SFEBES2023 Oral Communications Reproductive Endocrinology (6 abstracts)
Unsupervised steroid metabolome cluster analysis to dissect androgen excess and metabolic dysfunction in 488 women with polycystic ovary syndrome – results from the prospective DAISy-PCOS study
Eka Melson 1,2 , Thais P. Rocha 2,3 , Roland J. Veen 4 , Lida Abdi 1 , Tara McDonnell 5 , Veronika Tandl 6 , James M. Hawley 2,7 , Laura B. L. Wittemans 8 , Amarah V. Anthony 2 , Lorna C. Gilligan 2 , Fozia Shaheen 2 , Punith Kempegowda 2 , Caroline D.T. Gillett 2 , Leanne Cussen 5 , Cornelia Missbrenner 6 , Fannie Lajeunesse-Trempe 9 , Helena Gleeson 10 , Aled Rees 11 , Lynne Robinson 12 , Channa Jayasena 13 , Harpal S. Randeva 14 , Georgios K. Dimitriadis 9,15 , Larissa G. Gomes 3 , Alice J. Sitch 16,17 , Eleni Vradi 18 , Barbara Obermayer-Pietsch 6 , Michael W. O’Reilly 5 , Angela E. Taylor 2 , Michael Biehl 4 & Wiebke Arlt 1,17,19,2
1Medical Research Council London Institute of Medical Sciences (MRC LMS), London, United Kingdom. 2University of Birmingham, Birmingham, United Kingdom. 3Disciplina de Endocrinologia e Metabologia, Faculdade de Medicina da Universidade de São Paulo, São Paulo, Brazil. 4University of Groningen, Bernoulli Institute for Mathematics, Computer Science and Artificial Intelligence, Gronigen, Netherlands. 5Royal College of Surgeons in Ireland, Endocrinology Research Group, Department of Medicine, Dublin, Ireland. 6Medical University of Graz, Division of Endocrinology and Diabetology, Department of Internal Medicine, Graz, Austria. 7Wythenshawe Hospital, Manchester University NHS Foundation Trust, Department of Clinical Biochemistry, Manchester, United Kingdom. 8University of Oxford, Big Data Institute, Oxford, United Kingdom. 9King’s College Hospital NHS Foundation Trust, Department of Endocrinology, London, United Kingdom. 10Queen Elizabeth Hospital, University Hospitals Birmingham NHS Foundation Trust, Department of Endocrinology, Birmingham, United Kingdom. 11Cardiff University, Neuroscience and Mental Health Research Institute, School of Medicine, Cardiff, United Kingdom. 12Birmingham Women’s Hospital, Birmingham Women’s and Children’s Hospital NHS Foundation Trust, Birmingham, United Kingdom. 13Imperial College London, Department of Metabolism, Digestion and Reproduction, London, United Kingdom. 14University of Warwick, Warwick Medical School, Coventry, United Kingdom. 15King’s College London, Obesity, Type 2 Diabetes and Immunometabolism Research Group, Faculty of Cardiovascular and Metabolic Medicine & Sciences, School of Life Course Sciences, London, United Kingdom. 16University of Birmingham, Institute of Applied Health Research, Birmingham, United Kingdom. 17National Institute for Health and Care Research (NIHR) Birmingham Biomedical Research Centre, Birmingham, United Kingdom. 18Bayer AG, Berlin, Germany. 19Imperial College London, Institute of Clinical Sciences, Faculty of Medicine, London, United Kingdom
Introduction: Polycystic ovary syndrome (PCOS) affects 10% of women and is associated with a 2-3fold risk of type 2 diabetes (T2D), hypertension, fatty liver and cardiovascular disease. Androgen excess has been implicated as a major contributor to metabolic risk in PCOS. We aimed to identify PCOS sub-types with distinct androgen profiles and compare their cardiometabolic risk.
Methods: We cross-sectionally studied 488 treatment-naïve women with PCOS from UK & Ireland, Austria and Brazil (Age 28[24-32] years; BMI 27.5[22.4-34.6] kg/m2). Standardised assessments included bloods before and during 120min oral glucose tolerance test. We quantified 11-androgenic serum steroids by tandem mass spectrometry, followed by unsupervised k-means clustering of steroid data and statistical comparison of differences in clinical phenotype and metabolic parameters.
Results: Machine learning identified three distinct PCOS subgroups characterised by gonadal-derived androgen excess (GAE; 21.5% of women; lead steroids testosterone, dihydrotestosterone), adrenal-derived androgen excess (AAE; 21.7%; 11-ketotestosterone, 11-hydroxytestosterone) and comparably mild androgen excess (MAE; 56.8%), with similar age and BMI. Compared to GAE and MAE, the AAE cluster had the highest rates of hirsutism (76.4% vs. 67.6% vs. 59.9%) and alopecia (32.1% vs. 14.3% vs. 21.7%). The AAE cluster had the highest HOMA-IR and lowest Matsuda insulin sensitivity index (all P<0.01) and a 2-3fold higher incidence of impaired glucose tolerance (IGT) and newly diagnosed T2D. We achieved recruitment of 27% non-white women to the UK & Ireland cohort (n =208), with South Asian women more likely to be in the AAE cluster compared to white women (59% vs. 35%).
Conclusion: Unsupervised cluster analysis revealed three PCOS subtypes with distinct androgen excess profiles. The AAE cluster was characterised by highest insulin resistance, IGT and T2D, implicating 11-oxygenated androgens as drivers of metabolic risk. These results provide proof-of-principle for a novel metabolic risk prediction tool in PCOS that could guide future preventative and therapeutic strategies.
Volume 94
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