Related Experiment Videos
Machine learning and cox model-based prediction of CDK4/6 inhibitor outcomes in HR+/HER2 - metastatic breast cancer:
Binliang Liu1, Tao Wu2, Sijuan Ding3
1Department of Breast Cancer Medical Oncology, The Affiliated Cancer Hospital of Xiangya School of Medicine, Hunan Cancer Hospital, Central South University, No. 283, Tongzipo Road, Yuelu District, Changsha, 410013, P. R. China.
Background:
Cyclin-dependent kinase 4/6 inhibitors (CDK4/6is) are standard therapy for HR+/HER2 - metastatic breast cancer (MBC), yet treatment outcomes vary substantially and individualized prognostic tools based on real-world data remain limited. This study evaluated real-world effectiveness and developed prognostic models integrating Cox regression and machine learning.
Methods:
This multicenter retrospective study included 1,008 HR+/HER2 - MBC patients treated with CDK4/6is across 20 cancer centers in central China. Treatment patterns were analyzed in the overall cohort. PFS and prognostic factors were evaluated in patients receiving first- or second-line CDK4/6is using Kaplan-Meier and Cox regression analyses. These patients were randomly divided into training and validation cohorts (7:3). A Cox model and seven machine learning algorithms (GBM, RSF, Lasso-Cox, CoxBoost, XGBoost, SuperPC, and plsRcox) were developed and compared using time-dependent AUC, calibration, and decision curve analysis.
Results:
CDK4/6is were used as first- and second-line therapy in 65.68% and 24.60% of patients. Median PFS was significantly longer in first-line versus second-line treatment (38.0 vs. 18.8 months, P < 0.001). Independent prognostic factors included Luminal B subtype, liver metastasis, and second-line treatment, while HER2 IHC 1 + and longer disease-free interval were favorable predictors. The Cox model showed good discrimination (AUCs: 0.731, 0.719, and 0.704). Among machine learning models, GBM and RSF showed higher discrimination in the training cohort but only moderate performance in the validation cohort.
Conclusion:
CDK4/6is demonstrated substantial real-world effectiveness. Both Cox and machine learning models enabled individualized prognostic prediction. GBM and RSF showed relatively better predictive performance, but external validation is still required before clinical application.
Insights
Cyclin-dependent kinase 4/6 inhibitors (CDK4/6is) show effectiveness in metastatic breast cancer. Both Cox and machine learning models can predict patient outcomes, with GBM and RSF demonstrating promising predictive performance.
Area of Science:
- Oncology
- Pharmacology
- Biostatistics
Background:
- Cyclin-dependent kinase 4/6 inhibitors (CDK4/6is) are a standard treatment for HR+/HER2- metastatic breast cancer (MBC).
- Significant variations in treatment outcomes highlight the need for individualized prognostic tools using real-world data.
- Current prognostic models for CDK4/6i therapy in MBC lack comprehensive real-world validation.
Purpose of the Study:
- To evaluate the real-world effectiveness of CDK4/6 inhibitors in HR+/HER2- MBC patients.
- To develop and compare prognostic models integrating Cox regression and machine learning for predicting treatment outcomes.
- To identify key prognostic factors influencing progression-free survival (PFS) in this patient population.
Main Methods:
- A multicenter retrospective study of 1,008 HR+/HER2- MBC patients treated with CDK4/6is in central China.
- Analysis of treatment patterns, progression-free survival (PFS), and prognostic factors using Kaplan-Meier and Cox regression.
- Development and comparison of a Cox model and seven machine learning algorithms (GBM, RSF, Lasso-Cox, CoxBoost, XGBoost, SuperPC, plsRcox) using training and validation cohorts.
Main Results:
- CDK4/6is were utilized as first-line (65.68%) and second-line (24.60%) therapy, with longer median PFS in first-line treatment (38.0 vs. 18.8 months).
- Independent prognostic factors for poorer outcomes included Luminal B subtype, liver metastasis, and second-line treatment.
- The Cox model demonstrated good discrimination (AUCs: 0.731-0.704); GBM and RSF showed higher discrimination in training but moderate performance in validation.
Conclusions:
- CDK4/6 inhibitors exhibit significant real-world effectiveness in HR+/HER2- MBC.
- Both Cox regression and machine learning models offer potential for individualized prognostic prediction.
- While GBM and RSF show promising predictive capabilities, external validation is crucial before clinical implementation.
