Journal Article Ahead of print
PubMed: 42535554 DOI: 10.1093/ejendo/lvag141
Baseline biochemical evaluation for etiologic differentiation in ACTH-dependent Cushing's syndrome: machine learning and composite score analysis in a multicenter cohort of 566 patients.
Eur J Endocrinol
Authors: Mario Detomas, Juan P Prada Salcedo, Barbara Altieri, Isabel Stüfchen, Isabella Nasi-Kordhishti, Stefan Wolfsberger, Marco Losa, Marcus Quinkler, Otilia Kimpel, Greisa Vila, Jürgen Bernd Honegger, German Rubinstein, Martin Fassnacht, Timo Deutschbein
Affiliations
- Department of Internal Medicine I, Division of Endocrinology and Diabetes, University Hospital, University of Würzburg, Würzburg, Germany.
- Deptartment of Bioinformatics, University of Würzburg, Biocenter/Am Hubland, 97074 Würzburg, Germany.
- Medizinische Klinik und Poliklinik IV, Klinikum der Universität München, Munich, Germany.
- Department of Neurosurgery, University of Tübingen, Tübingen, Germany.
- Department of Neurosurgery, Medical University of Graz, Styria, Austria.
- Department of Neurosurgery, Medical University of Vienna, Vienna, Austria.
- Department of Neurosurgery, IRCCS San Raffaele Scientific Institute, Vita-Salute University, Milan, Italy.
- Endocrinology in Charlottenburg, Berlin, Germany.
- Clinical Division of Endocrinology and Metabolism, Department of Internal Medicine III, Medical University of Vienna, Vienna, Austria.
- Medicover Oldenburg MVZ, Oldenburg, Germany.
Abstract
BACKGROUND
Limited availability of corticotropin-releasing hormone (CRH) currently complicates the differentiation of adrenocorticotropin (ACTH)-dependent Cushing's syndrome (CS). The diagnostic value of common screening tests in distinguishing Cushing's disease (CD) from ectopic CS (ECS) remains unclear.
OBJECTIVE
To assess the diagnostic performance of screening tests, alone and in combination, in differentiating CD from ECS.
METHODS
Retrospective multicenter study enrolling patients with confirmed ACTH-dependent CS and available screening tests at diagnosis. Data are expressed as multiples of upper limit of normal (×ULN). Optimal cut-offs were determined using Youden's Index. Combination with composite score models and machine learning (ML) algorithm were performed.
RESULTS
A total of 566 patients were included (509 [90%] with CD). The optimal morning ACTH cut-off was 1.8×ULN (sensitivity 74%, specificity 77%, AUC=0.776 [CI-95% 0.688-0.853]). 24h-urinary free cortisol (24h-UFC) showed the best performance (cut-off 5.9×ULN, sensitivity 72%, specificity 83%, AUC=0.854 [0.816-0.923]), followed by the 1 mg dexamethasone suppression test (12.8×ULN, sensitivity 70%, specificity 86%, AUC=0.828 [0.740-0.906]). Using a composite score, a cut-off of 1.5 yielded 78% sensitivity and 92% specificity (AUC=0.865 [0.798-0.933]). Combining this score with pituitary MRI-findings, sensitivity and specificity were 88% and 85% (AUC=0.931 [0.887-0.974]). ML algorithm (balanced random forest) including screening tests-results yielded a sensitivity of 71% and specificity of 84% (AUC=0.853 [0.741-0.966]), while including the MRI-findings variant achieved a sensitivity and specificity of 76% and 91% (AUC=0.902 [0.805-0.999]).
CONCLUSION
Combining screening tests in composite scoring and ML differentiates ACTH-dependent CS and offers a potential diagnostic alternative to CRH-stimulation test.
MeSH
Algorithms, Hormones, Machine Learning, Learning, Adrenocorticotropic Hormone, Endocrine System Diseases, Cushing Syndrome, Sensitivity And Specificity