Risk Prediction Models for Chemotherapy-Induced Myelosuppression: A Systematic Review
Year: 2025
Type: review
Study design: Systematic Review
Abstract: Aims/Background Chemotherapy-induced myelosuppression (CIM) is associated with increased risks of life-threatening complications, treatment delays, and reduced therapeutic efficacy. However, the predictors contributing to these risks remain unclear. Therefore, this study aimed to systematically review the existing risk prediction models developed for CIM and to evaluate prognostic factors associated with patient outcomes. Methods We comprehensively searched domestic and international databases for literature on CIM risk prediction models, covering records from database inception to 31 December 2024. Two researchers independently performed literature screening and data extraction. The risk of bias and applicability of included studies were assessed using the Prediction Model Risk of Bias Assessment Tool (PROBAST). Results A total of 17 risk prediction models developed for CIM were identified. The area under the curve (AUC) of the selected models ranged from 0.708 to 0.95. Among the six models that underwent external validation, AUC values ranged from 0.708 to 0.95. Fifteen models reported discriminative performance metrics (AUC) exceeding 0.70. The included models incorporated between 2 and 16 predictors, with chemotherapy regimen intensity, baseline haematological parameters (platelet count, haemoglobin, and neutrophil count), and age being the most frequently selected variables. Conclusion Current CIM risk prediction models demonstrate promising predictive performance, with clinically relevant predictors. However, high bias risks necessitate future optimisation through multicenter prospective studies, the integration of dynamic biomarkers, and standardised validation frameworks to enhance the utility of clinical decision-making.
Authors: Jing Han, Hongling Yin, Na Liu, Fuxia Li
Language: English
Subfield: Oncology
Field: Medicine
Domain: Health Sciences
Open Access status: hybrid
OpenAlex ID: W4416396207