Isolation and characterisation of monoclonal antibodies for the treatment or prevention of antibiotic resistant Acinetobacter baumannii infections

The bacteria Acinetobacter baumannii causes severe lung and blood-borne infections in humans. It is one of the most highly resistant bacteria to antibiotics, and as a consequence A. baumannii infections have a very high mortality which approaches 50%. Overall, A. baumannii causes over 50,000 deaths per year across the globe, a number which is increasing. It is especially common in Asia with 15,000 deaths per year in Thailand alone, and at one of our research site hospitals the number of people with A. baumannii per year has increased from 100 in 2010 to over 500 in 2022. The World Health Organisation has made A. baumannii its top priority antibiotic resistant bacteria for which we need new treatments. One way of overcoming antibiotic resistance is to treat bacteria with antibodies, naturally occurring proteins that bind to invading microbes and boost the ability of the immune system to kill them. Antibody therapies are known to work for other microbes but are not available as yet for A. baumannii. We aim to fill this gap by developing an antibody treatment for A. baumannii. Over the past four years, we have identified antibodies to four A. baumannii proteins that we have shown bind strongly to the bacterial surface and can increase activity of the immune system against the bacteria. Importantly when given to mice these antibodies protected against A. baumannii infection, indicating they could be a good treatment for human infection. We now want to develop these antibodies for use in humans. To do so we need to obtain single specific antibodies for each of our protein targets which are called monoclonal antibodies, as these can then be produced in a factory in the large quantities needed for a treatment. We will isolate several monoclonal antibodies to each of our four A. baumannii proteins from either humans who have had previous A. baumannii infection and have developed an immune response to this bacteria, or from mice using vaccination experiments. We will then test each isolated monoclonal antibody to see how well they bind to and promote the immune systems ability to recognise and kill A. baumannii strains. We will also test each monoclonal antibody to see whether they can protect mice against A. baumannii infection. The most effective monoclonal antibodies will then be tested in combinations as our previous work suggests this will be more effective than a single antibody. In addition, we will collect data and samples on patients with A. baumannii infection at our hospital research site in Thailand. The information on the patients is needed so we can plan future clinical trials of a monoclonal antibody therapy; and samples from the patients will also help with the experiments investigating the monoclonal antibodies by providing white cells from which we can isolate monoclonal antibodies. At the end of the study we will have the data needed to decide which of the monoclonal antibodies and in which combination are likely to be the most effective treatment for A. baumannii infections. These monoclonal antibodies will in the future be developed into a clinical treatment for testing in humans.

We will identify the most effective monoclonal antibodies to our selected protein antigens (antigens 1, 3, 5 and 7) for treating or preventing severe A. baumannii infections. To ensure successful identification of monoclonal antibodies we will use two approaches. Firstly, we will isolate human B cell clones producing IgG to the selected antigens from patients with proven A. baumannii infection, an approach we have used successfully in proof of principle studies (Figure 7). As human B cell cloning is slower and less reliable than mouse B cell cloning, we will also use mouse B cell cloning as a second approach for up to two antigens. The efficacy of derived monoclonal antibodies at promoting immune recognition of A. baumannii will be assessed using: (i) ELISAs and flow cytometry assays to measure IgG binding and avidity to target proteins, and recognition of clinical A. baumannii strains; (ii) flow cytometry assays of IgG-dependent complement activity and neutrophil phagocytosis; and (iii) neutrophil and complement bacterial killing assays. Each monoclonal antibody will be tested individually against four representative clinical AMR A. baumannii strains. The protective efficacy of each monoclonal antibody will be assessed in mouse models of A. baumannii infection with two different clinical AMR A. baumannii strains. The best performing candidate antibodies will be tested in bivalent, trivalent, and quadrivalent combinations to identify potential synergistic effects, and against an extended panel of ten strains for the in vitro assays. We will also extend the existing Thai clinical study of A. baumannii infection to better define the clinical characteristics of infection, characterise the causative A. baumannii strains using genome sequencing, and identify new donors for B cell cloning experiments. The data obtained during the project will then be used to select the best four monoclonal antibodies (one per protein antigen) for future clinical development.