Prevention of Chronic Lung Disease (CLD - Prevention Study)

Background Chronic Lung Disease (CLD) is one of the most common complications of perinatal lung injury which represents a systemic condition with long term sequelae, such as persistent pulmonary dysfunction, asthma-like symptoms, and BPCO. CLD recognizes many prenatal risk factors, including maternal smoking, chorioamnionitis, and intrauterine growth restriction (IUGR), besides postnatal risk factors such as hyperoxia, parenteral nutrition, inflammation and mechanical ventilation. Exposure to inflammation in utero alters neonatal immune development and predisposes the fetal lung to a dysregulated prolonged response to invasive mechanical ventilation and supraphysiological oxygen. Responses to injurious (and protective) influences early in life are modulated by genetic and epigenetic mechanisms, that could result in permanent structural changes with significant consequences in adulthood. IUGR pregnancies are characterized by increased oxidative stress and IUGR infants are at high risk of cardiovascular disease, metabolic syndrome, lung dysfunction and chronic kidney and respiratory diseases in adulthood Despite the constant increase in knowledge of the mechanisms leading to the progression of lung damage, so far, no effective management has been developed that would lead to CLD prevention.

Aims

1. prospectively identify potential biomarkers able to predict the severe course of pulmonary function in the first 12 months of life and realize a new profile to early identify high risk newborns.
2. detect genetic variance causation model (by MiSeq Illumina platform) correlating with severe pulmonary dysfunction and asthma development.
3. detect MicroRNAs as well as anti- and pro-inflammatory cytokine variations (MIP-1a, MCP-1, IL-8, TNF-a, IFN-g, IL-10) correlating with the severity of pulmonary dysfunction in the first 12 months of life and the risk of asthma development.

Experimental design This is a multicenter longitudinal study, whose primary aim is to identify a new high-risk neonatal profile for Chronic Lung Disease (CLD) development, starting from perinatal life. In order to reach this objective all participant Units will perform vaginal swab to the enrolled mothers, will collect placenta and cord blood at birth and peripheral blood, bronchoalveolar lavage fluid (BALF), meconium and feces samples of all enrolled newborn infants. The study will begin with the evaluation of prenatal risk factors.

After delivery the placentas will be fixed in 10% buffered formalin. Subsequently macroscopic and microscopic analyses will be performed according to the Amsterdam placental workshop group consensus statement. A vaginal swab will be collected at the time of delivery for all the enrolled women. Three ml of cord blood (artery and/or vein) will be performed for analysis of miRNA, genetic variance and cytokines. In intubated infants only, BALF sample will be obtained by instilling 1 ml/kg of 0.9% sodium chloride in the endotracheal tube and suctioning the fluid into a sterile mucus trap. The first BALF sample will be collected within the first 24 h of life, whereas further BALF samples will be collected at 7 and 28 days of life in newborns who will be still intubated. First meconium sample after birth and additional feces samples will be collected at 7, 28 days of life, 6 and 12 months of life. Vaginal swabs, placenta, BALF, meconium and feces samples will be stored at -80°C until further processing. For each sample, bacterial DNA extraction will be performed in a strictly controlled level-2 biological safety workplace using DANAGENE MICROBIOME DNA kits (Danagen-Bioted) according to manufacturers instructions.

In cord serum at birth (artery) and at 48 hours of life, 8-10 candidate miRNAs, in particular miR-451, miR-29b, and miR-16 will be studied as it has been shown that they have an inhibitory effect on angiogenesis through vascular endothelial growth factor (VEGF) suppression and miR is associated to genetic susceptibility to CLD. The genetic variance causation model will be performed in all enrolled newborns at 48 hours of life (at least 500 μL of whole blood) using a MiSeq Illumina platform allowing application of Next Generation Sequencing technologies.

To detect anti- and pro-inflammatory cytokine variations correlating with the severity of disease over time, cord blood and peripheral blood samples (2 ml into an EDTA tube) will be obtained from the included infants at the admission (within 48 hours of life), at 7 and 28 days of life, at 6 and 12 months. Both anti- (INF-y) and pro-inflammatory (IL-6 and NGF) cytokines levels will be investigated. All samples will be stored at -80 °C until testing. Cytokines will be measured with a commercially available ELISA kit according to manufacturer's instructions.

The investigators will also try to explore the association between mother and placenta data with newborns analysis during the first days of life. Upon newborn admission to Neonatal Intensive Care Unit, considerable attention will be paid to ascertaining the degree and variations in hemo-oxygenation, as well as the presence of infection that is at the origin of Oxidative Stress (OS) and OS- related diseases. The following specimens will be collected: 1 ml of peripheral blood (within 48 hours of life). Additional peripheral blood samples will be collected at 7 and 28 days of life, at 6 and 12 months of life.

Oxidative stress (OS) and lipid mediators involved in OS will be evaluated both by biomarkers of oxidative protein damage (Advanced Oxidation Protein Products, AOPP) and lipid peroxidation (Isoprostanes, IsoPs). A further evaluation of oxidative stress profile will be performed by measuring non-enzymatic antioxidant molecules (vitamin E; glutathione, GSH, and ascorbic acid, AA) and enzymatic antioxidant molecules (superoxide dismutase, SOD, catalase, CAT, and glutathione peroxidase, GPx). To this end high-performance liquid chromatography (HPLC) and gas chromatography interfaced mass spectrometry (GC-MS) will be employed: HPLC to detect vitamin E, glutathione, GSH, and AA.

