Pneumonia, which can be acquired in the community (including influenza and COVID-19), is a leading cause of mortality. The risk of severe cardiovascular diseases events (stroke, myocardial infarction, pulmonary embolism) increases after infections, but causal mechanisms are not understood yet. There is an essential need for improved understanding of the relationship between pneumonia and cardiovascular diseases and early identification of patients at risk of cardiovascular events to develop tailored therapies.

The overall concept underpinning "Homi-lung" is to investigate the time course of host-microbiome interactions during & after pneumonia to i) understand the causal relationship between trained immunity, microbiome dysbiosis and cardiovascular and respiratory diseases (CVRD) progressions, ii) define endotypes of pneumonia associated with response to treatment & CVRD history; iii) develop biomarkers to predict the individual response to the treatment & CVRD progression, and iv) preclinically validate therapeutical approaches for CVRD during & after pneumonia.

Post-acute pneumonia syndrome People believe that there is a "modern pandemic" beyond the pandemic. This is called the post-acute COVID syndrome (PACS), and it is a constellation of symptoms and medical entities which emerge after acute infection by the new coronavirus SARS-CoV-2 (COVID-19). However, in this definition, people are attracted by the apparent symptomatology and ignore that long-term complications may be even more severe. In this regard, it is reported that the incidence of type 2 diabetes mellitus (T2DM) is increasing almost 1.56-fold after acute COVID-19, which may happen without any symptoms. The increase in the incidence of T2DM is supported by two large-scale meta-analyses involving more than 4.2 million patients during the post-COVID-19 follow-up period.

Recent evidence from the Hellenic Sepsis Study group suggests that circulating monocytes of patients after the acute COVID- 19 illness have increased ability for the biosynthesis of interleukin (IL)-1β, many of them do not present symptoms of PACS.

Taking into consideration the importance of IL-1β for the pathogenesis of T2DM through the destruction of β-pancreatic cell islets, it is evident that increased cardiometabolic (CV) risk may also be a counterpart of PACS. In the CANTOS randomized clinical trial published several years ago, survivors of a first myocardial infarct were randomized to treatment with a placebo or canakinumab, one monoclonal antibody targeting IL-1β, for five years. Results showed that anti-IL-1 treatment decreased by 15% the incidence of secondary cardiovascular events outscoring excess IL-1β production as a driver of CV risk. Consequently, it is reasonable to hypothesize that COVID-19 survivors who over-produce IL-1β may present with long-term CV events.

Beyond state of the art: respiratory dysbiosis, a complete reappraisal of the physiopathology of pneumonia for innovative treatments Healthy distal airways have long been considered sterile, and pneumonia was thus supposed to be caused by the contamination of the lungs by exogenous virulent pathogens (for CAP) or during micro- aspirations of the digestive contents in comatose patients (for HAP). Based on this physiopathology, numerous strategies to rapidly eliminate pathogens are recommended and widely used in Europe and worldwide. However, the limits of CAP and HAP treatments which increase bacterial or viral clearance, are highlighted in almost all randomized trials evaluating antibiotics or antiviral drugs in which the rates of treatment failure commonly exceed 30%, and by the 30%-rate of patients presenting with prolonged symptoms after pathogen clearance. A reappraisal of the physiopathology of pneumonia seemed necessary to overcome the relative failure and improve patient outcomes.

We have demonstrated that pneumonia outcomes depend on pathogen clearance and restoring healthy interactions between a weakened microbiome and altered immunity. Since CVRD progression is associated with disruption of the host-microbiome interactions, we hypothesize that the dysbiosis induced by pneumonia participates in the CVRD progression reported after the infection recovery. We thus propose to perform i) a longitudinal follow-up of host-microbiome interactions in large cohorts of patients cured of pneumonia to demonstrate clinically meaningful associations between immune reprogramming, microbiome dysbiosis and CVRD progression, and ii) preclinical investigations in calibrated mice models to demonstrate causality between dysbiosis and CVRD progression.

