Predicting the course of asthma from childhood until early adulthood

Purpose of review To communicate recent insights about the natural history of childhood asthma, with a focus on prediction of persistence and remission of childhood asthma, up to early adulthood. Recent findings Lung function around the age of 8–9 years is the strongest predictor: obstructive lung function predicts asthma persistence up to early adulthood, whereas normal lung function predicts remission. The ability to predict asthma remission improves when lung function is combined with blood eosinophil levels and degree of bronchial hyperresponsiveness. Interventions, such as inhaled corticosteroids and immunotherapy do not appear to alter the course of asthma. Epigenetic studies have revealed potential novel biomarkers of asthma remission, such as micro-RNA patterns in blood. Specifically, lower serum levels of mi-R221-5p, which is associated with lower IL-6 release and eosinophilic inflammation, predict remission. Higher levels of blood DNA-methylation of a CpG site in Peroxisomal Biogenesis Factor 11 Beta were associated with asthma remission. Summary Lung function, allergic comorbidity and polysensitization in childhood predict the course of asthma. Recent epigenetic studies have provided a better understanding of underlying pathological processes in asthma remission, which may be used to improve prediction or develop novel treatments aimed at altering the course of asthma.


INTRODUCTION
The clinical syndrome of childhood asthma is characterized by common symptoms, such as wheezing, shortness of breath, chest tightness and cough in combination with variable expiratory flow limitation and bronchial hyperresponsiveness (BHR) [1,2]. It has become increasingly clear that these symptoms arise from different (inflammatory) subtypes, making asthma a syndrome rather than one disease [3]. Variation exists not only in asthma subtypes but also in the course of asthma, from patients that experience complete remission to those with persistent, sometimes even life-long symptoms.
Although considered a chronic disease in childhood, the course of asthma varies. Almost 80% of patients with asthma experience symptoms during the first 6 years of life [4,5]. In children with asthma at the age of 7 years, 67-75% will be symptom free as adults [6][7][8]. In contrast, research suggests that approximately 3-5% of the general population will experience persistence of symptoms from childhood until early adulthood, accompanied by low lung growth [6,[8][9][10]11 & ]. In a population-based birth cohort followed until age 24 years, 6.4% of all children had adolescent-onset asthma with symptom onset after age 8 [11 comorbidity and baseline disease severity are important predictors of the course of asthma [12][13][14].
Investigation of factors associated with pediatric asthma persistence and remission is important as it not only provides modifiable characteristics that may be used as targets for intervention but also for guidance as to which individuals may or may not require more rigorous clinical follow-up. This review aims to summarize the latest insights into important factors related to the course of asthma from childhood to early adulthood.

METHODS
We performed a literature search in PubMed to identify studies investigating predictive factors associated with the course of asthma from childhood until early adulthood (Table 1). Backward citation was performed in all selected articles. All included articles were peer-reviewed and published in English. We prioritized articles published in the last 18 months. On the basis of retrieved literature, we selected the following predictors for discussion: lung function, BHR, sex, treatment, allergic comorbidity, genetics, and epigenetics (microRNAs and DNA methylation).

THE COURSE OF ASTHMA FROM CHILDHOOD TO EARLY ADULTHOOD
Approximately 25-40% of all children wheeze during the first 7 years of life and six longitudinal patterns of wheeze in the first years of life have been distinguished with differences in association with asthma later in life ( Fig. 1) [15,16]. For example, transient wheeze (wheezing in the first 3 years of life but not thereafter), is not strongly related to asthma later in life. In contrast, persistent (wheezing during the first 6 years of life), intermediate-onset and lateonset wheezing (wheezing starting at 1.5 and 3.5 years, respectively) during childhood has been associated with an asthma diagnosis at school-age. This suggests that these wheezing phenotypes likely reflect onset of conditions collectively identified as asthma [16][17][18]. Risk factors for these latter phenotypes include aeroallergen sensitization and sex. Before adolescence, male sex is associated with wheezing phenotypes and school-age asthma [19]. However, as puberty progresses, asthma prevalence and severity increase in the female population while the opposite is seen in males [4,20]. Other risk factors for asthma persistence include allergic sensitization during childhood, more severe asthma and a more obstructive lung function between the age of 5 and 12 [21,22]. Allergic sensitization plays an important role in the course of asthma, as suggested by investigations in the EGEA cohort, that reported that polysensitization was associated with asthma, allergic comorbidity and lower lung function outcomes [23]. In contrast, key factors associated with remission of asthma until early adulthood are milder initial disease, lack of allergic sensitization and comorbidity, lower degree of BHR and a better lung function [12,24]. Of individuals with persistent

KEY POINTS
The degree of airway obstruction during childhood is highly predictive of asthma remission and persistence until early adulthood.
Lung function, when combined with airway responsiveness and blood eosinophils may improve prediction of asthma remission.
Current evidence suggests that no treatment can change the natural course of asthma.
MicroRNAs could be important in predicting the course of asthma, with mi-R221-5p levels in blood being a promising candidate.
Distinct methylation profiles in blood or airway epithelium are associated with asthma remission; however, it remains to be seen if DNA methylation is the cause or consequence of remission.

