Exhaled Eicosanoids following Bronchial Aspirin Challenge in Asthma Patients with and without Aspirin Hypersensitivity: The Pilot Study

Background. Special regulatory role of eicosanoids has been postulated in aspirin-induced asthma. Objective. To investigate effects of aspirin on exhaled breath condensate (EBC) levels of eicosanoids in patients with asthma. Methods. We determined EBC eicosanoid concentrations using gas chromatography/mass spectrometry (GC-MS) and high-performance liquid chromatography/mass spectrometry (HPLC-MS2) or both. Determinations were performed at baseline and following bronchial aspirin challenge, in two well-defined phenotypes of asthma: aspirin-sensitive and aspirin-tolerant patients. Results. Aspirin precipitated bronchial reactions in all aspirin-sensitive, but in none of aspirin-tolerant patients (ATAs). At baseline, eicosanoids profile did not differ between both asthma groups except for lipoxygenation products: 5- and 15-hydroxyeicosatetraenoic acid (5-, 15-HETE) which were higher in aspirin-induced asthma (AIA) than inaspirin-tolerant subjects. Following aspirin challenge the total levels of cysteinyl-leukotrienes (cys-LTs) remained unchanged in both groups. The dose of aspirin had an effect on magnitude of the response of the exhaled cys-LTs and prostanoids levels only in AIA subjects. Conclusion. The high baseline eicosanoid profiling of lipoxygenation products 5- and 15-HETE in EBC makes it possible to detect alterations in aspirin-sensitive asthma. Cysteinyl-leukotrienes, and eoxins levels in EBC after bronchial aspirin administration in stable asthma patients cannot be used as a reliable diagnostic index for aspirin hypersensitivity.


Introduction
Exhaled breath condensate (EBC) is a simple, noninvasive technique for monitoring airway inflammation. The measurement of eicosanoids in the expired breath condensate has proven to be a useful noninvasive method for the assessment and monitoring of airway inflammation in inflammatory diseases such as asthma and other pulmonary diseases [1,2].
The aim of this study was to evaluate the changes in wide eicosanoid spectrum concentrations in EBC during asthmatic response following aspirin inhalation. We hypothesized that the profile of eicosanoids in EBC after local aspirin administrated is markedly different in aspirin-induced asthma patients as compared to asthmatics who tolerate aspirin well. These results were validated by specific analytical techniques, such as gas chromatography/mass spectrometry (GC-MS) or by high-performance liquid chromatography/mass spectrometry (HPLC-MS).

Subjects.
The study population consisted of 21 asthmatic patients sensitive to aspirin (AIA) and 23 asthmatics who tolerated aspirin well (ATA). The diagnosis of asthma was established according to GINA 2008 update. The patients' characteristics are presented in Table 1.
The diagnosis of aspirin intolerance was confirmed by oral aspirin provocation tests, performed during 36 months preceding the study. All ATA patients occasionally used aspirin without any adverse reactions. The patients had stable asthma and their baseline FEV 1 was >70% of the predicted value on the study day. None had experienced an exacerbation or a respiratory tract infection in the 6 weeks preceding the study. Nearly 70% of ATA patients had intermittent asthma, 13% mild persistent asthma, and 17% moderate persistent. In AIA group, 67% of patients had intermittent asthma, 19% mild persistent asthma, and 14% moderate persistent asthma. According to Asthma Control Test 33% of AIA patients and 26% of ATA patients had controlled asthma, and 33% and 48% had partly controlled asthma, 33% and 26% had uncontrolled asthma, respectively. In ATA group, 2 patients were current smokers and 3 exsmokers. There were 7 exsmokers and no current smokers in AIA group. The average level of FEV 1 and FEV 1 /FVC in AIA patients was 89.9% and 73.2%, in ATA patients was 91.5% and 78.5%, respectively.
The subjects were instructed to withhold medications that decrease bronchial responsiveness prior to aspirin challenge. Short-acting β 2 -agonists were not used 8 hours before the challenge. Long-acting β 2 -agonists and theophylline were  withdrawn for 24 hours. Short-acting antihistamines and cromones were stopped 5 days before the challenge. Inhaled steroids were allowed at a dose ≤2000 μg budesonide per day. None of the patients were treated with systemic corticosteroids or leukotriene modifying drugs. Baseline and following bronchial aspirin challenge, exhaled breath condensate eicosanoids (see Table 2) levels were measured in all subjects. The patients gave informed consent and the study was approved by the University Ethics Committee.

