UnloCKing the Role of Creatine Kinase in Childhood Asthma

Asthma affects more than 300 million people worldwide, with wide variation in prevalence across countries (1, 2). In the United States, asthma affects more than 5 million children, with pronounced disparities among different socioeconomic, racial, and ethnic groups (3). In the past few decades, researchers have made substantial strides in understanding the complex pathophysiology of asthma. We now understand asthma is not a homogeneous disease but rather an umbrella term that groups common symptoms and manifestations that arise from diverse mechanisms and pathways. Although advances in our understanding of the underlying pathobiology of asthma have led to better therapies for our patients (4), effective early diagnosis and sustained control of asthma symptoms with treatment remain an ongoing challenge. As such, novel insights into the pathways involved and the long-term potential for these findings to translate into clinical biomarkers or therapeutic targets remain of great interest.

In this issue of the Journal, Guerra and colleagues (pp. 544–552) report a novel molecular association: that CK (creatine kinase, also known as creatine phosphokinase or CPK) may be involved in the pathobiology of childhood asthma (5). In a cross-sectional analysis of data from eight longstanding and well-characterized birth cohorts (seven in Europe and one in the United States), they evaluated the association between serum concentrations of 88 proteins and asthma. Of 46 detectable proteins, 4 were associated with asthma, and, of those, lower concentrations of CK were associated with higher odds of asthma in all stages of their analysis. Furthermore, gene expression of both cytosolic forms of CK (CKM and CKB) were lower in children and young people with asthma than in control subjects. In a complementary murine model of asthma, the authors then demonstrated lower concentrations of CK protein after house dust mite sensitization; furthermore, CKB inhibition impeded the resolution of airway hyperresponsiveness and mucin production, suggesting a potential biological mechanism for the observed associations. The study by Guerra and colleagues has several strengths. Data were compiled from several well-characterized longitudinal cohorts from different backgrounds and environments. The analytical design comprised discovery, validation, and independent replication stages, which strengthens confidence in the reported findings—an approach that should be more widely implemented in epidemiological studies. In addition to serum protein concentrations, the investigators evaluated gene expression, which supported the findings. Finally, the authors included experimental data from a murine model that provided a mechanistic explanation to support the epidemiological findings.

At the same time, there are several limitations that the authors transparently address in their discussion. Although the multiplex panel used was comprehensive, it was presumably constructed based on prior biological knowledge and therefore did not represent an entirely unbiased “discovery” stage. Only 52% of proteins in the panel were detectable and analyzed; some proteins in the panel have been previously associated with asthma yet were not detected or not significant in the current study. Protein concentration z-scores were calculated within each cohort, hindering generalizability and preventing identification of a standard “cut-off” concentration to determine risk. Most importantly, the study did not perform a clinical validation of CK as a potential biomarker for asthma. Combining data from different cohorts may introduce variation in both the age at which CK is measured and the nature of the wheezing phenotypes encompassed by an “asthma diagnosis.” The authors compared concentrations in the lowest tertile to those in the upper tertile, but we do not know how well CK concentrations classify asthma versus health when assessed as a continuous variable. Most importantly, the study did not include disease controls, so it remains unclear whether the changes observed here in CK concentrations are specific to asthma or relate to the presence of chronic disease per se—respiratory or otherwise. CK is expressed in many cells and tissues, particularly in those with high energy requirements. Although CK-MB is more specific to cardiomyocytes and therefore serves as a biomarker for acute coronary disease, CK may be elevated in muscle damage or inflammation from different etiologies, including exercise; conversely, serum CK concentrations may be low in hospitalized or critically ill patients (6), conditions that decrease muscle mass, persons with low physical activity, connective tissue disorders (7), hyperthyroidism (8), or steroid administration (9, 10). Nonetheless, the findings from the current study are novel and intriguing; we could not find any studies on CK in asthma beyond isolated reports of elevated concentrations during severe exacerbations or with administration of certain asthma medications (11–13).

Despite these limitations, the results presented by Guerra and colleagues are interesting and thought provoking for future research. Replication of the observational findings will be important. Crucially, forthcoming studies must be designed to determine whether differences in CK concentrations are specific to asthma or nonspecifically related to chronic disease. Data from longitudinal studies could be leveraged to assess the utility of CK as a predictive biomarker (i.e., for incident asthma) and to identify potential cut-off concentrations with adequate sensitivity and specificity. Further understanding of the mechanisms by which CK concentrations are reduced is needed, as well as whether data from an established adult mouse model of asthma translates to juvenile mice, which may be more reflective of pediatric asthma. Differential responses between adult and juvenile mice are already well described (14). Moreover, mechanistic and clinical studies will be needed to assess whether differences in CK concentrations seen in the murine lung translate to the human airway and affect lung function parameters, including airway hyperresponsiveness. Whether CK can serve as an asthma biomarker or may help identify novel therapeutic agents will require substantial work, but in the meantime the current study represents a truly collaborative, multinational, translational study with results that deserve further investigation.