Exhaled Breath Condensate in Idiopathic Pulmonary Fibrosis
Exhaled Breath Condensate in Idiopathic Pulmonary Fibrosis
LPA has emerged as an important pro-fibrotic mediator in multiple organ systems, particularly the lungs, and the first clinical trial of an LPA receptor antagonist has recently been initiated in IPF patients (ClinicalTrials.gov identifier: NCT01766817). In this study, we analyzed the exhaled breath condensate (EBC) and plasma from eleven IPF patients and eleven controls without lung disease for the presence of lysophosphatidic acid (LPA), using liquid chromatography-tandem mass spectrometry (LC-MS/MS). We demonstrated that at least nine LPA species are detectable in EBC, and that one of these species, docosatetraenoyl (22:4) LPA, is significantly elevated in the EBC of IPF patients compared to controls. Thirteen LPA species were detectable in plasma; however, none of these differed significantly between the two groups.
Multiple species of LPA exist in biological fluids and are identified according to the composition of their fatty acid side chain. While all LPA species are thought to signal through LPA receptors, there are data indicating that the different species may have differing affinities for the various receptors. Very little is known about 22:4 LPA specifically, and it is unclear whether or not its signaling profile differs significantly from that of other LPA species. Notably, unsaturated LPA species appear to have higher affinity for most LPA receptors than do saturated species. In particular, long chain, polyunsaturated LPA species (like 22:4 LPA) have been shown to be the most potent activators of certain biological processes, such as platelet activation. Therefore, it is possible that 22:4 LPA may have more potent pro-fibrotic effects compared to other LPA species, and that the increase in 22:4 LPA in the EBC of IPF patients may be playing a role in driving the disease process. It should be noted, however, that the amount of 22:4 LPA in EBC was only a small fraction of total LPA, which may argue against a significant pathophysiological role for this particular LPA species in IPF. The increase in 22:4 LPA may instead indicate the generation of LPA from a specific a specific source, such as lung epithelial cells, which are known to contain high levels of polyunsaturated phospholipids.
In addition to being a therapeutic target, LPA may also serve as a useful biomarker for IPF. Elevations in LPA have been detected in the bronchoalveolar lavage (BAL) fluid from mice after intratracheal bleomycin administration and from humans with known IPF. 22:4 LPA was not specifically measured in this previous report of IPF patients, but it is detectable in BAL fluid, and it and other long-chain, polyunsaturated LPA species have been found to be elevated in BAL fluid in a mouse model of asthma and in human allergic airway inflammation. Our data suggest that EBC 22:4 LPA levels may be a useful biomarker for IPF diagnosis and/or prognosis. From a diagnostic standpoint, our data demonstrate minimal overlap between EBC 22:4 LPA levels in IPF patients and controls. To be of true value in the diagnosis of IPF, EBC 22:4 LPA levels would have to be able to differentiate between IPF and other forms of chronic interstitial lung diseases, most notably nonspecific interstitial pneumonia (NSIP) and chronic hypersensitivity pneumonitis (HP). As such comparisons were not performed in this study, further research would be needed to fully evaluate the potential role of EBC 22:4 LPA levels as a diagnostic biomarker in IPF.
It is notable that the EBC 22:4 LPA level in one patient was far outside the standard deviation of the mean, and that this patient was in the midst of an IPF exacerbation at the time of sample collection. This observation raises the hypothesis that EBC 22:4 LPA levels may be a useful biomarker of disease activity and/or acute exacerbations in IPF. Analysis of our data failed to reveal an association between EBC 22:4 LPA levels and disease severity or outcomes (decline in pulmonary function or mortality), although this study was likely underpowered to detect any such associations. Further study of EBC LPA levels in IPF, specifically in patients with rapidly progressive disease and those suffering from acute exacerbations, may shed light on the potential role of EBC 22:4 LPA levels as a prognostic biomarker in this disease.
While BAL has long been considered the optimal means of sampling the alveolar surfaces for analysis, it is invasive and not without risk, especially in subgroups of patients with advanced respiratory disease, such as those with pulmonary fibrosis. In comparison, EBC provides a method for non-invasive sampling of the lower respiratory tract. There are concerns regarding the accuracy with which EBC reflects the distal lung microenvironment, however, as there is risk of contamination with oral and gastrointestinal secretions, as well as an unknown (and potentially variable) dilution factor due to condensed water vapor. Recommendations regarding optimized EBC collection have been made to minimize contamination and variations in solute dilution. In our current study, it is reassuring that the two most abundant LPA species detected in EBC (16:0 and 18:0) were also the two most abundant species measured in BAL fluid from control subjects in the IPF (unpublished data) and asthma studies referenced above, suggesting that our EBC samples accurately reflect the distal lung compartments. Furthermore, the total LPA levels in our EBC samples are similar to those seen in BAL fluid, with respect to both the mean values and the standard deviations, suggesting that the dilution factors (and the variability thereof) may be similar for these two sample types.
Additional limitations to this study exist. Most notably, the sample size of the study was small. Though we were able to detect a difference in the amount of 22:4 LPA in IPF subjects versus controls, we were likely underpowered to detect differences in the other LPA species should any exist. While the subjects in the study population of interest all met current consensus guidelines for diagnosis of IPF, they differed in disease severity. Whether or not significant differences in LPA species could be detected when stratified for disease severity is not known. However, despite our limitations, we were able to detect a significant elevation in the amount of 22:4 LPA in EBC from IPF subjects. This study builds on previous work showing that LPA is increased in BAL fluid in IPF patients, and advances the current field of pulmonary research by showing that LPA can be extracted from EBC. Additional research is needed to determine any relationships between LPA species detectable in EBC and disease severity or progression in IPF.
