Evaluating anti-infective drugs

02 - 03 December 2024 09:00 - 17:00 The Midland Hotel Free
Illustration of a line graph

Theo Murphy meeting organised by Dr James Watson, Dr William Schilling and Professor Nicholas White.

Our objective is to draw upon experience and evidence across infectious diseases to derive common principles and approaches to in-vivo pharmacodynamic assessment and to discuss the optimal design, evaluation and interpretation of pharmacometric studies. We will compare the pharmacometric approaches and methodologies used across major infectious diseases to inform the development and rapid assessment of interventions for future pandemics.   

Attending the meeting

This event was intended for researchers in relevant fields, and was a residential meeting taking place at the Midland Hotel, Manchester. 

  • Free to attend
  • Advance registration essential
  • This was an in-person only meeting

Enquiries: contact the Scientific Programmes team.

Organisers

  • Dr James Watson

    Dr James Watson, University of Oxford, UK

    James Watson is a statistician and the Associate Director at the Infectious Diseases Data Observatory, a data re-use platform aimed at helping improve the treatment of major infectious diseases. His main work focuses on severe malaria, notably improving the definition of severe malaria in areas of high malaria transmission. He helped develop the statistical methodology for adaptive platform trials characterising antiviral effects of therapeutics in SARS-CoV-2, influenza, RSV, and dengue. He is an investigator on the CHARM project, which aims to develop a new methodology for the assessment of new antiparasitic drugs for the treatment of Chagas disease.

  • blank avatar

    Dr William Schilling, Mahidol Oxford Tropical Medicine Research Unit, Thailand

    Will is a an Infectious Diseases/ Microbiology Registrar working at St George’s Hospital London and was a research physician working at the Mahidol Oxford Tropical Medicine Research Unit (MORU) prior to this. At the time of reading this biography he may or may not now have a DPhil.

    Will’s work at MORU focussed on COVID-19. This was initially a large multinational prophylaxis study of hydroxychloroquine/ chloroquine but as the pandemic progressed, a platform pharmacometrics study assessing various therapeutics in early treatment- PLATCOV. Subsequently, Will has been involved in similar studies in influenza, RSV and dengue.

    Will has previous experience working in Medicine in Mbarara, Uganda. He hopes to one day do a 5K or 10K fun run.

  • blank avatar

    Professor Nick White FRS, Mahidol Oxford Research Unit

    Professor Nick White is Professor of Tropical Medicine at the Faculty of Tropical Medicine, Mahidol University, Thailand and Oxford University, UK. He is a Wellcome Trust Principal Research Fellow who chairs the Wellcome Trust Tropical Medicine Research Programmes in South East Asia. His main research focus is the pathophysiology and treatment of malaria. He also currently conducts pharmacometric research on COVID-19, influenza, and Chagas disease. Professor White has received the Prince Mahidol Prize for Medicine, the Canada Gairdner Foundation Global Health Prize, and the Royal Society of Tropical Medicine and Hygiene Manson medal. He has been a member of the WHO antimalarial treatment guidelines committee for 20 years (and co-chaired it for twelve) and he currently chairs the Scientific Advisory Committee of the Drugs for Neglected Diseases initiative and the Coalition for Equitable Research in Low Resource Settings.

Schedule

Chair

Professor Julie Anne Simpson

Professor Julie Anne Simpson, University of Melbourne, Australia

09:00-09:05 Welcome by the lead organiser
Dr James Watson, University of Oxford, UK

Dr James Watson, University of Oxford, UK

09:05-09:30 Malaria

Malaria is the most important parasite of man. Most deaths from falciparum malaria (estimated 2000 per day) occur in African children. Malaria disease results from the proliferation of intraerythrocytic Plasmodium parasites. The first symptoms of illness (fever) occur when the number of parasites in the blood reaches about 2 million/ kg. This corresponds with about 1 parasite per 200 white cells in a thick blood smear. Mortality rises when parasite burdens exceed 2 x 1010/ kg (corresponding to about 2% parasitaemia), although the relationship between intravascular parasite burden and parasite density in the peripheral blood smear is confounded by the sequestration of erythrocytes containing mature forms of the asexual parasites in the deep vasculature. The initial multiplication of malaria parasites in people with no immunity approximates 10 fold per asexual cycle (which is 48 hours in P. falciparum and P.vivax), but this stops abruptly at high densities. Antimalarial drugs which accelerate ring stage clearance can be readily assessed by measurement of parasite clearance rates. Intra-host modelling can also be used to assess in-vivo inhibitory concentrations and assess therapeutic responses. Simple pharmacometric models explain most aspects of the therapeutic response in malaria and can be used to optimise treatment regimens.

