Fungale mist in virale stormen doen optrekken: viraal-geassocieerde pulmonaire aspergillose ontwarren

Viral-associated pulmonary aspergillosis (VAPA) is a pulmonary superinfection with the fungus Aspergillus, affecting patients requiring intensive care for severe influenza (IAPA) or severe COVID-19 (CAPA). VAPA is associated with increased mortality rates, reaching approximately 50%. Diagnosis of VAPA mainly relies on bronchoalveolar lavage (BAL) sampling, and treatment currently consists of antifungals. Recognized as an important entity only in the decade following the 2009 H1N1 influenza pandemic, the pathophysiology and clinical characteristics of VAPA remain to be investigated. With this PhD project, we aimed to characterize the immunological and microbiota landscape of the lungs of patients with severe influenza or COVID-19 with or without VAPA in order to identify targets for future in-depth mechanistic and translational research. Moreover, we aimed to shed light on several clinical aspects of VAPA. In a first study, we set out to characterize the epithelial and myeloid innate immunity in VAPA, comparing patients with severe influenza or COVID-19 with or without aspergillosis, investigating multiplex gene expression and protein levels in BAL fluid and RNAscope and spatial transcriptomics in tracheobronchial biopsies. We found arguments for a triple barrier breach in antifungal immunity in patients with VAPA, with hampering of the epithelial barrier function, capacity to kill Aspergillus spores and tools to deal with Aspergillus hyphae. With this study, we identified several targets for follow-up studies to focus on in order to characterize the pathophysiology of VAPA more in-depth, including relative neutrophil abundance (reduced in CAPA patients compared to COVID-19-only patients), several Aspergillus-recognizing proteins (such as galectin-3, downregulated in IAPA patients compared to influenza-only patients), or immune pathways (such as type I and type II interferon signaling, downregulated in VAPA patients). In a second study, performing a post-hoc analysis of the first study, we compared the myeloid immunity profiles of patients with IAPA who survived versus those who died. We found that at time of IAPA diagnosis, non-survivors showed decreased expression of several genes linked to Aspergillus recognition and upregulation of genes linked to VEGF-signaling. The latter was corroborated by the finding that IAPA patients with the highest levels of VEGF at the protein level showed significantly increased 90-day mortality. The signatures found in this study imply a role for reduced Aspergillus recognition and increased lung fibrosis in the outcome of patients with IAPA, and could serve as starting point to develop host-based prognostic biomarkers. In a third study, we characterized the immune landscape of the lower respiratory tract of patients with severe COVID-19 with or without CAPA more in-depth, analyzing scRNA-seq data and neutrophil extracellular trap (NET) levels of BAL fluid. Main findings of this study were the depletion of neutrophils detected by scRNA-seq in patients with CAPA, caused by increased NETosis in response to the fungus. Higher NET levels were associated with increased survival among CAPA patients specifically, implying an important role for neutrophils in targeting CAPA. We also identified an increase in the fraction of hybrid neutrophils in patients with CAPA. These neutrophils show characteristics of antigen-presenting cells, and could pose a specialized response to the fungus. In a fourth study, we aimed to characterize and compare bacterial and Aspergillus superinfection in patients with severe influenza or COVID-19, by performing a retrospective study that included analyses on clinical data, as well as 16S rRNA gene sequencing, viral and bacterial load determinations, and host immunity analyses on BAL fluid of these patients. We found that IAPA could be detected significantly more often than bacterial superinfections in patients with severe influenza. Bacterial superinfections were identified sooner in patients with severe influenza compared to those with severe COVID-19. VAPA was associated with mortality in both patients with severe influenza or COVID-19, but detection of bacterial superinfection was associated with mortality only in patients with severe influenza. We found only minor differences in lung bacterial microbiota composition comparing patients with severe influenza versus COVID-19 with or without aspergillosis, and no differences in viral load. Bacterial load was higher in patients with CAPA, and could imply a worse mucociliary clearance in these patients as contributing factor to this finding and to the presence of aspergillosis. COVID-19 patients with bacterial superinfection showed a hyperinflammatory signature, which we did not identify in patients with influenza, implying that influenza tampers more severely with the capacity of the immune system to recognize and respond to secondary infections compared to COVID-19. In a fifth study, we wanted to compare incidences of CAPA before versus after widespread implementation of vaccination against COVID-19, given that vaccination has caused severe COVID-19 cases to plummet while increasing the fraction of severely immunocompromised patients among those who still end up in ICU for severe COVID-19. We performed a retrospective observational study among mechanically ventilated COVID-19 patients in University Hospitals Leuven, and found that the fraction of patients with CAPA increased dramatically transitioning to the vaccination era. In a sixth study, we analyzed data from almost 50 observational studies investigating VAPA incidence, and found that the large variation in reported VAPA incidence can at least partly be explained by large variation in use of bronchoscopy and BAL sampling. Finally, in a seventh study, we characterized the association of serum galactomannan positivity, reflecting angio-invasion, in the largest cohort of VAPA patients to date. Presence of serum galactomannan positivity among VAPA patients was lower than expected, but we found that positive (and especially if high-positive) serum galactomannan optical density indices were associated with a dismal outcome in these patients. Our findings corroborate that serum galactomannan should never replace lower respiratory tract sampling to find VAPA, and imply that more aggressive antifungal treatment strategies might be reasonable if serum galactomannan is found positive in a patient with VAPA. In conclusion, this project has provided the first insights in the pathophysiology and lung host immunological profiles of VAPA. A multi-level breach in antifungal immunity caused by the viral pneumonia forms a perfect storm for aspergillosis to occur and to become tissue-invasive. Hybrid neutrophils and NETs might play a beneficial role against VAPA. Moreover, this project showed the relative importance of VAPA over bacterial superinfections in patients with severe influenza or COVID-19 in the era of antibiotic use. VAPA is not a rare complication, as showcased by very high incidences of CAPA in mechanically ventilated patients since the vaccination era and its overall similar or even higher incidence compared to bacterial superinfection in patients with severe influenza or COVID-19. However, it is likely underdiagnosed and underreported due to lack of thorough sampling of the lungs. Use of galactomannan in blood does not form an alternative screening marker due to poor sensitivity, but this test shows prognostic value when positive, heralding poor outcome in patients with VAPA. With this work, we have opened doors to improve awareness and recognition of VAPA, and we have laid foundations for future in-depth research into host-directed medicine directed at these lethal superinfections.