Discussion
LPA has emerged as an important pro-fibrotic mediator in multiple organ systems, particularly the lungs, and the first clinical trial of an LPA receptor antagonist has recently been initiated in IPF patients (ClinicalTrials.gov identifier: NCT01766817). In this study, we analyzed the exhaled breath condensate (EBC) and plasma from eleven IPF patients and eleven controls without lung disease for the presence of lysophosphatidic acid (LPA), using liquid chromatography-tandem mass spectrometry (LC-MS/MS). We demonstrated that at least nine LPA species are detectable in EBC, and that one of these species, docosatetraenoyl (22:4) LPA, is significantly elevated in the EBC of IPF patients compared to controls. Thirteen LPA species were detectable in plasma; however, none of these differed significantly between the two groups.
Multiple species of LPA exist in biological fluids and are identified according to the composition of their fatty acid side chain. While all LPA species are thought to signal through LPA receptors, there are data indicating that the different species may have differing affinities for the various receptors. Very little is known about 22:4 LPA specifically, and it is unclear whether or not its signaling profile differs significantly from that of other LPA species. Notably, unsaturated LPA species appear to have higher affinity for most LPA receptors than do saturated species. In particular, long chain, polyunsaturated LPA species (like 22:4 LPA) have been shown to be the most potent activators of certain biological processes, such as platelet activation. Therefore, it is possible that 22:4 LPA may have more potent pro-fibrotic effects compared to other LPA species, and that the increase in 22:4 LPA in the EBC of IPF patients may be playing a role in driving the disease process. It should be noted, however, that the amount of 22:4 LPA in EBC was only a small fraction of total LPA, which may argue against a significant pathophysiological role for this particular LPA species in IPF. The increase in 22:4 LPA may instead indicate the generation of LPA from a specific a specific source, such as lung epithelial cells, which are known to contain high levels of polyunsaturated phospholipids.
In addition to being a therapeutic target, LPA may also serve as a useful biomarker for IPF. Elevations in LPA have been detected in the bronchoalveolar lavage (BAL) fluid from mice after intratracheal bleomycin administration and from humans with known IPF. 22:4 LPA was not specifically measured in this previous report of IPF patients, but it is detectable in BAL fluid, and it and other long-chain, polyunsaturated LPA species have been found to be elevated in BAL fluid in a mouse model of asthma and in human allergic airway inflammation. Our data suggest that EBC 22:4 LPA levels may be a useful biomarker for IPF diagnosis and/or prognosis. From a diagnostic standpoint, our data demonstrate minimal overlap between EBC 22:4 LPA levels in IPF patients and controls. To be of true value in the diagnosis of IPF, EBC 22:4 LPA levels would have to be able to differentiate between IPF and other forms of chronic interstitial lung diseases, most notably nonspecific interstitial pneumonia (NSIP) and chronic hypersensitivity pneumonitis (HP). As such comparisons were not performed in this study, further research would be needed to fully evaluate the potential role of EBC 22:4 LPA levels as a diagnostic biomarker in IPF.
It is notable that the EBC 22:4 LPA level in one patient was far outside the standard deviation of the mean, and that this patient was in the midst of an IPF exacerbation at the time of sample collection. This observation raises the hypothesis that EBC 22:4 LPA levels may be a useful biomarker of disease activity and/or acute exacerbations in IPF. Analysis of our data failed to reveal an association between EBC 22:4 LPA levels and disease severity or outcomes (decline in pulmonary function or mortality), although this study was likely underpowered to detect any such associations. Further study of EBC LPA levels in IPF, specifically in patients with rapidly progressive disease and those suffering from acute exacerbations, may shed light on the potential role of EBC 22:4 LPA levels as a prognostic biomarker in this disease.
While BAL has long been considered the optimal means of sampling the alveolar surfaces for analysis, it is invasive and not without risk, especially in subgroups of patients with advanced respiratory disease, such as those with pulmonary fibrosis. In comparison, EBC provides a method for non-invasive sampling of the lower respiratory tract. There are concerns regarding the accuracy with which EBC reflects the distal lung microenvironment, however, as there is risk of contamination with oral and gastrointestinal secretions, as well as an unknown (and potentially variable) dilution factor due to condensed water vapor. Recommendations regarding optimized EBC collection have been made to minimize contamination and variations in solute dilution. In our current study, it is reassuring that the two most abundant LPA species detected in EBC (16:0 and 18:0) were also the two most abundant species measured in BAL fluid from control subjects in the IPF (unpublished data) and asthma studies referenced above, suggesting that our EBC samples accurately reflect the distal lung compartments. Furthermore, the total LPA levels in our EBC samples are similar to those seen in BAL fluid, with respect to both the mean values and the standard deviations, suggesting that the dilution factors (and the variability thereof) may be similar for these two sample types.
Additional limitations to this study exist. Most notably, the sample size of the study was small. Though we were able to detect a difference in the amount of 22:4 LPA in IPF subjects versus controls, we were likely underpowered to detect differences in the other LPA species should any exist. While the subjects in the study population of interest all met current consensus guidelines for diagnosis of IPF, they differed in disease severity. Whether or not significant differences in LPA species could be detected when stratified for disease severity is not known. However, despite our limitations, we were able to detect a significant elevation in the amount of 22:4 LPA in EBC from IPF subjects. This study builds on previous work showing that LPA is increased in BAL fluid in IPF patients, and advances the current field of pulmonary research by showing that LPA can be extracted from EBC. Additional research is needed to determine any relationships between LPA species detectable in EBC and disease severity or progression in IPF.