Professor Nick White FRS, Mahidol Oxford Research Unit

Professor Nick White FRS, Mahidol Oxford Research Unit

09:30-09:45 Discussion
09:45-10:15 The application of clearance markers in clinical trials of fungal infections

Invasive candidiasis and cryptococcal meningitis are the most common invasive fungal infections globally, affecting patients in the ICU and with advanced HIV respectively. 

Early fungicidal activity (EFA), a continuous measure of fungal clearance from serial samples of cerebrospinal fluid- was first implemented by our research group in clinical trials in the early 2000s and has now been widely adopted as an endpoint in phase II and III  trials of cryptococcal meningitis.

The task is more challenging in other mycoses, including invasive candidiasis/ candidaemia, where the evaluation of novel antifungals in the pipeline has been hampered by the lack of treatment response biomarkers. Preliminary data on beta-D glucan (fungal cell wall antigen) clearance kinetics from an ongoing observational study in candidaemia will be presented for discussion.

Professor Tihana Bicanic

Professor Tihana Bicanic

City St George's University of London, UK

10:15-10:30 Discussion
10:30-11:00 Break
11:30-11:45 Discussion
11:45-12:15 HIV
Professor Saye H Khoo, University of Liverpool, UK

Professor Saye H Khoo, University of Liverpool, UK

12:15-12:30 Discussion
13:30-14:00 Global challenges in antimicrobial therapy and AMR
Professor William Hope OBE, University of Liverpool, UK

Professor William Hope OBE, University of Liverpool, UK

Chair

Dr James Watson

Dr James Watson, University of Oxford, UK

13:30-14:00 Modelling to support drug repurposing against hemorrhagic fever viruses

Jeremie will discuss the use of models to optimise drug repurposing against Ebola, Lassa, and Nipah viruses. He will specifically focus on the use of experimental animal models and translation to humans. The talk will also discuss recent similar developments in the context of mpox infection.

Jeremie Guedj, French National Institute of Health and Medical Research, France

Jeremie Guedj, French National Institute of Health and Medical Research, France

14:00-14:15 Discussion
14:15-14:45 Treatment of influenza

This talk will cover antiviral agents and immune-modulating agents in treating influenza.

Professor David S Hui, The Chinese University of Hong Kong, Hong Kong

Professor David S Hui, The Chinese University of Hong Kong, Hong Kong

14:45-15:00 Discussion
15:00-15:30 Break
15:30-16:00 Dengue nonstructural protein 1 as a surrogate marker: enhancing evaluation of antiviral efficacy alongside viremia levels

Dengue virus (DENV) infection remains a significant global public health challenge. Currently, no effective antiviral drug exists for DENV, and available vaccines show inconsistent efficacy across all four serotypes, with concerns about antibody-dependent enhancement in seronegative recipients. As such, developing antiviral treatments remains a critical need. In early-phase clinical trials, virological efficacy is typically assessed by measuring reduction in viremia levels and time to viremia clearance, which indicate effective drug action. However, symptomatic dengue patients generally seek medical care when viremia is already in decline, limiting the window for drug administration and complicating efficacy assessment. Furthermore, viremia often becomes undetectable within days without any intervention.

Nonstructural protein 1 (NS1), which correlates with both viremia and disease severity, offers a complementary marker. NS1 remains detectable after viremia clearance, reflecting ongoing viral replication in infected cells, and has been shown to exert direct pathogenic effects in the host. This broader detection window for NS1 allows for extended evaluation of drug efficacy, beyond the short period when viremia is measurable. Moreover, a reduction in NS1 levels may not only indicate decreased viral replication but also suggest a potential reduction in disease severity due to its role in viral burden and pathogenicity.

Utilising NS1 antigenemia as an additional surrogate marker for viral replication could enhance the ability to assess antiviral efficacy more comprehensively. While early treatment remains key to achieving maximal antiviral benefit, integrating NS1 measurements could provide a more practical and informative tool for evaluating antiviral candidates during clinical trials.