Lung function
A low lung function early in life predicts asthma later in life [26 -28]. In addition to being regarded as an outcome of asthma, lung function is also a predictor of asthma remission and persistence [9]. In the Childhood Asthma Management Program (CAMP) study, a 10% higher FEV 1 /FVC during childhood was associated with an odds ratio (OR) of 4.62 for asthma remission during early adulthood (age 18 -23 years) [29 && ]. Similarly, fewer than 10% of individuals with a baseline FEV 1 /FVC ratio less than 80% would enter remission [29 && ]. In contrast, of individuals with an FEV 1 /FVC ratio above 90% during childhood, 54% of male individuals and 70% of female individuals would enter remission before early adulthood. These findings were replicated in two Dutch cohorts, showing that FEV 1 /VC measured at school age was significantly higher in individuals who later went into remission compared with individuals with persistence of asthma in adulthood [30]. In the CAMP cohort, of individuals with a FEV 1 /FVC less than 70%, only 1.4 and 11.1% of male and female individuals would enter asthma remission thereby being highly predictive of asthma persistence [29 && ]. Similarly, a 5% lower post-bronchodilation FEV 1 /FVC was associated with an OR of 2.36 for persistence of a severe asthma phenotype until early adulthood [22]. This is contrary to findings from a cohort of individuals with a severe asthma phenotype where neither a low nor an obstructive lung function at school age were predictive of asthma persistence during the 3-year follow-up [31 & ]. However, longer follow-up is required to assess outcomes in early adulthood.
Bronchial hyperresponsiveness BHR, defined as excessive airway constriction resulting from external stimuli is a key hallmark of asthma. In addition to facilitating asthma diagnosis, BHR may predict the long-term course of asthma. Less severe BHR between the ages of 5 and 12 years of age (defined by a higher methacholine PC 20 ) was associated with a higher odds of asthma remission in www.co-allergy.com early adulthood [24,29 && ]. Consequently, a lower degree of BHR during childhood may predict a more benign disease trajectory in a population of individuals with mild-to-moderate asthma.

Sex
Male children more frequently have asthma during early childhood whereas female children have a greater incidence and persistence of symptoms after puberty [4,12]. In a Swedish population-based cohort followed from mid-childhood until the age of 19 years, male sex predicted asthma remission in children with physician-diagnosed asthma [21]. This is in contrast to recent findings from the CAMP study in BHR-positive children with moderate-to-severe asthma where sex was not a predictor of asthma remission [29 && ]. The predictive value of sex in the course of asthma may, therefore, differ depending on the degree of asthma severity. This was also suggested by investigations in a cohort of children with a severe asthma followed over 3 years: remission of severe asthma was equally likely in both boys and girls during adolescence [31 & ]. The course of asthma is associated with puberty and sex, likely as the result of sex hormones affecting airway homeostasis and development [20]. Therefore, future research on this topic should employ sex stratification.

