Depending on age and severity, the GINA guidelines recommend using ICS monotherapy, or in combination as the preferred controller in addition to a rescue inhaler. ICS are commonly prescribed to treat asthma and, to an extent, COPD. Highly effective in reducing inflammation, ICS can significantly improve asthma symptoms and reduce morbidity in both children and adults. The mechanism of action involves decreasing inflammatory mediators such as mast cells, cytokines, and enzymes such as COX-2. Inhalation routes allow for minimal systemic absorption; thus, efficacy and safety are robust. However, side effects are not negligible, especially when used in children for an extended period. Osteoporosis, glaucoma, and metabolic disturbances are possible adverse events of systemic exposure (Figure 2).

Tantisira conducted a GWAS in 2011 and identified the glucocorticoid-induced transcript 1 (GLCCI1) gene and the SNPs, rs37972 and rs37973 (Table 5) [51]. Non-Hispanic white children selected from the CAMP study were given budesonide, and the response was measured by the change in FEV1 from baseline. Although little is known about GLCCI1, those homozygote for the mutant T allele at SNP, rs37972, yielded a poorer response compared to those who were heterozygote (CT) or homozygote (CC) (OR, 2.36; 95% CI, 1.27–4.41) [51]. Additionally, linkage disequilibrium was found with the SNP, rs37973 (r² = 0.99). Individuals that were homozygote for the wild-type A allele had a better response to budesonide compared to those homozygotes for the mutant G allele (3.2 ± 1.6% vs. 9.4 ± 1.1% increase in FEV1) [51]. Similar findings were seen in a trial conducted by Thompson et al. [52]. Included in that study were 402 European children on long-term ICS treatment (>6 months). Higher steroid doses and increased hospitalization were seen in SNP, rs37973, homozygous for the mutant allele compared to those heterozygous or homozygous wild-type [52]. Unlike Tantisira [51], Vijverberg saw no association with individuals having SNP, rs37972 and ICS response in Northern European children [53]. Additionally, Vijverberg utilized hospital visits and oral steroid use as markers for exacerbation and concluded that the mutant T allele was not associated with an increased risk when treated with budesonide [53]. In contrast, Huang et al. found associations in the GLCCI1 gene in 263 Chinese children. SNPs, rs37969 GG, rs37972 CC, and rs37973 AA produced favorable changes measured by maximal mid-expiratory flow (p ≤ 0.05) [54]. Stress-induced phosphoprotein 1 (STIP1) was also explored in this study and was shown to suppress inflammatory mediators. Although a significant association between SNP rs2236647 CC and the risk of developing asthma was found, there was no association with any SNP and ICS response [54].

It was hypothesized that histone deacetylase I and 2 (HDAC1, HDAC2) play a positive role in airway responsiveness and inflammation. Kim studied 35 adults and 70 children with the HDAC1 SNP, rs1741981 [55]. Bronchodilation was measured as a % change in FEV1 from baseline [55]. Adults with the CC genotype produced diminished responses compared to the CT or TT genotypes when administered systemic corticosteroids (p = 0.018). The same was true for the children given inhaled corticosteroids. Those with the CC genotype had a FEV1 of 14.1% compared to either the CT or TT (19.4%, p = 0.035).

The corticotropin-release hormone (CRH) is a known mediator to stress and its corresponding receptor, corticotropin-releasing hormone receptor 1 (CRHR1), is hypothesized to have a role in ICS response [56]. In comparison to the unfavorable responses seen in the GLCCI1 variants, Tantisira observed an increase in response in those with the CRHR1 variants among three cohorts of both children and adults treated with an ICS. Response was measured by dFEV1 over a span of eight weeks in the CAMP replication study. Those with the rs242941 mutant TT genotypes displayed a change of 17.80 + 6.77 in FEV1 compared to a change of 7.57 + 1.50 in those with the wild-type CC. This was not seen in the initial adult study treated with flunisolide. SNP, rs1876828 variant, was also associated with triamcinolone responses in the adult replication study but was statistically insignificant in children [54].

