Perioperative predictors of biochemical remission in acromegaly: a predictive modeling study
Authors' conclusion
Predictive modeling may help estimate biochemical remission after surgery for acromegaly.
Study context
- Population
- Patients with acromegaly who underwent endoscopic endonasal transsphenoidal resection of GH-secreting primary pituitary adenomas, January 2018 to January 2025
- Participants
- 86 patients
- Outcome
- Biochemical remission after transsphenoidal resection
Methodological detailspredictive modeling, machine learning, random training/test split, Missing data addressed by median and KNN imputation, Five ML models (GBM, RF, GLMNET, KNN, Nnet) trained with exhaustive feature subset selection, evaluated by AUROC and accuracy
- Predictive modeling
- Machine learning
- Random training/test split
- Missing data addressed by median and KNN imputation
- Five ML models (GBM, RF, GLMNET, KNN, Nnet) trained with exhaustive feature subset selection, evaluated by AUROC and accuracy
Main finding
As reported in the abstract
- Achieved biochemical remission
- 62 (72.1%)
- Remission associated with older age
- p = 0.016
- Remission associated with round tumor shape
- p < 0.0001
- Remission associated with gross total resection
- p < 0.0001
- Remission associated with lower Knosp grade
- p < 0.0001
- Remission associated with dense CAM 5.2 staining
- p = 0.037
- Best model (RF, 4 features: gender, tumor shape, extent of resection, CAM 5.2) accuracy, test set
- 0.8235
- Best model (RF) AUROC, test set
- 0.8542
Observational data: associations, not effects.
Original publication
Perioperative predictors of biochemical remission in acromegaly: a predictive modeling study. Pituitary. Published 27 July 2026. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →