MASS cohort: Multicenter, longitudinal, and prospective study of the role of microbiome in severe pneumonia and host susceptibility

The MASS cohort comprises 2000 ICU patients with severe pneumonia, covering community‐acquired pneumonia, hospital‐acquired pneumonia, and ventilator‐associated pneumonia, sourced from 19 hospitals across 10 cities in three provinces. A wide array of samples including bronchoalveolar lavage fluid, sputum, feces, and whole blood are longitudinally collected throughout patients' ICU stays. The cohort study seeks to uncover the dynamics of lung and gut microbiomes and their associations with severe pneumonia and host susceptibility, integrating deep metagenomics and transcriptomics with detailed clinical data.


STUDY DESIGN
MASS is designed as a prospective, multicenter cohort comprising 26 ICUs across 19 medical centers, spanning 10 cities of three provinces, with a total of more than 800 ICU beds (Figure 1A and Table S1).These ICUs possess extensive expertise in the diagnosis and treatment of severe pneumonia, equipped with skilled medical staff.Among these medical centers, 14 are situated in seven cities of Zhejiang province, four in two cities of Henan province, and one in Shaanxi province.We plan to utilize patients in Zhejiang province as the discovery cohort (ZJ cohort), while patients from other provinces serve as external validation cohorts (non-ZJ cohorts).Alternatively, considering the potential geographic variation in microbiomes, we plan to allocate 20% of patients of each province as the respective validation cohort.The clinical trial has received approval from the Ethics Committee of the First Affiliated Hospital of Zhejiang University School of Medicine (ethic ID: IIT20230371B and IIT20230183B-R1) and registered on ClinicalTrials.gov(NCT06114784).
We plan to recruit 2000 ICU patients for longitudinal sample collection and follow-up visits (Figure 1B and Table S2).Upon enrollment, their basic information, medical history, and imaging data are documented (see "Collection of clinical data" in Supporting Information).Simultaneously, BALF, sputum, stool, and blood samples are collected for laboratory culture and next-generation sequencing (NGS), among other analyses (see "Methodologies of biological sample processing" and "Methods of bioinformatics analysis" in Supporting Information).When applicable, these samples will be collected longitudinally at 4, 7, 14, and 21 days after ICU admission, alongside the ventilator parameters recorded on those days.Drug usage will be documented on Days 14 and 28, and complications on Day 28.Additionally, follow-up visits will be conducted on Day 90 and 365 (Table S2).The primary aims of the MASS cohort are: (1)  investigating the dynamic characteristics of the lung and gut microbiomes in severe pneumonia patients and (2) exploring potential associations between microbiomes and host susceptibility to severe pneumonia.
The secondary aims of the MASS cohort include: (1) examining the relationships between lung microbiome and treatments, drug usage, and exposure; (2) investigating the variations in drug-resistance genes and horizontal gene transfer (HGT) within the lung and gut microbiomes of severe pneumonia patients; (3) identifying the disparities in host characteristics and microbiomes in CAP, HAP, and VAP; (4) charting the lung virome landscape; (5)  predicting the 28-day mortality for ICU patients based on lung microbiome and (6) dynamic monitoring of lung microbiome to predict secondary bacterial infection.
A timeline of the cohort has been established based on the current progress, detailed in the "Progress of the cohort" section (Figure 1C).

RECRUITMENT
The cohort commenced recruitment in September 2023 (Figure 1C).At the time of writing the announcement (February 2024), we have successfully recruited 286 patients from 15 hospitals (see "Progress of the cohort").We aim to enroll up to 2000 ICU patients with severe pneumonia to delineate the infections caused by various pan-kingdom pathogens.Since a definitive diagnosis of severe pneumonia may not be possible at enrollment, patients with suspected lung infections will also be included (see "Definitions" in Supporting Information).Following enrollment, Dr. Lingtong Huang, the project leader, and the principal investigator at each medical center will independently conduct secondary diagnoses for all patients.In cases of diagnostic disagreement, Dr. Hongliu Cai from the First Affiliated Hospital of Zhejiang University School of Medicine will participate in the final evaluation.All enrolled patients are fully informed and provide signed informed consent, without receiving additional financial compensation.For unconscious patients, the consent form will be signed by his/her next of kin.The inclusion and exclusion criteria are outlined below: Inclusion criteria: Patients with newly acquired lung infections that meet one of the following criteria: (1) receiving invasive or noninvasive mechanical ventilation for acute respiratory failure, with positive end-expiratory pressure (PEEP) ≥ 5 cm; (2) receiving high-flow oxygen therapy with a fraction of inspired oxygen (FiO 2 ) ≥ 50% and arterial oxygen partial pressure (PaO 2 ) to FiO 2 ratio <300.Patients with suspected lung infections (see "Definitions" in Supporting Information).
Exclusion criteria: (1) patients with expected length of stay in ICU less than 1 day; (2) hospitalized in other ICUs for more than 7 days before being transferred; (3) pregnant; (4) under 18 years old.

