ECE2022 Poster Presentations Endocrine-Related Cancer (41 abstracts)
Urinary steroid metabolomics for adrenocortical cancer diagnosis. Comparison of gas chromatography mass spectrometry to liquid chromatography mass spectrometry
Angela Taylor 1 , Irina Bancos 2,3 , Lorna Gilligan 1 , Rick van Veen 4 , Vasileios Chortis 1 , Fozia Shaheen 1 , Carl Jenkinson 1 , Donna M O’Neil 1 , Beverly Hughes 1 , James M Hawley 1,5 , Brian Keevil 5 , Cedric H L Shackelton 6 , Jonathan Deeks 7 , Alice J Sitch 7 , Michael Biehl 4 & Wiebke Arlt 1
1Institute of Metabolism and Systems Research, Medical School, University of Birmingham, birmingham, United Kingdom; 2Division of Endocrinology, Diabetes, Metabolism and Nutrition, Mayo Clinic, Rochester, United States; 3Mayo Clinic, Division of Endocrinology, Metabolism, Diabetes and Nutrition, Department of Internal Medicine, United States; 4Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, University of Groningen, Groningen, Netherlands; 5Wythenshawe Hospital, Biochemistry Department, Manchester, United Kingdom; 6UCSF, Beniott Children’s Hospital, Oakland, United States; 7Institute of Applied Health Research, University of Birmingham, Birmingham, United Kingdom
Introduction: Gas chromatography mass spectrometry (GC-MS) is the gold standard method for urinary steroid profiling. However, GC-MS requires chemical derivatisation, long run times, is labour intensive, expensive, and unsuitable for rapid multi-sample analysis, limiting its use in routine clinical practice. GC-MS urinary steroid metabolomics, the combination of steroid profiling and machine learning (Generalized Matrix Learning Vector Quantization) was shown to have superior specificity and sensitivity for adrenocortical carcinoma (ACC) diagnosis compared to imaging technologies (1). The method has subsequently been transferred to liquid chromatography tandem mass spectrometry (LC-MS/MS), selecting 15 diagnostically relevant steroids, decreasing the complexity and the cost of the assay. This method was applied to the EURINE-ACT cohort, 2017 prospectively recruited adrenal tumours patients through an ENS@T collaboration (2). Here we compare GC-MS to LC-MS/MS to evaluate the differences in quantitation and ACC diagnostic ability.
Experimental and Results: After deconjugation the steroid extract was either derivatised for GC-MS analysis (Agilent MSD 5975 with a DB1 column) or run directly via LC-MS/MS (Waters-Xevo with Acquity uPLC, HSS T3 column). Correlation between the two technologies was investigated by comparing steroid quantitation in 481 urines from a range of endocrine conditions, including a healthy control cohort 129 urines (75/54 female/male, 20-81 years). Correlation plots and Bland-Altman plots were used to assess method agreement. To compare diagnostic ability urines from 40 patients with adrenal carcinoma (17/23 female/male, 22-79 years, tumour size 50-230 mm) and 99 patients with non-cancerous adrenal tumours (61/38 female/male, 29-83 years, tumour size 9-55 mm) were assessed. Diagnostic ability was determined via calculation of the area under receiver operated characteristic curve (AUROC). There were statistically significant correlations between the methods for all steroids. The diagnostic ability, AUROC for 31 steroids by GC-MS was 0.969, (SD=0.044), and for 15 steroids by LC-MS/MS, was 0.954 (0.067). The highest estimated sensitivity=specificity was LC-MS/MS for 15 steroids (0.901), followed by GC-MS 31 steroids (0.890).
Conclusions: Despite differences in sample preparation and mass spectrometer design GC-MS and LC-MS/MS showed significantly similar quantitation for all steroids. Reduction of the number of analytes from 31 by GC-MS to 15 by LC-MS/MS does not impact the diagnostic ability for ACC diagnosis. LC-MS/MS should now be introduced into clinical biochemistry laboratories as a routine test for the diagnostic work up for patients with adrenal tumours.
1. Arlt W, et al. J Clin Endocrinol Metab. 2011 96(12):3775-84.
2. Bancos I Taylor AE, et al. Lancet Diabetes Endocrinol. 2020 8(9):773-781.
Article tools
My recent searches
My recently viewed abstracts
Authors
Endocrine Abstracts
ISSN 1470-3947 (print) | ISSN 1479-6848 (online)
© Bioscientifica 2024 |
Privacy policy |
Cookie settings
BiosciAbstracts
Bioscientifica Abstracts is the gateway to a series of products that provide a permanent, citable record of abstracts for biomedical and life science conferences.
Find out more