A bioreactor to predict the efficacy of antifungal therapies

A project grant awarded to Professor William Hope, from the University of Liverpool, has been used to create a bioreactor to predict the efficacy of antifungal therapies for Aspergillus infections, reducing the use of animals.

Research details

Principal Investigator: William Hope, Professor of Therapeutics and Infectious Diseases
Organisation: University of Liverpool
Award: £210,664, in 2007, over 24 months
Title: An in vitro model of the human alveolus to predict the efficacy of systemic antifungal therapy

Read more about Professor Hope's research

VIDEO: Professor Hope explaining his research and the impact on the 3Rs

Case study

Atmospheric Aspergillus causes lung infections in immunocompromised patients

Fungal spores are present in the atmosphere and are normally not detrimental to health. They can, however, cause fatal infections in immunocompromised patients, such as those receiving organ transplants or with haematological cancers. Each year in the UK there are an estimated 4,000 cases of lung infection caused by the fungus Aspergillus fumigatus. Approximately half of these patients will die. Providing improved or new therapeutics is essential.

Determining the clinical severity of fungal infections 

Clinical breakpoints are terms used to categorise microorganisms, including Aspergillus, as clinically susceptible, intermediate or resistant to antimicrobial agents. This information is used by clinicians to select appropriate therapeutics and doses for patients and is important in preventing drug resistance.

Breakpoints are determined using information on the pharmacokinetics (PK) and pharmacodynamics (PD) of the drug. This requires a model of infection which allows the efficacy of the drug to be assessed and the PK/PD relationship to be measured. Studies are typically performed in animals but the infection and response to drugs can differ from humans. This can make it challenging to establish safe and effective dose regimens for managing infection in patients because if a patient has drug levels below the clinical breakpoint they may not respond to treatment.

Large numbers of animals are used to establish the efficacious dose of antifungal drugs

Up to 2,000 mice are used to characterise the PK/PD relationship of each antifungal drug. For lung infections, in vitro models of the human alveolus have been developed but these are static and not suitable for PK/PD analysis of drugs.  In 2007, Professor William Hope, University of Liverpool (previously University of Manchester), was awarded an NC3Rs grant to develop a dynamic in vitro model to study invasive pulmonary aspergillosis and to measure the PK/PD relationship of antifungal drugs.

A microfluidic in vitro lung model of Aspergillus infection

Professor Hope has designed and constructed a bioreactor to house a cellular bilayer consisting of human alveolar epithelial cells and pulmonary endothelial cells grown on a semipermeable, polyester membrane. The bilayer delineates an upper compartment (representative of the air space of the alveolus) and a media-filled lower compartment (representative of the pulmonary capillary) connected to a central reservoir (representing the blood).

Aspergillus spores are introduced into the alveolar compartment, where they germinate to form hyphae (the invasive forms of the fungus) which quickly invade the cellular bilayer. Fresh cell culture media is pumped into the reservoir while spent media is removed at the same rate. Antifungal drugs can be then injected into the circuit, mimicking systemic drug administration. By sampling the media it is possible to determine first order PK.

The model also incorporates the biomarker galactomannan for assessing PD. This biomarker is a component of the Aspergillus cell wall which is released during growth. Its presence in blood is used to diagnose invasive aspergillosis in patients. The bioreactor can be sampled repeatedly for galactomannan allowing the generation of a rich data set that if performed in vivo would require large numbers of animals and serial blood sampling.

Translation to benefit patients

Voriconazole is an antifungicide widely used as a first-line therapy for the treatment of invasive pulmonary aspergillosis. A detailed understanding of its PK/PD relationship has been difficult to establish in animals.  Professor Hope has been able to use the new model to investigate the PK/PD relationship of voriconazole, establishing its clinical breakpoint. This was published in the scientific literature and on the website of the European Committee on Antimicrobial Susceptibility Testing, which is responsible for setting clinical breakpoints.

Professor Hope has secured additional funding from Astellas Pharma to use the model to define breakpoints for the novel antifungal compound, isavuconazole. The drug is now in Phase III clinical trials as a treatment against a range of medically important fungal pathogens. The model is also being used as part of collaborations with Canadian investigators, and for characterising novel antifungal compounds with the UK biotechnology company F2G. Its use could potentially be extended to other experimental contexts such as the immunopathology of infection.

There is one journal article describing the new model and one technical note arising from this award. Another five articles report the use of the model to evaluate different drugs and combination therapies.

This case study was published in a review of our research portfolio in November 2013.