Combining metabolomics and machine learning models as a tool to distinguish non-classic 21-hydroxylase deficiency from polycystic ovary syndrome without adrenocorticotropic hormone testing

Artificial intelligence in paediatric endocrinology: conflict or cooperation

Artificial intelligence (AI) in medicine is transforming healthcare by automating system tasks, assisting in diagnostics, predicting patient outcomes and personalising patient care, founded on the ability to analyse vast datasets. In paediatric endocrinology, AI has been developed for diabetes, for insulin dose adjustment, detection of hypoglycaemia and retinopathy screening; bone age assessment and thyroid nodule screening; the identification of growth disorders; the diagnosis of precocious puberty; and the use of facial recognition algorithms in conditions such as Cushing syndrome, acromegaly, congenital adrenal hyperplasia and Turner syndrome. AI can also predict those most at risk from childhood obesity by stratifying future interventions to modify lifestyle. AI will facilitate personalised healthcare by integrating data from 'omics' analysis, lifestyle tracking, medical history, laboratory and imaging, therapy response and treatment adherence from multiple sources. As data acquisition and processing becomes fundamental, data privacy and protecting children's health data is crucial. Minimising algorithmic bias generated by AI analysis for rare conditions seen in paediatric endocrinology is an important determinant of AI validity in clinical practice. AI cannot create the patient-doctor relationship or assess the wider holistic determinants of care. Children have individual needs and vulnerabilities and are considered in the context of family relationships and dynamics. Importantly, whilst AI provides value through augmenting efficiency and accuracy, it must not be used to replace clinical skills.

Steroid profiling in adrenal disease

The measurement of steroid hormones in blood and urine, which reflects steroid biosynthesis and metabolism, has been recognized as a valuable tool for identifying and distinguishing steroidogenic disorders. The application of mass spectrometry enables the reliable and simultaneous analysis of large panels of steroids, ushering in a new era for diagnosing adrenal diseases. However, the interpretation of complex hormone results necessitates the expertise and experience of skilled clinicians. In this scenario, machine learning techniques are gaining worldwide attention within healthcare fields. The clinical values of combining mass spectrometry-based steroid profiles analysis with machine learning models, also known as steroid metabolomics, have been investigated for identifying and discriminating adrenal disorders such as adrenocortical carcinomas, adrenocortical adenomas, and congenital adrenal hyperplasia. This promising approach is expected to lead to enhanced clinical decision-making in the field of adrenal diseases. This review will focus on the clinical performances of steroid profiling, which is measured using mass spectrometry and analyzed by machine learning techniques, in the realm of decision-making for adrenal diseases.

The serum steroid signature of PCOS hints at the involvement of novel pathways for excess androgen biosynthesis

CONTEXT

Polycystic ovary syndrome (PCOS) is defined by androgen excess and ovarian dysfunction in the absence of a specific physiological diagnosis. The best clinical marker of androgen excess is hirsutism, while the best biochemical parameter is still a matter of debate. Current consensus guidelines recommend, among other hormones, serum free testosterone as an important serum parameter to measure androgen excess. Recently, however, novel active androgens and androgen metabolic pathways have been discovered.

OBJECTIVE

To assess the contribution of novel androgens and related steroid biosynthetic pathways to the serum steroid pool in PCOS women in comparison to healthy controls.

DESIGN

This is a case control study, wherein PCOS was diagnosed according to the AE-PCOS 2009 criteria. Serum steroid profiling was performed by liquid chromatography high-resolution mass spectrometry.

SETTING

Yeditepe University and associated clinics in Istanbul, Turkey, together with Bern University Hospital Inselspital, Bern, Switzerland.

PARTICIPANTS

42 PCOS women and 42 matched, healthy control women.

MAIN OUTCOME MEASURES

Assessment of 34 steroids compartmentalized in four androgen related pathways: the classic androgen pathway, the backdoor pathway, the C11-oxy backdoor pathway, and the C11-oxy (11β-hydroxyandrostenedione) pathway.

RESULTS

Metabolites of all four pathways were identified in healthy and PCOS women. Highest concentrations were found for progesterone in controls and androstenedione in PCOS. Lowest levels were found for 11-ketotestosterone in controls compared to PCOS, and for 20α-hydroxyprogesterone in PCOS compared to controls. PCOS also had higher serum testosterone levels compared to the controls. PCOS women had overall higher levels of steroid metabolites of all four androgen pathways compared to healthy controls.

CONCLUSIONS

Novel alternative pathways contribute to the androgen production in healthy and PCOS women. Hyperandrogenism in PCOS is characterized by an overall increase of serum androgens in the classic, backdoor and C11-oxy pathways. While monogenetic disorders of steroid biosynthesis can be recognized by a specific pattern in the steroid profile, no diagnostic pattern or classifier was found in the serum for PCOS.