Treatment
Anti-inflammatory therapy with inhaled corticosteroids (ICS) is the cornerstone of asthma treatment. Investigation into the association between asthma treatment modalities and longitudinal outcomes is complicated as therapy regimens change regularly and confounding by indication (start of treatment in more severe cases) may complicate interpretation in observational studies. In the CAMP study, which started as a randomized clinical trial (RCT) comparing ICS with placebo on long-term lung function, treatment with ICS for at least 4 years did not affect asthma persistence or remission later in life [22]. These findings suggest that the role of ICS treatment is limited in affecting asthma outcomes. In the population-based OLIN cohort constituted of asthmatic children enrolled at age 7-8 years, treatment with ICS was associated with persistence of asthma [21]. However, these findings are difficult to interpret when studied in a nonrandomized setting as individuals with more severe asthma have a higher prevalence of ICS use [32,33].
Allergen immunotherapy (AIT) has been associated with better, short-term, clinical outcomes in children with asthma and allergy. Despite this, GINA guidelines do not recommend AIT in children with asthma, partly because of studies employing unvalidated symptom and medication scores [2,34]. It is uncertain to which extent this disease-modifying treatment may alter the course of asthma. In a 3year natural history study on asthmatic children sensitized to HDMs (house dust mites), the association between 3-year and 5-year courses of specific AIT and asthma remission was reported. Remission was defined as 12 months with no asthma symptoms requiring medication and a negative bronchial provocation test. After 3 and 5 years of treatment, 50 and 54.2% of the treatment group had entered remission as opposed to 3.3% in the control group receiving AIT [35]. These findings suggest that AIT could play a role in control of asthma symptoms and possibly alter the natural course of asthma. In a RCT of 812 pediatric individuals with allergic rhinoconjunctivitis without a history or signs of asthma, despite beneficial long-term clinical outcomes, sublingual immunotherapy (SLIT) did not affect the time to onset of asthma when compared with the control group [36]. This suggests that AIT, despite its therapeutic value for allergic rhinoconjunctivitis, does not prevent asthma. More high-quality, double-blinded, long-term follow-up studies are required to establish if AIT alters the natural evolution of asthma onset and persistence [37].

Allergic comorbidity
Allergic rhinitis and eczema have been associated with a lower likelihood of asthma remission [38,39]. The role of allergy in asthma persistence was supported by findings from the Swedish OLIN population-based cohort of individuals with asthma where allergic sensitization to furred animals was associated with a lower likelihood of remission in early adulthood [21]. Likewise, the Tasmanian Asthma Study reported that early-onset asthma and allergy phenotypes were associated with worse clinical outcomes, such as a lower and more obstructive lung function later in life in addition to a greater predisposition to Chronic Obstructive Pulmonary Disease (COPD) when compared with individuals with 'minimal or least asthma and allergies' [40 && ]. Consequently, allergic sensitization and the presence of allergic comorbidity are risk factors for persistence of asthma. Future research should investigate the effect of targeted treatment of allergic comorbidities, including biologicals, on longitudinal asthma outcomes.

Genetic factors
The pathogenesis of asthma is a complex interplay between environmental and genetic factors. Findings from a genome-wide association study (GWAS) have shown that over 100 genetic loci may be associated with the development of childhood and adult-onset asthma [41]. However, limited research has been performed on genetic prediction of asthma persistence and remission. A GWAS in a Dutch hospital-based cohort constituted of 790 asthmatic patients followed from preadolescence of which 179 individuals were in clinical remission and 55 individuals were in complete remission investigated single nucleotide polymorphisms (SNPs) associated with remission [42]. Replication was performed in two independent cohorts followed by expression quantitative loci analysis (eQTL) in lung tissue for identified SNPs. For complete remission, three SNPs were replicated with one (rs6581895) almost reaching genome-wide significance. This SNP was further identified as an eQTL for fibroblast growth factor receptor 2 (FRS2) and for chaperonin containing TCP1 subunit 2 (CCT2). The association between FRS2 and asthma remission may suggest that this gene is involved in resolution of inflammation [43]. Another SNP (rs1420101) associated with complete remission, was a cis-eQTL for IL-1RL1 and IL-18R1 [43]. The childhood asthma susceptibility gene IL-1RL1, which encodes the receptor of the epithelial alarmin IL-33, is known to promote airway inflammation [44,45]. More research is needed on variation of IL-1RL1 in different asthma phenotypes to establish, which individuals may benefit from treatments, such as anti-IL 33 [46]. Recently, a phase II clinical trial with anti-IL-1RL1 (astegolimab) showed that this drug significantly reduced asthma exacerbations in a broad population of patients, with inadequately controlled, severe asthma [47].

Epigenetic factors: microRNAs
MicroRNAs (miRNAs), which are small noncoding RNA that have regulatory function, have shown promise as a possible biomarker for predicting the course of pediatric asthma. In a recent study, several blood miRNAs were associated with the risk of asthma exacerbations in the CAMP cohort [48]. In the same cohort, miRNAs could also predict asthma remission [49 & ]. In the multivariate analysis, a higher blood expression of mi-R221-5p at baseline was associated with a lower likelihood of asthma remission by early adulthood. In addition to promoting IL-6 release and IgE-mediated mast cell degranulation, higher mi-R221-5p expression has been linked to higher eosinophilic airway inflammation [50,51]. Therefore, blood mi-R221-5p levels could serve as a potential biomarker for asthma remission.
The role of miRNA expression in individuals with different asthma outcomes has also been studied in adult populations [52 & ]. In a recently published Dutch study, total RNA was obtained from bronchial biopsies in 14 individuals with complete asthma remission, 46 patients with persistent asthma and 82 healthy controls. When comparing complete remission to persistent asthma and complete remission to healthy controls, 10 and 77 miR-NAs were differentially expressed. Furthermore, by applying a Bayesian network analysis, a network of miRNAs and long noncoding RNAs characteristic of complete remission was described. Of interest, this work showed that asthma remission was clearly different compared with a healthy state.