It is now well established that airways harbour a rich and diverse microbiome in healthy controls. Respiratory tract invasion by pathogens rapidly causes a loss of microbiome diversity and an impoverishment of host-microbiome interactions. These respiratory microbiome alterations play an essential role in the development of lung inflammation during pneumonia and reflect variation in baseline lung innate immunity. In published studies of mechanically ventilated patients, it has been demonstrated that lung microbiota are correlated with alveolar inflammation and that disruption of the gut microbiome (via anti-anaerobic antibiotics) increases patients risk of prolonged mechanical ventilation and mortality. These studies demonstrate the clinical significance of the microbiome in recovery from lung injury. Alterations of the gut microbiome derived metabolites also participate in the long-term immune reprogramming observed after sepsis.

In the healthy state, respiratory mucosal immunity actively controls the commensal bacterial agents in the airways. Numerous studies have revealed profound immune alterations in septic patients, considered immunocompetent" at the time of hospitalization. Partner Nantes Université has demonstrated that pneumonia induces prolonged immune reprogramming, characterized by the formation of paralyzed dendritic cells (DCs) and low phagocytic alveolar macrophages (MAC), that lasts for months in humans and is associated with prolonged susceptibility to bacterial and viral respiratory infections. These results demonstrate that the required immune response to contain respiratory pathogens and interact with the microbiome is rapidly dampened during pneumonia and that this immune reprogramming lasts for years.

Critically ill patients are highly variable in their recovery from lung injury. Much of this variation is attributable to the differential recovery of alveolar epithelial cell integrity and function. An improved understanding of lung epithelial recovery will be necessary to identify therapeutic targets for resolving lung injury and preventing CVRD progression. Lung epithelial cells are subject to constant exposure to 1) microbiota within the respiratory tract and 2) metabolites and translocated bacterial products from the lower gut microbiome. Yet the role of the microbiome in alveolar epithelial recovery is undetermined.

As a summary, we propose a reappraisal of the physiopathology of pneumonia based on the concept of dysbiosis between a weakened microbiome and sepsis-induced immunosuppression, which have the potential to explain the prolonged susceptibility to non-communicable diseases, notably by sustaining epithelial injuries.

Role of host-microbiome interactions in CVRD progression Current pharmaceutical interventions designed to slow the progression of atherosclerosis focus almost exclusively on reducing plasma cholesterol levels. However, clinical and experimental data support an additional critical role for inflammation in atherothrombosis. Notably, inflammation inhibition targeting the central NLRP3 inflammasome to IL-1 to IL-6 pathway of innate immunity is an emerging method for atherosclerosis treatment and prevention. Macrophage accumulation within the vascular wall is a hallmark of atherosclerosis, and in atherosclerotic lesions, macrophages respond to various environmental stimuli, such as modified lipids and cytokines. We have demonstrated that trained immunity develops early during pneumonia, correlates with the inflammatory response, and can help to predict long-term outcomes after viral pneumonia.

This innate immune reprogramming, lasts for months after sepsis recovery and is characterized by exacerbated inflammatory response and prolonged decrease phagocytic activity of monocytes and macrophages during secondary immune stimulation. We thus propose that the functional reprogramming of monocytes and macrophages observed after pneumonia, can alter the control of atherosclerosis plaques, increasing the risk of major CVD events. The gut microbiome has also emerged as a central factor affecting type 2 diabetes, obesity and the progression of atherosclerotic cardiovascular disease.

Integration of host-microbiome interactions to model the response to pneumonia and identify patients at risk of unfavourable outcomes early Each patient likely responds differently to therapeutic intervention and might recover differently after pneumonia. Indeed, some subgroups of patients suffer rapid CVRD progression, and others return to baseline conditions. Several biomarkers have been associated with pneumonia outcomes, but none have reached the accuracy required for clinical implementation. This is mainly because they are usually developed in small mono-centre cohorts and analyzed separately in the microbiome and host status. So, pneumonia treatments and rehabilitation care are "one-fits-all patients" approach leading to a large proportion of treatment failures and CVRD progression.

There is a critical need for reliable biomarkers for the stratification of patients predicting therapy success/failure and risk of CVRD progression/severe. The gold standard to reach these objectives is to use large cohorts of patients, bar coding of the samples, and high-throughput analysis followed by unbiased algorithm-guided analysis. In this setting, the description that the integration of the host response and the microbiome composition have a fair accuracy for the diagnosis of pneumonia demonstrates the potential of protocols investigating the host/microbiome interactions for the development of personalized treatment for respiratory infections.