Variants in the T-box transcription factor (TBX21), which encodes for transcription factor T-bet, also displayed improvements in ICS response. Children with glutamine at amino acid position 33 in place of histidine (rs2240017) displayed better PC20 [57]. PC20 is the provocative concentration resulting in a 20% drop of FEV1 post methacholine challenge. After four years, PC20 for the 33Q individuals given budesonide was measured to be 27.7 mg/mL, placing them in the non-asthmatic range [57]. Tantisira also conducted a study with a focus on the Fc fragment of IgE low affinity II receptor (FCER2) gene [58]. FCER2 encodes for CD23, causing IgE mediated responses which have been associated with severe asthmatic exacerbations. In this study, 311 Caucasian children were assigned to inhaled budesonide and observed over four years. The SNPs identified to be associated with increased IgE levels and exacerbations were: rs4996974, rs7249320, and rs28364072. Children with the SNP, rs28364072, homozygous for the mutant C allele were at a higher risk of severe exacerbations in both African American and Caucasian populations (HR 3.08 and 3.95, respectively) [58]. Koster replicated these findings in two pediatric cohorts treated with ICS. Children’s homozygotes for the variant C allele had a higher risk of severe exacerbations (OR 2.38, 95% CI 1.47–3.85, p = 0.0004), increase in hospital visits (OR 1.91, 95% CI 1.08–3.40, p = 0.03), and were more likely to use higher ICS doses (OR 2.46, 95% CI 1.38–4.39, p = 0.002 [59]. These findings support the need for earlier detection of FCER2 polymorphisms to identify those resistant to steroid treatment and provide timely anti-IgE treatment.

Located on chromosome 5q31-q32, nuclear receptor subfamily 3, group C, member 1 (NR3C1) is a protein-coding gene for glucocorticoid receptors. Keskin studied 82 children with a mean age of 9.6, given inhaled fluticasone propionate specifically evaluating the role of NR3C1 [60]. Of the 82 children, 26 had the rs41423247, GG genotype and displayed a better response compared to those with the CG or CC variations, measured by FEV1 (24.2% vs. 7.9%, p = 0.006) [60]. Stockman also conducted a study involving 734 Caucasian children receiving inhaled fluticasone propionate. Variations among cytochrome p450 (CYP) 3A4, 3A5, and 3A7, which are involved in fluticasone metabolism, were assessed for symptom control using the asthma control score. CYP3A5 and CYP3A7 were not found to have an association, however, CYP3A4 *22 children displayed better asthma control due to the reduced activity of the metabolic enzyme, leading to increased therapeutic outcomes [61]. These findings were not seen in a follow-up study of 64 children treated with beclomethasone [62]. They observed that children with CYP3A5 *3/*3, commonly found among Caucasian populations, displayed better asthma control, compared to either *1/*1 or *1/*3 variations [62]. Pharmacogenomics testing on these CYP enzymes has the potential to provide guidance on future regimens. Children with the CYP3A4 *1/*1 or *1/*3 variation may be advised to use a non-beclomethasone medication, such as budesonide, as their initial ICS. Similarly, those with the CYP3A4 *22 variation may be advised to start therapy with fluticasone.

Leukotrienes are involved in inflammatory processes and bronchoconstriction, both of which are implicated in asthmatic symptoms. Leukotriene modifiers are considered second line after ICS and LABA for chronic asthma according to the GINA guidelines [3]. Administered orally, these LTMs include zileuton, montelukast, and zafirlukast which are subsequently classified into leukotriene synthesis inhibitors or leukotriene receptor antagonists. Zileuton inhibits the 5-lipoxygenase (5-LO) enzyme, halting the conversion of arachidonic acid into LTA4. Montelukast, zafirlukast, and pranlukast are cysteinyl-leukotriene receptor (CysLTR) inhibitors, antagonizing the effects of LTC4, LTD4, and LTE4 [63,64].

Similar to bronchodilators and ICS, limitations of leukotriene modifiers are highlighted by genetic variations, emphasizing inter-patient variability leading to exacerbated symptoms [65,66]. In addition, nearly half of the side effects in children taking a LTM are attributed to psychiatric disturbances such as hallucinations and agitation [64]. Genes such as arachidonate 5-lipoxygenase (ALOX5) and cysteinyl-LTs (CysLTs) previously identified and replicated in adult studies have not yet been fully confirmed in children. Identifying variants associated with positive leukotriene modifier response can further benefit children who cannot gain symptom control with ICS or beta agonists.