Epigenetic factors: DNA methylation
Advancements made within the field of (epi)genomic studies may also provide valuable insights into the pathological mechanisms underlying the course of asthma ] performed DNA methylation analyses of nasal brushes and whole blood, of which whole blood DNA methylation was replicated in two separate cohorts. Qi et al. identified distinct CpG sites associated with clinical (cg13378519 Chr. 1) and complete asthma remission (cg24788483, Chr. 10). The association of these CpG sites with asthma remission suggests that peroxisome proliferation and Wnt signaling may play a role in asthma remission. In nasal brushes, 25 CpG sites were associated with asthma remission phenotypes; these findings require replication. These findings also emphasize that the methylation profile of individuals having undergone remission differs compared with healthy individuals, thereby suggesting that resolution of symptoms does not result in the return to a methylation profile identical to that of nonaffected individuals. This is illustrated for blood CpG methylation of cg13378519 (Fig. 2). However, it remains to be determined if methylation is the cause or consequence of asthma remission. Future studies should, therefore, adopt a prospective design to evaluate the predictive value of DNA methylation on asthma outcomes.

Combined prediction models for the course of asthma
Prediction models combine clinical parameters to predict the course of asthma. By combining FEV 1 / Predicting the course of asthma Koefoed et al.  FVC (! 85%), airway responsiveness (PC 20 of !1 mg/ ml) and blood eosinophil measurements (<500 cells/ml) in childhood, Wang et al. [29 && ] were able to correctly predict asthma remission in 82.6% of cases until early adulthood (Fig. 3). However, in two Dutch asthma cohorts, this model correctly predicted asthma remission in only 40% of cases [30]. This highlights how combining readily available data may be used to predict future outcomes. To further analyze additional heterogeneity in asthma phenotypes, prediction model development may benefit from incorporating machine learning approaches [55].

CONCLUSION
An obstructive lung function during childhood, allergic comorbidity and polysensitization are important predictors for persistence of asthma. Improved prediction of persistence is needed to identify patients who require more intense clinical follow-up. Recent epigenetic findings show that asthma remission represents a new state different to that of nonaffected individuals. However, the causality of these findings remains to be determined. The epigenetic studies have provided a better understanding of underlying pathological processes in asthma remission, and this may be used to develop novel treatments aimed at altering the course of asthma.
Boudewijn IM, Roffel MP, Vermeulen CJ, et al. A novel role for bronchial microRNAs and long noncoding RNAs in asthma remission. Am J Respir Crit Care Med 2020; 202:614-618. This study investigated differential miRNA expression in an adult population of subjects with complete asthma remission, persistence of asthma and healthy control subjects. This study highlights the differential expression of miRNA is subjects with asthma remission compared with healthy controls suggesting that remission represents a new state as opposed to a return to a healthy state.

53.
& Vermeulen CJ, Xu CJ, Vonk JM, et al. Differential DNA methylation in bronchial biopsies between persistent asthma and asthma in remission. Eur Respir J 2020; 55:. This study investigated differential DNA methylation using EWAS with samples obtained from bronchial biopsies in subjects with asthma remission, persistence and healthy controls. It highlights the differences in DNA methylation amongst subjects with asthma remission compared with healthy controls suggesting that asthma remission represents a new state as opposed to a return to a healthy state.

54.
& Qi C, Vonk JM, van der Plaat DA, et al. Epigenome-wide association study identifies DNA methylation markers for asthma remission in whole blood and nasal epithelium. Clin Transl Allergy 2020; 10:60. This study identified differentially methylated CpG sites obtained from whole blood and nasal brushing samples and their association with clinical and complete asthma remission in an adult population. It highlights the possible role of peroxisome proliferation and Wnt-signaling pathway in remission of asthma. 55. Kothalawala DM, Kadalayil L, Weiss VBN, et al. Prediction models for childhood asthma: a systematic review. Pediatr Allergy Immunol 2020; 31:616-627.