The development and validation of endotypes to better understand the functional mechanisms associated with CVRD progression will help clinicians to adapt treatment and better prevent these conditions. The definition of phenotypes will also help identify patients at risk early. We thus propose to combine host background (sex/gender, age, ethnicity, medical history and genetic susceptibility, vaccination), CVRD risk factors (dyslipidemia, diabetes, obesity), inflammation and soluble mediators (metabolome, cytokines), immune status (epigenetic regulation) and microbiome composition during and after pneumonia to capture the complexity of the hostmicrobiome interaction time course and define endotypes and phenotypes associating pneumonia with CVRD.

Inclusion Criteria:

Group A (healthy controls)

1. Adults (18 years or more) of both genders (Female/Male: 50/50 ratio)
2. No history of severe pneumonia (sCAP, COVID-19 or HAP)
3. Presence of no or one of the following comorbidities: obesity (defined as body mass index over 35 kg/m2), type 2 diabetes mellitus, hypercholesterolemia, essential arterial hypertension, or familial history of CVD.

Group B (CVRD controls)

1. Adults (18 years or more) of both genders (Female/Male ratio: 50/50)
2. No history of severe pneumonia (sCAP, COVID-19 or HAP)
3. At least two of the following comorbidities: obesity (defined as body mass index over 35 kg/m2), type 2 diabetes mellitus, hypercholesterolemia, essential arterial hypertension, or familial history of CVD

Group C (COVID-19 survivors)

1. Adults (18 years or more) of both genders (Female/Male ratio: 50/50)
2. Survivors from severe COVID-19 pneumonia at hospital discharge; all patients had consolidation in chest X-ray or chest computed tomography during acute infection and were treated for pneumonia
3. SoC treatment for acute COVID-19 with dexamethasone

Group D (sCAP survivors)

1. Adults (18 years or more) of both genders
2. Survivors from sCAP pneumonia; these patients may be either hospitalized in the ward with pO2FiO2 ratio less than 300 or require admission and hospitalization in the Intensive Care Unit.
3. SoC treatment for sCAP with antibiotics

Exclusion Criteria:

Group A (healthy controls)

1. Presence of two or more comorbidities
2. Any other co-existing disorder generating CVRD symptoms
3. Limited chance of survival for at least six months due to co-existing comorbidity (-ies) according to the judgement of the attending physicians
4. Pregnancy or lactation

Group B (CVRD controls)

1. Any other co-existing disorder generating CVRD symptoms
2. Limited chance of survival for at least six months due to co-existing comorbidity (-ies) according to the judgement of the attending physicians
3. Pregnancy or lactation

Group C (COVID-19 survivors)

1. Medical history of severe congestive heart failure (Stage III-IV)
2. Medical history of stage III or IV dyspnoea according to the New York Heart Association classification before the acute COVID-19
3. Limited chance of survival for at least six months due to co-existing comorbidity (-ies) according to the judgement of the attending physicians
4. Pregnancy or lactation

Group D (sCAP survivors)

1. Medical history of severe congestive heart failure (Stage III-IV)
2. Medical history of stage III or IV dyspnoea according to the New York Heart Association classification before the sCAP
3. Limited chance of survival for at least six months due to co-existing comorbidity (-ies) according to the judgement of the attending physicians
4. Pregnancy or lactation

The study needs to enroll four groups of patients to cover the magnitude of the spectrum of community-acquired pneumonia and enroll the required controls:

• Group A: Controls with no or one comorbidity, predisposing to significant CV events and without a medical history of severe pneumonia. This group is important to provide baseline biological values and control samples to investigate normal host-microbiome interactions. No follow-up is performed for healthy controls.
• Group B: Controls with comorbidities predisposing to major CV events and without a medical history of severe pneumonia. This group is important to provide normal CVRD progression in patients at risk and control samples for the investigation of host-microbiome interactions in patients with stable CVRD.
• Group C: Patients cured of acute COVID-19.
• Group D: Patients cured of severe community-acquired pneumonia.