Study
Design. The single-blind, placebo-controlled bronchial challenge test with aspirin was carried out during one day in all study patients [29]. The test began with the inhalation of 7 breaths of placebo (saline). FEV 1 was measured at 10 and 20 minutes after placebo inhalation. The postsaline FEV 1 obtained at 20 minutes was used as "postsaline baseline" value.
The consecutive doses of lysine-aspirin were inhaled every 30 minutes by increasing the concentration of lysineaspirin and by changing the number of breaths (increasing doses of 0.18, 0.36, 0.90, 2.34, 7.20, 16.2, 39.60, 115.20 mg, at 0.5 hour intervals, up to the cumulative dose of 181.98 mg). FEV 1 was measured at 10, 20, and 30 minutes after each dose. The challenge procedure with aspirin was interrupted, if a bronchospastic reaction occurred (FEV 1 dropped ≥ 20%), or if the maximum cumulative dose of aspirin was reached. The cumulative dose of aspirin causing a 20% fall in FEV 1 was calculated and recorded as PD 20 (provocation dose of aspirin). FEV 1 and extrabronchial symptoms were recorded at baseline, before the challenge tests, and then every 30 minutes until 6 hours after the last dose of aspirin.
In patients with positive bronchial aspirin challenge (AIA), exhaled breath condensate samples were collected for wide eicosanoid spectrum (see Table 2) estimations at baseline and at the time of appearance of the bronchial symptoms (time 0). In ATA patients, whose aspirin challenge was negative, exhaled breath condensate samples were collected at baseline and 0.5 hours after the last aspirin dose, that is, when the cumulative doses of 181.98 mg was reached (time 0).

Lung Function.
Pulmonary function tests were performed on a flow-integrating computerized pneumotachograph (Pneumoscreen, E. Jaeger, Germany).

Exhaled Breath Condensate (EBC)
. EBC was collected according to ATS/ERS [1] using ECO Screen instrument of Jaeger (GmbH Hoechberg, Germany). Following tidal breathing for 15-20 min, 1-2 mL of clear fluid was collected and immediately deeply frozen.

Biochemical Assays.
Exhaled breath condensate concentration of eicosanoids was measured by gas chromatography/mass spectrometry (GC-MS) and by high-performance liquid chromatography/tandem mass spectrometry (HPLC-MS 2 ) or both; see Table 2. Results of eicosanoids were recalculated as parts per million (ppm) of palmitic acid (PA) Journal of Allergy 5 or expressed as picograms per milliliter (pg/mL). Detection limits for eicosanoids measurements were between 0.17 pg/mL for 12-HETE and 0.89 pg/mL for PGD 2 . Intraassay coefficients of variance were less than 10% and interassay coefficients of variance were less than 15%. Accuracy of measurements were better than 98.7%. Detailed analytical procedure and deuterated standards used were described elsewhere [3,27].
2.6. Statistical Analysis. Summary statistics were expressed as mean (M), standard deviation (SD), median (Me), and 25% and 75% percentiles. General linear model (GLM) including repeated measures analysis of variance, which takes into account the fact that the outcome measurements are repeated over time within subject was used for multiple comparisons. Logarithmic transformation was used when needed as variance stabilizing transformation. Correlation between variables was estimated with the Spearman rank order correlations. A P-value ≤ 0.05 was considered statistically significant.

Clinical Reactions.
There was no statistical difference in the clinical characteristics between patients with aspirininduced asthma (positive aspirin challenge test), and those who tolerated aspirin well (negative aspirin challenge test) except for blood eosinophil count, Table 1. None of the patients developed symptoms after administration of placebo. In aspirin-sensitive asthmatics, bronchial reactions developed after 0.18 mg in 1 subject, after 0.36 mg in 2 subjects, after 0.9 mg in 3 subjects, following 2.34 mg in 2, after 7.2 mg in 5 subjects, after 16.2 mg in 4 subjects, following 39.6 mg in 2 subjects, and after 115.2 mg in 2 subjects. The mean cumulative dose of aspirin was 32.108 mg. All the symptoms were relieved by short-acting β 2 -agonists. None of ATA patients developed any clinical symptoms following aspirin challenges.

Cyclooxygenase Products
Exhaled Breath Condensate Concentration of Prostanoids. At baseline (Table 2), exhaled breath condensate levels of PGD 2 , such as its metabolite 9α11βPGF 2 , and PGF 2 α, 6-keto-PGF 1 α and nonenzymatic isomer (8-iso-PGF 2 ), did not differ significantly between the study groups. Statistically, results did not differ dependently on the methods used (HPLC/MS/MS or GC/MS) for marking eicosanoids and results shown (ppm of PA or pg/mL).
Following aspirin administration, no significant differences in EBC levels of PGD 2 measured by HPLC/MS/MS (results were recalculated as ppm of PA) and 9α11βPGF 2 measured by GC/MS (results were expressed as ppm of PA) were found in AIA (P = 0.13 and P = 0.82, resp.) and in ATA subjects (P = 0.69 and P = 0.43, resp.). The same marked result (no differences) was observed when PGD 2 was measured by GC-MS.
In cases when 11-dehydro TXB 2 resulted as ppm of PA, at baseline (Table 2) and following aspirin challenge (ANOVA, P = 0.66) exhaled breath condensate level of 11-dehydro TXB 2 did not differ significantly between the study groups.

