AMR in pulmonary infections

The use case is presented using the example of 2 types of lung infection corresponding to 2 different clinical and ecological contexts:

1° Chronic lung infections that require repeated and long-lasting antibiotic courses.  The use of the ABRomics platform will be illustrated through the BEACH study (funding: PHRC-I API18/B/072 + Vaincre la Mucoviscidose), a prospective multicentre (n=11 centres in France) study (P.I. G. Héry-Arnaud) which aims to characterise the pulmonary and intestinal microbiome of newborns with cystic fibrosis (CF). The longitudinal metagenomic data generated by this program will enable the deciphering of the evolution of the bronchopulmonary community-acquired resistome during the first antibiotic cures (oral and inhaled routes) received by CF children in the first 2 years of life.

2° Acute lung infections, which are infections with a vital prognosis in the short term and which require rapidly effective antibiotic cures (short-lasting antibiotic courses, mostly intravenously). The use of the [name] platform will be illustrated through the HAP² study (funding: H2020 #847782), a prospective international study on hospital-acquired pneumonia led by an international consortium (P.I. A. Roquilly), and the COVARDS study (funding ANR Flash-COVID), a French cohort of patients with severe COVID-19 pneumonia. Samples and raw data from these studies are already available.  Multi-omic approach that integrates host-microbiota interactions generated by this program will enable the characterisation of the pulmonary ARGs distribution of hospital-acquired pneumonia (HAP) in intubated adult patients in a nosocomial context (HAP²) and in community-acquired pneumonia (COVARDS).

Pseudomonas aeruginosa is one of the most frequent and threatening bacteria linked with these both clinical entities, and belongs to the WHO global priority list of antibiotic-resistant bacterial species. Multidrug-resistant P. aeruginosa (MDR-PA) is an important and growing issue in the care of patients with CF. In VAPs, a high attributable mortality is observed despite adequate antimicrobial treatment that is increased in the presence of MDR-PA strains. The Pseudomonas NRC (headed by P. Plésiat) is currently developing a program to establish ARGs from clinical strains of different sources (including VAP and CF) and with well-defined antibiotypes. These WGS-derived data will enable PA-specific ARGs to be identified in the metagenomic data generated by the BEACH and HAP² studies. 

The three projects (BEACH, HAP2, COVARDS) will address specific lung AMR issues that will be resolved by the ABRomics PF.

1. Preliminary issue: How naïve (in regards with AMR) is the lung ecosystem of a CF newborn/patient at the admission in an ICU?
2. Issue 1: What are the differences between several classes of pulmonary antibiotics in terms of selection pressure in the lungs?
3. Issue 2:How does the commensal microbiome act on pathogens-related ARGs?
4. Issue 3:How does the host act on ARGs development in the lung?

Lung infection use case diagram describing how physicians, microbiologists and National Reference Centres (NRC) will find the support of the ABRomics-PF to integrate their data and answer the 3 essential questions (circled in black) brought by the BEACH, COVARDS and HAP2 studies on the issue of antimicrobial resistance specific to the pulmonary ecosystem

Deliverable 1: lung microbiome characterization

Deliverable 2: lung host transcriptome and metabolome characterization

Deliverable 3: ARGs characterization of antimicrobial-naïve lungs in children and adult patients

Deliverable 4: ARGs characterization after lung antimicrobial courses

Deliverable 5: ARGs expansion in the lung under antimicrobial selection pressure

Deliverable 6: benchmarking of MustARD database and PCM workflow in the lung ecosystem

Deliverable 7: P. aeruginosa ARGs characterization from lung clinical strains isolated in different epidemiological contexts (community- or hospital-acquired infections)

Deliverable 8: lung commensals ARGs characterization

Deliverable 9: benchmarking of Meta-PanGenomics Graph application in the lung ecosystem

Deliverable 10: development of an ARGs database and workflows specific to the lung ecosystem taking into account the temporal and multi-omics scales