The final step will be prospectively identify a new profile of high risk newborns for severe pulmonary dysfunction in the first 12 months of life, at higher risk of developing asthma and COPD in later age. A long term outcome of all enrolled babies will be performed to unravel the relationships between placenta analysis, miRNA expression, inflammation and microbiota - as part of the in utero exposome (chorioamnionitis or fetal growth restriction) and oxidative stress with inflammation - as part of postnatal factors (invasive respiratory support, Patent Ductus Arteriosus, pneumonia, sepsis) - with the pulmonary outcome in terms of abnormal pulmonary function at 6 and 12 months of age.

At 6 and 12 months of age, the infants will receive clinical and pulmonary evaluation. Pulmonary function test (PFT) will be performed in all enrolled infants and will include Tidal breathing analysis and Nitrogen Washout test. The infants will be placed in the supine position, during quiet, natural sleep, according to American Thoracic Society/European Respiratory Society recommendations with measurement of lung volumes, flow, functional residual capacity, Lung Clearence Index 5% (LCI 5), Lung Clearence Index 2.5% (LCI 2.5), time to peak tidal expiratory flow/expiratory time ratio (tPTEF/tE). Lung ultrasound will be performed to rule out lung abnormalities. After allowing adaptation to the mask, tidal breathing, flow volume loops for >2 minutes or >20 artifact free breaths will be recorded. PFT will be made in order to relate the functional respiratory data with those of generated biochemical profile. Any possible relationships among pulmonary function, biochemical markers just before birth and in the infants at 6-12 months of age, and anthropometric measurements will be studied. Pulmonary development and function during extrauterine life will be related with prenatal measurements. Data in appropriate for gestational age (AGA) infants, small for gestational age (SGA) infants and large for gestational age (LGA) infants will be compared.

Methods of data collection Several types of data will be collected including quantitative, qualitative, generated from imaging, tissue and blood samples. Clinical data for Aim 1, 2 and 3 will be collected and managed using REDCap electronic data capture tools hosted at UO1- Fondazione Policlinico Universitario A. Gemelli-IRCCS (https://redcap-irccs.policlinicogemelli.it/).

A dedicated electronic case report form (eCRF) will be developed. Pseudo-anonimyzed data will be collected from the units involved. The Investigator will be responsible to ensure that the eCRF is properly and completely filled in. Sources of clinical information include the physicians patient record, hospital notes, original laboratory records, pharmacy records, results of ultrasound examination etc.

Statistic plan Primary endpoint of the study is to identify whether clinical, demographic, laboratory, and other parameters are able to predict the development of chronic lung disease at 36 weeks of PMA and/or of abnormal pulmonary function at 12 months of age. Assuming an acceptable accuracy level, with an Area Under the ROC Curve of at least 0.75 ± 0.15, a sample size of 42 patients is required.

Statistical analysis The sample will be described in its clinical and demographic characteristics using the appropriate descriptive statistics indices. In depth, qualitative data will be expressed as absolute and relative percentage frequency, while quantitative variables as either mean and standard deviation (SD) or median and interquartile range (IQR), depending on the case. To verify the Gaussian distribution of quantitative variables the Shapiro-Wilk test will be applied. Differences between groups at baseline will be evaluated, with regard to qualitative data, by either the Chi-Squared test or the Fisher-Freeman-Haltons exact test, as appropriate. Quantitative data will instead be compared using Student's t test for independent samples or Mann Withney's nonparametric U test, depending on the data distribution. Violin plots will be used to graphically represent significant or clinically relevant differences. The evaluation of the difference in terms of primary outcome at 12 months will be evaluated by Kaplan-Meier survival analysis. In particular, the log-rank test will be applied, and appropriate cumulative incidence curves will be drawn. In order to assess potential predictors of the primary outcome at 12 months, uni- and multivariable Cox regression models will be fitted. In depth, the potential predictors of the outcome will be evaluated by means of ordinary proportional hazard Cox regression models, and the Hazard Ratios (HR) and the 95% confidence intervals (CIs) will be consequently reported. The proportionality of the hazard functions will be evaluated by visual inspection of the hazards and Schoenfeld residual plots. In case of doubtful proportionality, Cox weighted regression models will be fitted. Predictors to be included in the multivariable model will be selected based on the univariable analysis (p<0.05 or suggestive, i.e. 0.05 p <0.10) and expert opinion, consistent with the rule of at least 10 events by outcome variable, and TRIPOD recommendations. The performance of the model will be evaluated by several indices, such as Cindex, Somers Dxy-rank correlation, Nagelkerke's R2 value, calibration intercept and slop. The C-index can be interpreted as an AUC, i.e. a measure of the accuracy of the model. Statistical significance is set for values of p<0.05. Suggestive p values (0.05, p < 0.10) will also be reported. Statistical analyses will be conducted using STATA (StataCorp, USA) and R (https://www.r-project.org/).