical trials you ideally want to do for antifungals, compared to those you can feasibly do.
Challenges include the difficult of conducting superiority trials, narrow margins for showing non-inferiority, and choosing appropriate clinical or microbiological endpoints. Example data packages illustrated how different clinical development might look for two different indications.
Main takeaways:
Radu Botgros talked through the current clinical trial guidance for antifungals, highlighting that this guidance has been challenged by drug developers because randomised trials for rare fungal infections are unfeasible and often unethical.
Current clinical trial guidance still focuses on pathogen-specific indications, but this is problematic for antifungal drug discovery because unlike bacteria, fungal targets are poorly defined and PK/PD data is sparse.
Alternative single-arm study designs using external controls, i.e. using data from patients receiving SOC in recent clinical studies, provide an alternative, but this is still challenging, as cohorts are small, diagnostic tests evolve, SOC treatments advance, and there are variations in patient populations and risk factors over time.
Main takeaways:
This session, chaired by Robert Skov and Holger Rohde, explored the role of improved diagnostics in tackling AMR and how to ensure global implementation.
Chantal Morel outlined the tough market for diagnostic innovators, explaining how labs are looking for reduced turnaround time and costs, and reimbursors want to see proof of clinical utility. Even if tests are rapid, transport time to the laboratory remains an issue.
An alternative is to introduce point-of-care tests, but this comes with safety and regulatory concerns issues with laboratory space. Siloed budgets and health financing structures are additional barriers – it’s unclear where the incentives to introduce better diagnostics lie.
Main takeaways:
Possible solutions to challenges of diagnostic development:
Lotte Steuten introduced the STRIDES framework, which is designed to help payers and policymakers systematically evaluate AMR diagnostics.
Existing frameworks focus on individual patient outcomes and short-term cost benefits, and overlook broader, long-term population level benefits.
STRIDES builds on an existing STEDI framework covering spectrum, transmission, enablement, diversity and insurance by adding broader elements of value, identified through literature reviews and expert.
The STRIDES framework now includes research and surveillance, to reflect the added values of improving trial recruitment and tracking AMR trends.
Main takeaways:
Robert Skov shared outcomes of a recent two-day workshop co-hosted by Africa CDC, The AMR Action Fund, ESCMID, AUROBAC-Tx and bioMérieux addressing the under-valuation and under-utilisation of diagnostics across different healthcare settings.
Forty-five participants from 39 organisations including researchers, regulators, professional societies, policymakers and the private sector mapped out challenges, identified actions and developed campaign plans.
Main takeaways:
Closing the second day with a Special Lecture, evolutionary microbiologist Vaughn Cooper introduced us to the future threat of ‘mirror bacteria’.
Mirror life has long been discussed as a concept, but technological advances mean mirror bacteria - built from achiral biological molecules - could plausibly be created by life scientists in the next 10-30 years.
These bacteria have little practical benefit, but pose unprecedented risk to humans, plants, animals and ecosystems.
A panel of experts convened to examine the potential of mirror bacteria, including numerous specialists actively engaged in mirror life research, and collectively published a series of recommendations.
Main takeaways:
The final day of GAMRIC 2025 focused on understanding interactions in healthcare centres that contribute to AMR in low-, middle- and high-income settings.
Edith Namakula and Mbabazi Phoebe Kalungu from the Mulago National Referral Hospital in Uganda outlined the AMR burden in Uganda.
Research carried out at Mulago hospital revealed suboptimal access to antibiotics, low staffing levels, pharmacy working hours and health worker attitudes were contributing to inappropriate antibiotic use.
An antibiotic stewardship programme involving clinicians, pharmacists, laboratory personnel and infection prevention control (IPC) processes introduced in 2024 included a 24-hour pharmacy dispensing medicines free of charge, annual antibiogram development to monitor antibiotic susceptibility patterns, and health worker training on AMR. Existing IPC processes were assessed to identify gaps for intervention.
There have since been improvements in diagnosis of malaria and TB, but diagnostic capability for bacterial infections is still lacking, and consistent supply of consumables such as antibiotic discs and costs of blood culture has proved challenging.
Uganda has high prevalence of sepsis and STIs, including MDR-resistant gonorrhoea where syndromic management may be driving resistance and high costs prevent microbial laboratory confirmation. There is also a high burden of viral infections in Uganda, including viral meningitis, and invasive fungal diseases.