Lipoxygenation Products Exhaled Breath Condensate Concentration of Cysteinyl-Leukotriene (Cys-LTs). At baseline (
In both study groups, following aspirin challenge, EBC level of LTC 4 decreased significantly (ANOVA, P = 0.003). No significant differences in EBC levels of LTD 4 was found in AIA and in ATA group patients (ANOVA, P = 0.67), where the level remained unchanged and at a constant. Exhaled LTE 4 after aspirin challenge increased significantly in both study groups (ANOVA, P = 0.03). However, total level of cys-LTs (the sum of LTC4, LTD4, LTE4) showed no changes in either group studied. Statistically the results of cys-LTs in EBC were identical independent of units (measurement) used (ppm of PA or pg/mL).
The dose of inhaled steroid used by study patients and FEV 1 values had no effect on magnitude of the response of the cys-LTs and its duration.

Exhaled Breath Condensate Concentration of Leukotriene B 4 .
At baseline (Table 2), exhaled breath condensate level of leukotriene B 4 did not differ significantly between the AIA and ATA groups (P = 0.36).

Exhaled Breath Condensate Concentration of HETE. At baseline
At baseline (Table 2), exhaled breath condensate levels of 12-HETE did not differ significantly between the AIA and ATA groups (P = 0.13). Following aspirin administration, EBC levels of 12-HETE decreased significantly only in ATA group (P = 0.03, due to outliers Wilcoxon Matched pairs test was used).   The dose of inhaled steroid used by patients and FEV 1 values had no effect on magnitude of the response of the 5and 15-HETE. At baseline, negative correlation was found between the doses of steroids and EBC levels of 12-HETE only in aspirin-sensitive subjects (r = −0.45, P = 0.04).
The dose of aspirin, had no effect on the magnitude of response of 5-and 15-HETE.
Exhaled Breath Condensate Concentration of Eoxins. At baseline and following aspirin challenge, exhaled breath condensate levels of eoxins C 4 , D 4 , and E 4 did not differ significantly between the AIA and ATA groups (see Table 2).
At baseline and following aspirin challenge no correlation was found between provocation doses of aspirin, inhaled steroid therapy and FEV 1 values, and eoxins EBC levels in aspirin-sensitive and aspirin-intolerant patients.

