Crynodeb
Introduction: Respiratory diseases kill over 4 million people a year globally. The diagnostic tools for respiratory diseases are currently limited by such factors as lack of accuracy, cost, or patient discomfort. Respiratory diseases could be better monitored through the development of new biomarkers that could be assessed non-invasively. In this project, metabolomic methods were used to assess if they could detect changes in urines from people suffering from lung cancer (LC), Chronic pulmonary obstructive disease (COPD), pulmonary fibrosis (PF) or asthma. These could suggest biomarkers that could be used in the diagnosis and management of pulmonary diseases based on an easily obtained liquid biopsy.Methods: 204 samples (42=HC, 27=asthma, 48=LC, 81=COPD and 5=PF) and 142 samples (88=LC and 54=HC (aged matched)) were analysed using a flow infusion electrospray ion mass spectrometry (FIE-MS). The derived data were assessed using the R-based platform Metaboanalyst, using principal component analyses (PCA), partial least squares discriminant analysis (PLS-DA) and the receiver operating characteristics (ROC) curves.
Results: The PCA and PLS-DA from the first sample set indicated that the metabolites of COPD and asthma groups were similar to those of health controls. In contrast, LC and PF formed a distinctive cluster from the asthma-control group and each other. The sources of variation were highlighted in pairwise comparisons of LC vs HC and PF vs. HC which included T-test and ROC curve assessments. Top ranking significant (P = <-0.001) differences between LC and HC in the first sample set were galactosylceramide d18:1/24:1(15Z) (AUC= 0.73918), Creatine riboside (AUC = 0.72561), 3-(Carboxymethyl)-3hydroxypentanedioic acid (AUC = 0.7203), Adenosine phosphosulfate (AUC = 0.71479), 7,8dihydroneopterin 3'-phosphate (AUC = 0.70948), cis,cis,cis-10,13,16-Docosatrienoyl-CoA (AUC = 0.70712), Inosine (AUC = 0.70712), N-(2,4- Dinitrophenyl)-2,4-dinitroaniline (AUC = 0.70515), Prostaglandin M (AUC = 0.70201) and pentadecenoic acid (AUC = 0.70142). These differences were consistent with change in
nucleotide (adenosine phosphosulfate, inonine and creatine riboside), fatty acid processing (galactosylceramide d18:1/24:1(15Z), 3(Carboxymethyl)-3-hydroxypentanedioic acid, pentadecenoic acid) and possibly inflammatory metabolism (Prostaglandin M). Top ranking significant (P = <-0.001) differences between PF and HC were hypotaurine (AUC =0.95122), Quinaldic acid (AUC = 0.95122), Creosol (AUC = 0.89268), Isopropylmalic acid (AUC = 0.89268), L-Tyrosine ethyl ester (AUC = 0.89268), 13-Hydroxyoctadecadienoic acid (13HPODE; AUC=0.88293), Uridine 5'-monophosphate (AUC = 0.88293), asparaginyl-cysteine (AUC = 0.87317), phenylalanyl-Alanine (AUC = 0.86829), phosphodimethylethanolamine (AUC = 0.85854) Phosphatidylserine (18:0/15:0) (AUC = 0.85854). Whilst the numbers of PF samples were low these changes were consistent with differences in protein/amino acid/phenolic (cresol, asparaginyl-cysteine , phenylalanyl-alanine) metabolism which could be linked to fibrotic remodelling of the extracellular matrix. Also, prominent were changes in long chain fatty acids (13-HPODE, which would be linked to inflammatory events.
Conclusions: There are currently poor diagnostic tools for respiratory diseases. Metabolomic analyses have suggested that urine could be used to help in the diagnosis of certain respiratory diseases. These analyses identified metabolomic changes which were biologically coherent being either associated with cell division (nucleotide synthesis), inflammation (long chain lipid processing) or protein/amino acid/phenolic processing which aligned with features of each pathology. Thus, metabolomics using urine, which is minimally invasive, has great potential to be used as a future diagnostic test, especially in LC.
| Dyddiad Dyfarnu | 2023 |
|---|---|
| Iaith wreiddiol | Saesneg |
| Sefydliad Dyfarnu |
|
| Noddwyr | Knowledge Economy Skills Scholarships |
| Goruchwyliwr | Luis Mur (Goruchwylydd) |