Main takeaways:
These are urgent unmet needs in Uganda for:
Anand Anandkumar outlined how India, the world’s most populated country and consumer of most antibiotics, has both the best and most abysmal healthcare in the world. If an intervention works in India, it is likely to work in other healthcare settings, because it needs to be affordable and work against the toughest microbial isolates, in some of the world’s most rural, deprived regions.
Vijay Richard, Senior Consultant, Hospital Infection Control, Narayana Health City in Bangalore, a large hospital with high patient throughput, requiring efficient management of resources and prevention of infections to avoid creating additional demand on services.
Their approach to infection prevention involves clear and simple protocols, constant training, close supervision, audit and feedback. But although effective, these conventional IPC skills assessments are time-consuming, impractical for covering many staff and affected by inter-observer variability.
To address this, they are exploring the use of technology assisted training, skills assessment and practice audits – by using AI-based surveillance of critical care areas that can provide immediate feedback on IPC and upskill healthcare professionals.
George Varghese, a Senior Consultant in Infectious Diseases, shared case studies of two complex surgical case studies, and how they, as infection preventionists, worked with the critical care teams and surgeons to identify the infection source and treat appropriately.
Mallika Reddy outlined how they are using state-of-the-art diagnostic tools available in their setting to monitor antibiotic sensitivity patterns.
Taslimarif Saiyd is coordinating the India AMR Innovation hub to identify and address AMR challenges and argued that although India is a hotspot for Gram negative bacteria it is also part of the solution.
Main takeaways:
Like the LIC setting, India highlighted the following unmet needs:
Kartik Cherabuddi, Heinz Salazar, Suzane Silbert, and Kristen Zeitler from Tampa Hospital, USA, shared their process for evaluating new diagnostics and therapeutics and what it takes to implement these into hospital workflows.
When considering whether to introduce a new technology to a hospital workflow, they look at clinical use, performance, feasibility and cost. Clinical use might mean something that can’t be done in-house, an improvement to an existing diagnostic or whether it will influence antimicrobial therapy decisions.
A further consideration is whether a patient who started on a product in hospital can be transitioned home, or whether barriers such as frequency of administration, stability, formulation, prevent this.
The key barriers to introducing a new product tend to be cost and developing internal protocols. For example, multiplex PCR panels are relatively easy to use and adapt but translating these innovations into use can take time.
Staffing can also be an issue, as not all hospitals have 24/7 microbiology laboratory coverage which can impact the success of bringing in new technologies.
When building the business case for a new technology, it’s essential to combine data on the product’s performance, clinical utility and cost and align this to the hospital’s priorities. It can be helpful to engage with colleagues from other disciplines to help determine if a new product is really needed.
Business cases need to show how the product will be implemented in the hospital workflows and the value it adds to the institution, ensuring time and cost savings are quantified – including costs of readmissions, length of hospital stay and additional drugs. It’s essential to ensure the executive board understands and buys into the value framework being used.
As priorities shift from making money to saving money, it can be helpful if the decision-makers see that you can sustain cost-savings with the new intervention.
While the US team has seen huge gains in improving sepsis mortality, there was no progress in tackling pneumonia, and implementation science is a big issue. Barriers included a lack of de- or re-prescribing if a laboratory test result comes back at night, infectious disease doctors feeling uncomfortable to speak to surgeons about appropriate antibiotics, and junior doctors being overcautious.
Having open communication is key and as healthcare systems become increasingly digital, workflows need to include space for face-to-face conversations. The take-home message was to get the right people into conversations about AMR.
Affordability and accessibility of antibiotics is an issue for low- and high-income countries alike.
In Uganda, the drugs simply aren’t available, while US patients often can’t cover the costs of continuing treatment once they leave hospital.
In all healthcare settings, adoption of new technologies remains a challenge, and more ‘implementation science’ is needed.
Lessons learned from innovations that work in healthcare systems with budget constraints and remote, rural deprived communities will be relevant to high-income settings as well. While the challenge is global, the solutions will need to be regional.
The second annual GAMRIC conference will be held in Portugal from 22 to 24 September 2026. Programme details and registration information will be updated on our website as soon as possible.
References