-
1
This study evaluated CCI, ACE-27, and CIRS-G comorbidity scoring systems for predicting molecular responses in CP-CML patients treated with flumatinib.
-
2
The XGBoost model achieved the highest predictive performance with an AUC of 0.852, indicating strong predictive capabilities for treatment outcomes.
-
3
Integrating CIRS-G into the baseline model provided significant incremental value, outperforming both CCI and ACE-27 in predicting treatment efficacy.
-
4
SHAP analysis indicated that a CIRS-G score of 8 or higher may severely compromise therapeutic efficacy, highlighting the importance of comorbidity.
-
5
The study emphasizes the need for personalized treatment approaches in CP-CML, considering the complex interactions between age, comorbidities, and drug dosing.