Associate Professor Panisadee Avirutnan, Mahidol University, Thailand

Associate Professor Panisadee Avirutnan, Mahidol University, Thailand

Dr Junjie Ding

Dr Junjie Ding

Mahidol University, Thailand

16:00-16:15 Discussion
16:15-16:45 Rickettsial infections

Rickettsial infections have a global distribution and are transmitted to humans by ectoparasites such as ticks, mites, and fleas. Clinical presentations range from a mild febrile syndrome to, rarely, life-threatening sepsis or meningo-encephalitis. In Southeast Asia, species causing most human disease are Orientia tsutsugamushi and Rickettsia typhi. Accumulating evidence from clinical studies suggests doxycycline-based treatment regimens are optimal; however, defining suitable clinical endpoints for trials is challenging. Investigating the rate of bacterial killing in patients on different antimicrobials would improve our understanding of the therapeutic response.

In Laos a randomised pharmacokinetic-pharmacodynamic study is underway to characterise clearance of Orientia tsutsugamushi and Rickettsia typhi from the blood of patients with scrub typhus or murine typhus treated with either doxycycline or azithromycin. Preliminary results of the rate of bacterial clearance estimated from serial qPCR measurements and clinical outcomes will be presented.

Professor Elizabeth Ashley, Lao-Oxford- Mahosot Hospital-Wellcome Trust Research Unit, Laos

Professor Elizabeth Ashley, Lao-Oxford- Mahosot Hospital-Wellcome Trust Research Unit, Laos

16:45-17:00 Discussion

Chair

Professor Elizabeth Ashley

Professor Elizabeth Ashley, Lao-Oxford- Mahosot Hospital-Wellcome Trust Research Unit, Laos

09:00-10:00 Short talks

Modelling viral load kinetics during the COVID-19 pandemic: public health applications and lessons learned, Dr James Hay

One of the most controversial topics during the COVID-19 was the implementation of occupational health testing programmes and guidelines for at-home testing to detect, isolate and release infectious individuals. Much of the evidence to guide these policies initially came from mathematical modelling studies in lieu of sufficient data on viral kinetics; a topic which saw little focus from infectious disease modellers prior to the pandemic. In addition, as the landscape of SARS-CoV-2 infection changed with new variants, immunity from vaccination and infection, and availability of antivirals, models and assumptions required constant updating. In this talk, James will give a brief overview of key mathematical modelling results and data on SARS-CoV-2 viral kinetics (timing, peak and variation of viral load following infection) and testing which he and others generated during the COVID-19. He will specifically focus on data from the USA National Basketball Association occupational health testing programme, which provided an unprecedented volume of longitudinal RT-qPCR testing data through the wildtype, Alpha, Delta and Omicron waves. Quantifying the time course of acute respiratory virus infections in this way has important implications for public health policies around screening, isolation, and surveillance, particularly in the context of vaccines and therapeutics.

TREAT-GNB, a platform trial for severe Gram-negative bacterial infections , Dr Mo Yin

Current clinical trial models for carbapenem-resistant Gram-negative bacterial (CRGNB) infections pose significant barriers to translation including overly stringent inclusion criteria that focus on limited types of infections, rigid randomisation treatment options, and a lack of access to patient cohorts at-risk of CRGNB infections. The ADVANCE-ID (Advancing Clinical Evidence for Infectious Diseases) network was jointly established by the Wellcome Trust, the University of Oxford, Mahidol University, Christian Medical College Vellore (CMC Vellore), and Singapore academic institutions to promote transnational research collaboration and capacity building for a sustainable pipeline to conduct large-scale clinical trials in infectious diseases. We challenge the existing paradigm with our platform trial, TREAT-GNB, which adopts a novel trial design: the Personalised RAndomised Controlled Trial (PRACTical) design. The main strength of the PRACTical design is a flexible randomisation list which tailors treatment based on individual patient requirements. Of all the regimens under consideration, study participants would be randomly assigned only to those regimens that were considered clinically reasonable for that patient at that time, incorporating antimicrobial susceptibility, toxicity profile, and physician judgement (ie using a personalised randomisation list). The set of patients with the same personalised randomisation list would then form the unit analogous to a trial in network meta-analysis. The design allows for existing antibiotic combinations to be included in the randomisation list, with potential to engage industry partners to add novel FDA- or European Medicines Agency-approved antibiotics to the randomisation lists, as they become available, so that they can be ranked against other regimens. Regimens will be ranked according to efficacy, resistance phenotype, safety and cost. In this talk, the speaker will introduce the PRACTical study design and its application for CRGNB infections. 

