AI Science Tracker
EN

Evaluating comorbidity scoring systems for flumatinib therapy in chronic myeloid leukemia: a machine learning and SHAP-based predictive analysis.

Guizhou Medical University, China; Department of Hematology, People's Hospital of Qianxinan Buyi and Miao Autonomous Prefecture, China · 28 May 2026 · medicine

Discovery

Evaluating comorbidity scoring systems for flumatinib therapy in chronic myeloid leukemia: a machine learning and SHAP-based predictive analysis.

Source check: Source checked. Not independent scientific confirmation.

Finding: Among 559 retrospectively studied patients with chronic-phase chronic myeloid leukemia treated with flumatinib, XGBoost predicted 12-month major molecular response (AUC 0.852). Adding CIRS-G comorbidity scoring produced the largest reported incremental AUC (0.078; p=0.006) among the evaluated scores. This single-center retrospective association may overestimate prospective validity.

Technology: XGBoost predicted 12-month major molecular response; SHAP interpreted nonlinear feature associations and interactions, while CCI, ACE-27 and CIRS-G scores were compared for incremental predictive value.

Publication: Peer-reviewed paper · Test context: retrospective or benchmark. Catalog categories, not an evidence grade.

AI contribution
XGBoost predicted 12-month major molecular response; SHAP interpreted nonlinear feature associations and interactions, while CCI, ACE-27 and CIRS-G scores were compared for incremental predictive value.
Test and result
The publisher abstract reports retrospective data from 559 patients treated during 2018–2024, XGBoost AUC 0.852 for 12-month major molecular response, and the largest incremental value when CIRS-G was added to the baseline model (ΔAUC 0.078, p=0.006). SHAP analyses were exploratory; the study notes that temporal variation and retrospective design may overstate prospective validity.
Limitation / still unknown
The single-center retrospective model does not establish prospective performance or clinical benefit; the authors caution that temporal variation may overestimate validity.
Evidence sources and annotation context

Evidence context (descriptions follow the cited sources; test type is a catalog classification):

Evidence sources: https://www.frontiersin.org/journals/medicine/articles/10.3389/fmed.2026.1849735/full

Source checked · editorial details

The source page was checked against this entry. Independent confirmation is not assessed by this label.

Source type: paper. Paper: not linked in this catalog; this does not mean no paper exists.

Review date: 22 Nov 2026. Review scope: Primary result-bearing material for the stated method, result and evaluation, with bibliographic version information..

No follow-up recorded; this is not a claim that no follow-up exists. Follow-up review scope and date are not recorded. No source-supported relationships recorded.

Editorial assessment: Impact 2/5 · Validation 2/5 · Novelty 3/5. Impact and novelty are editorial judgments; validation follows the annotated test type where one exists. Not an evidence grade. Scoring criteria.

Open in the full catalog for filters, charts and related discoveries. Browse medicine discoveries.