Precision Medicine in Childhood Asthma

Asthma is a complex and multifactorial respiratory disease with a high prevalence in the pediatric population. Variation in treatment response to asthma therapies has been described among patients, and difficult-to-treat asthma carries both high healthcare and socioeconomic burden to the patients and society. Omic studies can be used to discover the molecular mechanisms underlying asthma susceptibility and treatment response, contributing to a better knowledge and definition of asthma pathogenesis and therefore, to the development of precision medicine. This entry aims to summarize the recent findings of omic studies of treatment response in childhood asthma. Between 2018-2019 a total of 13 omic studies has been performed involving genomics, epigenomics, transcriptomics, metabolomics, and the microbiome. These have been focused on the response to three common asthma medications: short-acting beta agonists, inhaled corticosteroids, and leukotriene receptor antagonists. Novel associations of different biomarkers with asthma treatment response have been described. However, stronger evidence and more consistent results are required to implement these molecular biomarkers into clinical practice by establishing the most appropriate therapy for each patient.


Introduction
Pharmacogenomics investigates the association of genetic markers across the genome with treatment response. This approach usually focuses on single nucleotide polymorphisms (SNPs), which are changes in the DNA sequence affecting one nucleotide shared by >1% of the population. This type of marker is usually coinherited with other genetic variants due to the linkage disequilibrium (LD) patterns allowing the indirect study of millions of genetic variants that are inferred from hundreds of thousands of SNPs assessed by genotyping arrays. Thus, common genome-wide genetic variation can be studied without any prior hypothesis by genome-wide association studies (GWAS) . Recently, whole-genome sequencing (WGS) emerged as a high-resolution method to study both common and rare genetic variation. Although WGS detects genetic variation not tackled by genome-wide genotyping arrays, its high cost has still limited its application to pharmacogenomic studies .
Recent pharmacogenomic studies of childhood asthma have not been limited to European-descent populations as in the past , but they have mostly analyzed two minority populations from the United States with a high prevalence of asthma and low treatment response : African Americans and Hispanics/Latinos. Specifically, several publications within the period reviewed have been focused on two studies: Study of African Americans, Asthma, Genes and Environments (SAGE) and the Genes-Environment and Admixture in Latino Americans (GALA II) (Table 1). In these studies, genomic analyses of SABA treatment response were carried out by means of GWAS using genotyping arrays and WGS data . Additionally, given that African Americans and Hispanics/Latinos are admixed populations, admixture mapping can be applied in order to identify genomic regions in which local ancestry is associated with pharmacogenetic traits . This complementary approach is possible because of the differences in allele frequency of the SNPs depending on their ancestral background . These studies have been focused on bronchodilator drug response (BDR), a measurement of the change in lung function after SABA administration , and also on the presence/absence of severe asthma exacerbations despite ICS treatment as a proxy of ICS response . analyzing imputed data from genotyping arrays. A SNP from 9q21 (rs73650726) was genome-wide significantly associated with BDR exclusively in African-admixed populations. Moreover, a trans-ethnic meta-GWAS across African American and Hispanic/Latino children and youth (n = 2779) identified genome-wide associations of three SNPs from the protein kinase cGMP-dependent 1 (PRKG1) gene with SABA response. Interestingly, PRKG1 is a biologically plausible gene for SABA response given that it encodes a protein involved in the nitric oxide/cGMP signaling pathway, acts as an important regulator of airway inflammation in response to SABA, and participates in smooth muscle relaxation .
Furthermore, PRKG1 has been associated with lung function measurements and asthma risk .
However, these results need to be replicated in independent studies. association of a SNP from the intergenic region of the genes encoding for the apolipoprotein B mRNA editing enzyme catalytic subunits 3B and 3C (APOBEC3B-APOBEC3C) was revealed, a result that was replicated in European children with asthma (n = 1697). However, this SNP did not reach the genomewide significance threshold in a meta-analysis across all populations. Despite this, a consistent association of the SNP from the APOBEC3B-APOBEC3C intergenic regions was found with the change in the forced expiratory volume in the first second (FEV ) after ICS treatment during 6 weeks in a subset of patients with such data available (n = 166), as an alternative measurement of ICS response. APOBEC3B and APOBEC3C are subunits of a cytidine deaminase that carries out RNA editing and participates in the immune response to viruses. In the same study , an attempt of replication of genomic regions previously associated with ICS response was performed. The association of the region that included the L3MBTL4 and ARHGAP28 was validated, which has also been associated with post-bronchodilator lung function .
Epigenetics is the study of the mechanisms involved in gene expression regulation without modifying the DNA sequence. These include DNA methylation (DNAm), microRNA (miRNA) regulation, and histone modifications. Nowadays, whole-genome epigenetic changes can be analyzed through high-throughput techniques. DNAm is the most studied epigenetic modification, involving the methylation of a cytosine base followed by a guanine (CpG site) .