Influenza, Dr Phrutsamon Wongnak

Viral respiratory infections represent a major global health threat, contributing significantly to morbidity and mortality, particularly in high-risk populations. Developing a robust pharmacodynamic framework to measure and compare the efficacy of antiviral agents is essential for informing intervention strategies aimed at mitigating the public health impact of these infections. This presentation outlines the use of adaptive open-label, phase 2, randomised controlled platform trials and the measurement of viral clearance rates, rather than time to PCR negativity, as a primary outcome in assessing antiviral efficacy. This approach is being applied to infections caused by SARS-CoV-2, influenza, and respiratory syncytial virus.

Once randomised, daily serial oropharyngeal swab samples were collected from participants, and viral densities were measured daily using qPCR. A hierarchical Bayesian modelling framework was employed to estimate the viral clearance rate, defined as the slope of the log-linear decay in viral densities. The antiviral effect is defined as the relative change in the viral clearance rate among participants receiving treatment interventions compared to those who did not receive any treatment.

Additionally, in infections with highly variable viral density dynamics over time, such as SARS-CoV-2, the impact of these temporal variations on head-to-head comparisons of antiviral efficacy and strategies for optimising sampling designs are discussed.

A within-host model of chronic Trypanosoma cruzi infection to help inform drug and regimen selection, Miss Somya Mehra

Trypanosoma cruzi is a kinetoplastid parasite causing Chagas disease, a major neglected tropical disease thought to affect over 6 million people worldwide. The current recommended treatment is 8 weeks of twice daily benznidazole, a regimen which is poorly tolerated in around a third of patients. There has been recent interest and funding in exploring intermittent treatment regimens and developing novel anti-parasitic drugs for T cruzi, with two candidates now in phase 1 trials. However, assessment of anti-parasitic efficacy in vivo for the treatment of chronic disease remains challenging. While correlates of blood-stage parasitaemia can be obtained using PCR methods, patients with chronic Chagas disease typically harbour very low parasite densities in the blood, close to the lower limit of quantification. Parasite biomass in the tissue additionally remains unobserved with current methods. 

In order to characterise the dose-response relationship for drugs against tissue stage and blood stage parasites, it is necessary to develop stochastic within-host models of T cruzi parasites. Multitype branching processes yield a natural theoretical framework to explore the cycling of T cruzi parasites between the bloodstream (trypomastigotes) and a tissue reservoir (amastigotes), including determinants of the steady state parasite burden; the probability of stochastic extinction at low parasite densities; and the potential consequences of drugs with differential effects on tissue and blood stage parasites. Interrogating simple models of T cruzi parasite clearance and replication can yield insights of practical value.

Optimal Resource Allocation for Infectious Diseases, Dr David Price

Design of experiments is not a new concept. However, much of the classical experimental design work was developed for the purpose of hypothesis testing, ie how many resources are required to ensure sufficient statistical power to test a formal hypothesis. In contrast, many disciplines of research, including discovery science and public health, are often more concerned with estimation of a quantity or learning about a system, rather than hypothesis testing.

This talk will introduce the optimal design framework, a decision-theoretic approach to get the most information from the limited available resources. We will consider scenarios where the purpose is to either estimate parameters or discriminate between competing models as well as possible. Examples in each case will relate to learning about infectious disease dynamics.

Dr James Hay

Dr James Hay

University of Oxford, UK

Dr Mo Yin, University of Singapore, Singapore

Dr Mo Yin, University of Singapore, Singapore

Dr David Price, University of Melbourne, Australia

Dr David Price, University of Melbourne, Australia

Dr Phrutsamon Wongnak

Dr Phrutsamon Wongnak

Mahidol Oxford Research Unit, Thailand

Miss Somya Mehra, Mahidol Oxford Tropical Medicine Research Unit, Thailand

Miss Somya Mehra, Mahidol Oxford Tropical Medicine Research Unit, Thailand

Professor David Price, Vice-Provost, Research, UCL

Professor David Price, Vice-Provost, Research, UCL

10:00-10:30 Discussion
10:30-11:00 Break
11:00-11:30 Bridging the gap: pharmacokinetic-pharmacodynamic modelling to optimise antimalarial dosing

The success of malaria control relies on the availability of highly effective antimalarial drugs. The widespread emergence of drug-resistant parasites now threatens the efficacy of first-line treatments, necessitating the urgent development of novel regimens and combinations of existing and new therapeutic agents to ensure adequate cure of malaria. 