BIOLOGICAL SAMPLING PROCEDURES
We will collect BALF, sputum or endotracheal aspirate (ETA), stool or rectal swab, whole blood, and serum from patients on Days 1, 4, 7, 14, and 21 after admission, when applicable (Figure 1B and Table S2).Sample collection may occur within a 24-h window around the planned sampling time.For example, if Day 4 sampling is impractical, collection may be rescheduled for Days 3-5 (Day 4 ± 24 h).To minimize the batch effect of supplies, all participating medical centers will use standardized nucleic acid preservation solutions, sampling tubes, and saline solutions supplied quarterly by the First Affiliated Hospital of Zhejiang University School of Medicine.Given the typical low microbial yield in BALF, each hospital will establish negative environmental and instrument controls.Sample collection will conclude if a patient is discharged or transferred from the ICU.The detailed protocols for the collection of various biological samples and negative controls, along with the rest of the analytical protocols can be found in Supporting Information.
By the end of February 2024, the cohort had amassed 502 BALF, 484 sputum, 386 stool, 282 whole blood, and 292 serum samples (Figure 2F).Concurrently, 30 negative control samples from each batch of consumables and saline solutions were also collected.The BALF specimens were sent to the National Institute of Pathogen Biology, CAMS & PUMC for aliquoting, using phosphate-buffered saline as a negative control during this process.

DISCUSSION
This study aims to utilize a multicenter, large-cohort, longitudinal approach to robustly characterize the severe pneumonia microbiome across Chinese ICU settings, integrating deep sequencing and advanced analytics to explore the lung microbiome's diversity and its role in respiratory diseases.
The study's limitations include the potential lack of generalizability beyond Chinese ICU patients.The absence of a nonsevere pneumonia control group restricts comparative analyses.An ongoing healthy respiratory virome project, led by Professor Chao Jiang, can provide a healthy control for this study (supported by the National Natural Science Foundation of China, grant no.82341109).The study's observational nature prevents establishing direct causality between microbiomes and severe pneumonia outcomes.Future studies should include diverse demographics [12], control groups, and experimental validation.Antibiotic use and varying comorbidity and treatment profiles present additional confounding factors, although extensive data collection and propensity score matching may mitigate these effects.Additionally, given the limited research on how environmental factors affect severe pneumonia [13, 14], we plan to incorporate exposome monitoring into the MASS cohort to deepen our understanding [15-17].

CONCLUSION
In conclusion, the MASS cohort establishes a rich resource for exploring the microbiome's influence on severe pneumonia and host susceptibility, integrating metagenomic, transcriptomic, and phenomic analyses.This cohort will significantly enhance our understanding of emerging pathogens, risk factors for poor prognosis, and determinants of survival among ICU patients.

F
I G U R E 1 Study design of the MASS cohort.(A) Geographical distributions of the 19 medical centers across 10 cities (red) in three provinces of China.Fourteen centers are located in seven cities of Zhejiang province (zoomed-in view on the right), four in Henan province, and one in Shaanxi province.(B) Overview of clinical data and biological sample collection.(C) Timeline of MASS cohort.

CORRESPONDENCE | 3 2
Progress in patient recruitment and biological sampling for the MASS cohort.(A) Geographic distribution of the 286 patients' residences.ZJ represents the patients admitted to medical centers in Zhejiang province, whereas non-ZJ represents those admitted to medical centers in Shaanxi and Henan provinces."+" indicate the location of the medical centers.The cities where medical centers are located are painted orange.(B) Number of patients diagnosed with community-acquired pneumonia, hospital-acquired pneumonia, and ventilator-associated pneumonia in each medical center (bar plots) and the overall proportions of different pneumonia (pie chart).The "undefined" group includes four patients with trauma-associated lung injury and four patients who were considered highly unlikely to have a lung infection.(C) Density plots of patients' age and body mass index.The vertical dashed line denotes the median.(D) The distribution of patients' genders, the proportion of patients having a drinking and smoking history, and the proportion of patients receiving extracorporeal organ support.(E) Distribution of length of stay in intensive care unit (ICU) and patients' sources and outcomes.The light blue dots indicate the time of admission to the ICU, and the blue circles indicate the time of discharge.The "ongoing ICU treatment" group includes hospitalized patients as of April 1, 2024.(F) The number of biological samples collected at five-time points.