Discussion
In this study, we used a validated analytic platform [27,30,31] to analyze eicosanoids in EBC of asthmatic patients. A highly sensitive method of gas chromatography/mass spectrometry or high-performance liquid chromatography/mass spectrometry or both were used to measure spectrum of eicosanoids-nonvolatile compounds present in EBC [3,27,31]. Novel concepts for the standardization of EBC material measurements have been introduced to obtain characteristics of eicosanoid patterns produced by asthmatic lungs [27]. The assessment of palmitic acid content in EBC among many other methods [32] seems to be a convenient solution for compensating the "dilution factor" [3]. It has recently been demonstrated that dilution of nonvolatile compounds in water differs between the subjects by more than 1 order of magnitude and depends on ventilation mechanics [33,34]. For that reason, in this paper, data are recalculated as parts per million of palmitic acid. For better understanding, EBC eicosanoid levels are given in both applicable units: ppm of PA (Table 2) and in pg/mL (Table 3) independently of performed assay.
The aim of this study was to compare a wide profile of eicosanoids released to the epithelial surface of the asthmatic lung in subjects with and without aspirin hypersensitivity at baseline and following bronchial aspirin challenge, most of them on chronic inhaled steroid therapy. We demonstrated for the first time exhaled eicosanoids following local administration of aspirin in aspirin-induced asthma patients.
Comparing subjects with AIA and ATA no significant differences were observed in EBC levels of cyclooxygenase pathway prostanoid products (PGD 2 and its metabolite 9α11βPGF 2 , PGF2α, 6-keto-PGF1α, and 11-dehydro TXB 2 ) and nonenzymatic isomer (8-iso-PGF 2 ) in neither baseline nor after aspirin inhalation. Opposing data has been earlier reported as lower [28] and other times higher [27] baseline PGD 2 metabolite, namely, 9α,11βPGF 2 , levels in EBC in AIA patients. Differences observed in this previous study between both asthma phenotypes [27] can possibly be explained by lower FEV 1 values and more severe of disease in patients with aspirin hypersensitivity compared to subjects who tolerated aspirin well. Moreover, a significant predictor of decreased FEV 1 was increased 9α,11β-PGF 2 only in AIA which did not correlate in ATA subjects [27]. In our study, all patients presented comparable FEV 1 values, and most likely, similar bronchoconstricting eicosanoid levels such as PGD 2 and its metabolite. Mast cells are probably the main source of PGD 2 overproduction. Higher global production of PGD 2 metabolite 9α,11βPGF 2 was also present in the blood and urine at baseline in AIA subjects [18,20]. Concentration of 9α,11β-PGF 2 was not changed by the oral-systemic [28] and how indicated our data bronchial-local administration of aspirin.
Our results revealed a significant upregulation of some arachidonate lipoxygenation products in asthmatic subjects with aspirin hypersensitivity, as manifested by high baseline levels of 5-, 15-HETE in EBC. This data is consistent with the latest observations [27]. These findings are related to an overexpression of lipoxygenases enzymes, particularly 5-and 15-LO in the asthmatic lung with aspirin hypersensitivity. These enzymes are expressed in eosinophils, activated macrophages, and also in lymphocytes and mast cells. Kowalski and colleagues have demonstrated that aspirin triggers specific generation of 15-HETE from nasal polyp epithelial cells [12] and peripheral blood leukocytes [35,36] from aspirin-sensitive but not aspirin-tolerant subjects with asthma/rhinosinusitis. Also, they have demonstrated that two alternatively spliced variants of COX-1 mRNA present in human leucocytes may be differently expressed in patients with asthma. The relative expression of those variants has been correlated to aspirin-triggered 15-HETE generation suggesting association of this phenomenon with the pathogenesis of aspirin-induced asthma [37].
No elevation of baseline 12-HETE in EBC comparing AIA and ATA subjects was observed. Whereas, after bronchial aspirin administration, there was a statistically significant decrease in 12-HETE concentration noted only in ATA subjects. On that basis, we assume blood platelets, the main source of 12-LO, may possibly play some role in pathogenesis of aspirin hypersensitivity. However, 12-LO was originally cloned from respiratory epithelia, where 15-LO activity was also found [38].
Our study did not demonstrated baseline local overproduction of cys-LTs in the airways in AIA and is consistent with an earlier study [28]. Baseline EBC levels of LTC 4 , LTD 4 , and LTE 4 did not differ between aspirin tolerant and hypersensitive subjects. Contrary to this finding in subjects with AIA, a higher baseline level of cys-LTs in EBC was also reported [26,27]. This might have been a result of low FEV 1 values and minor control of asthma (severity index) [27] or steroid-naive [26] in aspirin hypersensitivities compared to aspirin tolerant subjects. The levels of exhaled cys-LTs were lower in those AIA subjects who received inhaled steroid therapy [26]. Following bronchial aspirin challenge levels of particular cys-LTs showed some variations but the total concentration of cys-LTs remained unchanged in both study groups. Up till now, cys-LTs level in EBC has not been measured after local (inhaled) aspirin administration. However, it has been shown [28] that cys-LTs levels in EBC after oral (systemic administration) aspirin challenge increased significantly in subjects with AIA. Varying results of different studies implicate that levels of cys-LTs in EBC cannot be a convenient indicator of asthma phenotype as their level after challenge test possibly depends on manner of aspirin administration and probably the doses of steroids inhaled. Results can be surprising because, a key enzyme-LTC4 synthase, is overexpressed in bronchial mucosa of patient with AIA [39]. Furthermore, circulating steroid blood eosinophils-main source of LTE 4 -carry more mRNA transcripts for this enzyme [40].
As was reported in childhood asthma [41] and adult asthmatics [27], 15-LO analogues of cys-LTs, eoxins C 4 , D 4 , and E 4 , showed no increase at baseline in asthmatic subjects independent of aspirin hypersensitivity. Their concentrations were not changed by the bronchial aspirin challenge. Role of those eicosanoids has been recently investigated in aspirin hypersensitive patients.
The high-sensitivity eicosanoid profiling of lipoxygenation products (5HETE, 15HETE) in EBC makes it possible to detect alterations in asthma, especially in its distinct phenotype characterized by hypersensitivity to aspirin and other nonsteroidal anti-inflammatory drugs. Cysteinyl-leukotriene levels in EBC after aspirin challenge in stable asthma patients, not steroid naive, most probably cannot be used as a reliable and sensitive index for aspirin hypersensitivity. In stable AIA patients on chronic inhaled steroid therapy of global (urinary) rather than in local (breath condensate) production of postchallenge cys-LTs is of greater and more sensitive value for aspirin hypersensitivity. We believe that quantitate cell analysis and measurements of released eicosanoides in induced-sputum will be more applicable for that purpose.