To addressing this challenge, mechanistic within-host models that integrate the blood antimalarial drug concentrations with the parasite-time profile (also known as antimalarial pharmacokinetic-pharmacodynamic (PK-PD) models), provide a valuable decision tool for determining optimal antimalarial dosing regimens.

This presentation will provide an overview of published antimalarial PK-PD models, highlighting the differences in the mechanistic structure to capture malaria biology, and the relationship between drug concentration, drug-drug interaction and pharmacodynamic effect. Additionally an evaluation of the statistical approaches used for model validation against clinical data will be described, and a three step framework for the development and evaluation of PK-PD models recommended.

Professor Julie Anne Simpson, University of Melbourne, Australia

Professor Julie Anne Simpson, University of Melbourne, Australia

11:30-11:45 Discussion
11:45-12:15 Opportunities for therapy in Chikungunya infection

Chikungunya is an arboviral infection whose clinical presentation is very similar to those of dengue and zika. Fever occurs in the great majority of individuals and viremia is measured in all patients till 72 hours of the beginning of symptoms. Viral decay is homogeneous and most individuals are negative at 4 days after they are first seen. Although most symptoms are limited to the first days of infection, at 7-9 months after infection, 50% of patients who had chikungunya still reported some degree of discomfort and 25% needed to take some kind of pain killer to decrease symptoms. In patients with Chronic Chikungunya infection, there are 3 major clinical syndromes: (i) Rheumatoid arthritis-like; (ii) Spondyloarthropathy-like  and (iii) fibromyalgia-like syndromes. This classification of clinical presentation is useful for therapeutic purposes and follow-up of patients. Most individuals report their disease is significantly better at 18 months after infection. Experimentally, Chikungunya virus infection of mice is associated with transient viremia and viral infection of the site of infection and significant persistent inflammation and TNF expression at the dorsal root ganglia that innervates the infective site. These studies show that the viremia that follows chikungunya infection is high and has a kinetics compatible with the testing of antiviral compounds. In participants, with chronic infection, aggressive anti-inflammatory and pain management is needed for return to normal activities, especially because there is much impact on quality of life and the disease tends to very limited in duration.

Professor Mauro Martins Teixeira, Federal University of Minas Gerais, Brazil

Professor Mauro Martins Teixeira, Federal University of Minas Gerais, Brazil

12:15-12:30 Discussion

Chair

blank avatar

Professor Nick White FRS, Mahidol Oxford Research Unit

13:30-14:00 Reductio ad unum: the single patient RCT

Pharmacometric studies allow for greater efficiency in determining effects which translate to smaller sample sizes. Certain rare diseases with particular profiles, and particular drugs, discussed in this talk, may allow for sample sizes of 1.

Dr William Schilling, Mahidol Oxford Tropical Medicine Research Unit, Thailand

Dr William Schilling, Mahidol Oxford Tropical Medicine Research Unit, Thailand

14:00-14:15 Discussion
14:15-14:45 Perspective on PK/ PD and QSP modelling for anti-infective drug discovery and development
Suzanne Gaudet, Novartis Institutes for Biomedical Research, USA

Suzanne Gaudet, Novartis Institutes for Biomedical Research, USA

14:45-15:00 Discussion
15:00-15:30 Break
15:30-16:00 Regulatory perspective

Regulatory agencies have been adjusting their requirements for early clinical development of anti-infective agents, taking into account scientific advances in the area of PK/ PD so make dose finding studies as efficient as possible. PK/ PD evaluations are considered now standard practice for antibacterial medicines. Proper preclinical evaluation of PK/ PD together with Phase I PK and safety data could allow skipping formal dose finding studies in the context of antibacterial agents development. In other areas, clinical trial to select appropriate doses for confirmatory clinical trials are still needed. Surrogate endpoints based on a pharmacodynamic marker could be considered based on the level of confidence that has been gathered with respect to specific microbiological evaluations, eg viral loads measurement in upper respiratory tract for new antivirals for respiratory viruses. More work is needed to advance translational medicine that could increase the efficiency of clinical development for new treatments for infectious diseases.

Dr Marco Cavaleri, European Medicines Agency, Netherlands

Dr Marco Cavaleri, European Medicines Agency, Netherlands

16:00-16:15 Discussion
16:15-17:00 Panel discussion and overview