Presented at the Neonatal Society 2018 Autumn Meeting.
Montaldo P1, Kaforou M1, Markati D1, Swamy R1, Mendoza J1, Oliviera V1, Bandiya P2, Benakappa N2, Mangalabharathi S3, Mondkar J4, Manerkar S4, Sreelar A3, Ramji S5, Shankaran S6, Herberg J1, Thayyil S1
1 Imperial College London, UK
2 Indira Gandhi Institute of Child Health, Bangalore, India
3 Madras Medical College, Chennai, India
4 Sion Medical College, Mumbai, India
5 MAMC, New Delhi, India
6 Wayne State University, USA
Background: Most putative neuroprotectants, including cooling therapy, are effective only when initiated within few hours of birth, hence early identification of ‘at risk’ encephalopathic infants is vital. We have recently shown that babies with neonatal encephalopathy (NE) have a different gene expression profile, when compared with age matched healthy newborn babies1. In this study we examined if babies who later develop adverse neurodevelopmental outcome after NE have a unique host gene expression (transcriptomic) profile at birth.
Methods: We performed next generation sequencing on the RNA extracted from whole blood collected from the first 47 babies (aged <6 hours) with moderate/severe NE recruited to the HELIX (Hypothermia for Encephalopathy in Low and Middle- Income Countries) trial2. We excluded babies with culture positive sepsis from the analysis. We used principal component analysis (PCA) to determine the significant sources of variability and fitted quasi-likelihood negative binomial generalized log-linear models to identify significantly differentially expressed genes, which were then subjected to pathway analysis (Ingenuity Pathway Analysis). We adjusted the analysis for the randomisation allocation (masked as ‘A’ and ‘B’). Adverse outcome (primary outcome) was death or moderate or severe disability (motor disability and Bayley III) at 2 years. Imperial College REC and ICMR approved the study.
Results: The primary outcome was available from all 47 babies; 2 were excluded due to blood culture positive sepsis. Of the remaining 45 babies, 23 (51%) had adverse outcome. PCA (Figure 1) showed separate clusters in babies with adverse outcome (red) and those with good outcome (green). Regression model identified 320 genes that discriminated between the NE babies with adverse from the ones with good outcome. The most significant pathways identified were related to Melatonin (-log p value 2.61) and T cell receptor signalling (-log p value 2.4). The upregulated genes are shown in red and the downregulated genes are shown in green (Figure 2). Heatmap (Figure 3) showed separate clusters in babies with adverse outcome (red) and those with good outcome (green).
Conclusion: These preliminary data suggest that babies with NE who have adverse outcome may have a unique host RNA signature at birth. Gene expression profiling is a promising disease stratification tool and may provide insights into the underlying molecular and cellular mechanisms.
Corresponding author: p.montaldo@imperial.ac.uk
References
1. Montaldo et al. Whole Blood Gene Expression Reveals Specific Transcriptome Changes in Neonatal Encephalopathy. Neonatology 2018; 115(1): 68-76
2. Thayyil et al. Hypothermia for encephalopathy in low and middle-income countries: study protocol for a randomised controlled trial. Trials 2017; 18: 432