6 Polycystic ovarian syndrome in adolescents: discovery proteomics and the search for novel non-invasive biomarkers

Abstract

Background Polycystic ovarian syndrome (PCOS) is common, affecting up to one-fifth of females and is associated with significant comorbidity. Despite this, it is poorly understood, and diagnosis and management remain challenging in adolescents. Proteomics enables the better understanding of disease mechanisms and facilitates the identification of novel biomarkers. Objectives To describe the clinical phenotype of PCOS in adolescents and undertake discovery proteomic urine profiling using ultra-performance liquid chromatography-mass spectrometry (UPLC-MS/MS) to identify novel non-invasive biomarkers of PCOS. Method This prospective longitudinal study recruited adolescent females meeting NIH diagnostic criteria for PCOS. The following were measured at baseline and annual follow-up: hormonal and metabolic markers including an oral glucose tolerance test, psychological, pubertal and anthropometric parameters, and pelvic ultrasounds. We have undertaken UPLC-MS/MS and developed new methods for discovery proteomic profiling of urine in an attempt to identify new disease mechanisms, drug targets and potential biomarkers. Results We recruited 40 participants (median age 15.0 years, range 12.5-18.3), with two-thirds completing annual follow-up. Clinical signs at presentation included acne (89%), hirsutism (78%) and acanthosis nigricans (49%). Two-thirds of participants had depressive or anxiety symptoms yet only one-third were known to mental health services. Metabolic dysfunction was common at baseline; overweight/obesity (86%), elevated body fat (88%) and dyslipidaemia (35%). These parameters persisted at follow-up. Insulin resistance was almost universal at baseline and follow-up (91%). Impaired glucose metabolism was common but improved from baseline (29%) to follow-up (10%; p=0.11). Over two-thirds had elevated anti-Mullerian hormone, three-quarters had an elevated free androgen index. Raised inflammatory markers (CRP/ESR) were present in 40% participants. Only three participants had definitive ultrasonographic evidence of PCOS. Interventions included lifestyle advice (27%), combined oral contraceptive pill (COCP) ± anti-androgen (16%), metformin (30%) or metformin + COCP ± anti-androgen (27%). Conclusion and future directions Adolescents withPCOS are at high risk of metabolic dysfunction, inflammation and mental health disorders. Therefore, early diagnosis and intervention are imperative. However, current diagnostic and surveillance methods are suboptimal. We have used urinary proteomics to study metabolic pathways affected in PCOS and aim to identify novel non-invasive biomarkers. Subsequently, we will create a clinically translatable assay to aid diagnosis and stratify management of this common adolescent condition.