Two epigenome-wide association studies (EWAS) were performed during the period reviewed evaluating the effects of ICS response on DNA methylation patterns in peripheral blood cells (PBCs) . In the first study , non-Hispanic whites from the Childhood Asthma Management Program (CAMP, n = 154), Europeans from the Children, Allergy, Milieu, Stockholm, Epidemiology (BAMSE, n = 72), and Hispanic/Latinos from the Genetic Epidemiology of Asthma in Costa Rica Study (GACRS, n = 168) were analyzed. Treatment response was measured as two outcomes related to asthma exacerbations in the past 12 months while the patients were treated with ICSs. Hypomethylation of a CpG site near interleukin 12B (IL12B) was protective for severe exacerbations in a meta-analysis across non-Hispanic whites and Hispanic/Latinos. and this hypomethylation was correlated with lower IL12B expression in blood cells in one of the studies. IL12B encodes a subunit of two cytokines (IL-12 and IL-23) involved in immune response and airway hyperresponsiveness, and its expression level has been associated with corticosteroids response in bronchial biopsies from asthma patients . Moreover, in the same study, Wang et al. found 13 CpG sites that were significantly associated with the absence of OCS use in a meta-analysis across non-Hispanic whites (n = 322). An interaction analysis identified that hypermethylation of a CpG near the genes encoding for the cortistatin (CORT) and centromere protein S (CENPS) was associated with the absence of OCSs use, and also with higher CORT expression in CAMP . Interestingly, cortistatin is a peptid involved in the anti-inflammatory process regulated by the hypothalamic-pituitary-adrenal axis and regulates endogenous corticosteroid levels .   (Table 3). The strongest association was described for miR-206, which had higher serum expression levels in non-exacerbators compared to exacerbators. In addition, a predictive model for asthma exacerbations based on three miRNAs (miR-206, miR-146b-5p, and miR-720) and clinical variables had an improved prediction area under the receiver characteristics curve (AUC) than a model exclusively based on clinical parameters (0.81 vs 0.67). Strikingly, these three miRNAs have been related to asthma in cellular and animal models , and miR-146b-5b and miR-206 had also been associated with baseline FEV / F V C . Moreover, four biological pathways regulated by the three miRNAs were revealed by Kho et al., including two previously associated with asthma .  [38] [39] [40] [41] [42] [43] [44] Encyclopedia 2020 doi: 10.32545/encyclopedia202005.0007.v3 in the antiviral response regulated by one of the interferon pathways .
Metabolomics provides the profile of the whole metabolite composition (metabolome) in biological samples by means of high-throughput analytical techniques, such as mass spectrophotometry, nuclear magnetic resonance, or spectroscopic methods. interestingly, metabolomics can be performed in noninvasive samples such as exhaled air (breathomics) and this recent omic has been applied to asthma with promising results .
Kelly et al. examined the interaction of age and 501 serum metabolites on BDR after albuterol administration in blood samples from children with asthma from CAMP at three-time points every four years, with mean ages of 8.8 (n = 560), 12.8 (n = 563), and 16.8 (n = 295), respectively . A total of 39 metabolites, mainly lipids, showed a nominal interaction with age on BDR, being the strongest interaction observed for the 2-hydroxyglutarate. Results of this metabolite were replicated in Hispanic children with asthma from GACRS (n = 320), with a mean age of 9.1 years. However, these results would not be considered significant after multiple comparison adjustment.
Kelly et al. also conducted a multi-omic study of lung function in Hispanic/Latino children with asthma (n = 325) simultaneously analyzing gene transcripts and metabolites with WGCNA . Four transcript modules and five metabolite clusters were found to be related to lung function and seven of them were found to interact among them. Interestingly, one transcriptomic module was associated with BDR and with a lipid metabolomic module, and had as a hub the well-known pediatric-asthma loci ORMDL3, which is involved in sphingolipid biosynthesis . Furthermore, a SNP from ORMDL3 was found to be an eQTL of ORMDL3 and was associated with several lipids included in the metabolomic module. Reinforcing these results, findings were replicated in children with asthma from CAMP (n = 207).
The study of the microbiome involves the analysis of the genetic composition of the microbial communities, known as microbiota. fully explain the differences in treatment response in childhood asthma. This is expected since asthma is a complex and multifactorial disease, and different phenotypes may not share the same underlying mechanisms. However, further advances in omic studies will aid to understand how different biological layers are implicated in disease pathogenesis and also how they could interact among them to determine the response to pharmacological therapies. family with sequence similarity 129 member A (FAM129A) as an asthma steroid response modulator. Journal of