Non-invasive blood-based detection of endometriosis can improve standard-of-care by facilitating early diagnosis and clinical management among symptomatic women - PubMed
4 days ago
- #biomarkers
- #endometriosis
- #diagnosis
- Non-invasive blood-based detection of endometriosis improves standard-of-care by enabling early diagnosis and clinical management.
- Study aimed to develop and validate a blood-based diagnostic assay for endometriosis, accurate across menstrual cycle phases.
- Multi-center case-control study with machine learning classification and independent cohort validation involved 298 reproductive-age women.
- The assay integrated microRNAs, protein biomarkers, steroid hormone, age, and BMI using a random forest model.
- Achieved AUC of 0.944, sensitivity of 0.80, and specificity of 0.975 in the independent validation cohort.
- Consistent performance across menstrual cycle phases: proliferative (AUC 0.935) and secretory (AUC 0.993).
- Identified 61.5% of histologically confirmed endometriosis cases missed by imaging modalities like transvaginal ultrasound and MRI.
- Potential to improve early detection and guide timely clinical intervention, with ongoing prospective validation in diverse populations.