Urine: A Lens for Asthma Pathogenesis and Treatment?

In this issue of the Journal, Kolmert and colleagues (pp. 37–53) in the U-BIOPRED (Unbiased Biomarkers for the Prediction of Respiratory Diseases Outcomes) Study Group report urinary eicosanoid levels from healthy control subjects, subjects with mild–moderate asthma, and subjects with severe asthma (SA) (1). The rationale for this study is that there is a lack of predictive biomarkers for which patients with asthma may be stratified based on the pathobiological mechanisms that lead to disease severity, and such biomarkers may be able to be used to improve treatment selection. As the authors note, the amount of eicosanoids in each urine collection represent the integration of the systemic load of these mediators since the last urination, therefore providing an ongoing assessment of their production. Before we discuss the results of their study, it is important to understand the context in which eicosanoids are currently understood to have a role in asthma pathogenesis.

Leukotrienes and prostaglandins are lipids produced from arachidonic acid metabolism that have pleiotropic biologic functions in the lung. For specialists in pulmonary medicine and allergy/immunology, these mediators have a particular importance in that they regulate many aspects of asthma pathophysiology (2). For instance, the cysteinyl leukotrienes (cysLTs), measured in the U-BIOPRED study, can be synthesized as a result of allergen-induced, IgE-mediated reactions by mast cells and basophils (3). Eosinophils are also important producers of cysLTs (4). The cysLTs consist of LTC4, LTD4, and LTE4, which are sequentially produced, as shown in Figure 1. The half-lives of LTC4 and LTD4 are very short, making it challenging to measure them in biologic fluids; however, LTE4, the end product of cysLT metabolism, is stable and can be measured in the urine (5), thus providing an opportunity to quantify cysLT production as performed by Kolmert and colleagues. The cysLTs cause bronchoconstriction and induce airway epithelial cell mucin expression, both cardinal features of allergic asthma (6, 7).

Figure 1.

Metabolism of arachidonic acid in the production of leukotrienes and prostaglandins. 5-LO = 5-lipoxygenase; COX = cyclooxygenase; LT = leukotriene; PG = prostaglandin; PLA2 = phospholipase A2; TXA2 = thromboxane A2.

Although cysLTs are products of arachidonic acid metabolism through the 5-LO (5-lipoxygenase) pathway, arachidonic acid may also be metabolized through the COX (cyclooxygenase) pathway to produce the prostaglandins, also shown in Figure 1. PGD2 is the major prostaglandin produced by IgE-mediated mast cell activation, whereas basophils, eosinophils, and macrophages are inflammatory cells in the airway that can also synthesize PGD2 (8, 9). PGD2 promotes allergic inflammation in multiple ways by signaling through the receptor DP2, which is also known as CRTH2. DP2 is expressed on eosinophils, basophils, CD4 T-helper cell type 2 (Th2) cells, and group 2 innate lymphoid cells (ILC2) (10, 11). DP2 signaling in eosinophils augments their release from bone marrow, increases their respiratory burst, stimulates the chemotactic response to other chemokines such as eotaxin, and primes them for degranulation (12). PGD2 signaling through DP2 stimulated human peripheral blood ILC2 to secrete large amounts of IL-13 to the same level produced in response to IL-25 and IL-33, whereas the addition of IL-25 and IL-33 to PGD2 synergistically increased IL-13 expression by ILC2 (13), and PGD2 increased ILC2 expression of the IL-33 and IL-25 receptor subunits, ST2 and IL-17RA, respectively (10). Importantly, there seem to be synergistic effects of PGD2 and cysLTs in promoting allergic inflammatory responses. For instance, LTE4 enhanced the activation of ILC2 and type 2 cytokine production by PGD2 (14). Therefore, understanding the possible contribution of cysLTs and PGD2 to asthma, and in particular SA, would provide insight into disease pathogenesis.

In their study, Kolmert and colleagues stratified the results of the subjects with SA into groups that are in the highest or lowest 25th percentile for urinary eicosanoids. Those subjects with SA who were in the highest 25th percentile of urinary LTE4 and PGD2 metabolites had significantly lower lung function yet had increased levels of exhaled nitric oxide and blood and sputum eosinophils, in addition to other markers of type 2 inflammation, such as periostin. The authors interpret these results as justifiably suggesting that there is increased mast cell activation in SA. The authors also report that males had higher levels of the urinary PGE2 metabolite than females. This is important because, as the authors point out, there is strong data that PGE2 signaling through receptors that activate cyclic AMP downregulates allergic inflammation and bronchoconstriction (15, 16), thus suggesting a potential mechanism as to why adult females have a greater incidence of asthma as well as more severe disease.

The authors further stratified subjects into those that were treated with oral corticosteroids and found, somewhat surprisingly, there was no difference in the majority of the eicosanoid measurements based on usage of this medication. Interestingly, when patients were stratified based on omalizumab treatment, the authors found that, in contrast to the oral corticosteroid data, omalizumab use significantly decreased the urinary levels of LTE4 as well as metabolites of PGD2 and thromboxane. Based on these results, the authors suggest that urinary eicosanoid levels possibly could be used as a predictive biomarker of response to biologics such as omalizumab. However, there is no information that the subjects treated with omalizumab had a response to this medication; therefore, in this instance, we have no idea as to whether the change in the urinary eicosanoids signaled successful treatment to this biologic.

The merits of this manuscript include the enormous amount of data, particularly in the supplemental data section, that will be of use to other investigators at the intersection of the eicosanoid and asthma fields. Other major strengths of this project include the internal validation of the original results with a follow-up study of adults with SA 12–18 months after the original study. The authors went even a step further by performing an investigation of adolescents who had asthma, the results of which provided external validation of the data found in the adults with asthma. The data certainly provides evidence for both a role for eicosanoids as determinants of asthma severity and the possibility that urinary eicosanoids could be used to stratify asthma by pathobiology as an adjunct to clinical severity. The authors propose that urinary eicosanoids could be used to phenotype patients before treatment with biologics to predict response to these expensive drugs. The next step would be to measure urinary eicosanoids before entry into trials of biologic agents to determine if these could be predictors of success or failures, followed by prospective trials to confirm that urinary eicosanoids are indeed true biomarkers of response to therapy. Though there are many strengths of this project, there are some shortcomings. For instance, PGI2 metabolites were not examined because they were lost. PGI2 is a negative regulator of CD4 and ILC2 type 2 cytokine production, inhibits dendritic cells from inducing Th2 immune responses, and promotes immune tolerance in the airway (2). It would have been interesting to see if there was an inverse correlation of the stable urinary PGI2 metabolite with asthma severity, suggesting that PGI2 may be protective. Furthermore, lipids that have antiinflammatory effects and that promote resolution of inflammation, such as lipoxins, protectins, resolvins, and maresins, were not measured, and such data would have provided insight into the importance of these mediators in asthma pathogenesis (17). However, despite these limitations, the work by the U-BIOPRED study group is an important addition to the asthma field, and future clinical trials will be important in defining how this data can be used to direct precision treatments.