Response to: ‘Occupational asthma contribution to phenotyping adult asthma by using age-of-asthma onset clustering’

Response to: Antonino DS, Fabio RLM. Occupational asthma contribution in phenotyping adult asthma by using age-of-asthma onset clustering. Expert Rev Respir Med 2015;9(4):387-88

We thank the authors of this letter for their comments in relation to our review [1].

In their letter, the authors discuss the importance of relating clinical phenotypes to specific pathogenic mechanisms (endotypes). They further suggest that lessons learned from occupational asthma could help inform the use of age-of-asthma onset in future cluster-based studies.

The authors refer to an example in their previous work [2] – a detailed examination of toluene diisocyanate (TDI)-induced asthma. Their study demonstrated that two distinct subgroups of TDI-induced asthma, each with different inflammatory characteristics, could be identified based on duration of exposure required to induce asthma. Of note, this was a retrospective study with a small sample size of only 18 subjects. In light of the mean age-of-onset of the two groups, the authors assert that age 12 years, the most commonly used cut-off to delineate early- and late-onset adult asthma, is insufficient and that age 40 years may be more appropriate.

We agree with several points raised in this letter, and certainly occupational asthma forms a significant component of adult lung disease, with parallels to non-work-related allergic asthma. [3]. However, we argue that age 12 years is indeed an appropriate cut-off to delineate early-onset and late-onset adult asthma. As the authors note, studies describing adult asthma phenotypes should exclude non-current childhood asthma. Hence, our review excluded studies including subjects with remitted asthma. Having said this, it is also important to recognize that up to 40% of adults with current asthma start their disease as children [4].

The rationale behind the use of age 12 as a cut-off stems from epidemiological data on gender-specific differences in asthma prevalence over time [5]. Asthma prevalence in boys is approximately twice that of girls; however, in adolescence, this pattern reverses and we observe a ‘gender switch’ [5]. In a cross-sectional study, Venn et al. showed that this prevalence reversal occurred at age 12. After age 12, the prevalence of asthma and wheeze increased with age for females and decreased for males [6]. A review of gender-specific differences in asthma incidence and prevalence summarizes several similar studies with consistent findings [5].

The authors’ suggestion that age 40 might be an appropriate cut-off, however, is not without merit. A second ‘gender switch’ is observed later in life at around age 50–55 and occurs in parallel with menopause and andropause [7,8]. At this age, the male–female gap in asthma incidence decreases and even reverses after the 5th decade [7,8]. Asthma in older adults (≥65 years) clearly differs from asthma in middle age and the appropriate age-of-asthma onset cut-off in these older individuals is less clear.

For asthma in middle age, we maintain that age 12 is an appropriate cut-off to delineate early onset and late onset. Occupational asthma should instead be considered a specialized subgroup of late-onset asthma, of which further ‘phenotype–endotypes’ can be described, for example, the two distinct subgroups of TDI-induced asthma described by Di Stefano and colleagues [2]. Certainly, studies of occupational asthma can provide useful insight into asthma pathophysiology; however, the broad application of these concepts to the general population should be done cautiously.

These points aside, we wish to reiterate the authors’ salient comment that, clinical phenotypes should be coupled with underlying endotypes. Cluster-based analyses incorporating both clinical characteristics – asthma severity, age-of-onset, lung function – with genetic and biological markers may help establish a new taxonomy of asthma and airway disease and facilitate the development of phenotypic-specific treatment.

Financial & competing interests disclosure

SC Dharmage, EH Walters, CJ Lodge, AJ Lowe and MC Matheson are supported by National Health and Medical Research Council of Australia. JL Perret is supported by an Australian Postgraduate Award and Australian Lung Foundation. The authors have no other relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript apart from those disclosed.