Related Experiment Video
Updated: Nov 19, 2025
Constructing and Visualizing Models using Mime-based Machine-learning Framework
Published on: July 22, 2025
Proposed global prognostic score for systemic mastocytosis: a retrospective prognostic modelling study
Javier I Muñoz-González1, Iván Álvarez-Twose2, María Jara-Acevedo3
1Cancer Research Center-IBMCC-USAL-CSIC, Department of Medicine and Cytometry Service-Nucleus Platform, Centro de Investigación Biomédica en Red de Cáncer (CIBERONC), University of Salamanca, Salamanca, Spain; Biomedical Research Institute of Salamanca (IBSAL), Salamanca, Spain; Spanish Network on Mastocytosis, Toledo and Salamanca, Spain.
A new Global Prognostic Score for Mastocytosis (GPSM) accurately predicts progression-free survival (PFS) and overall survival (OS) in systemic mastocytosis patients. This validated model uses accessible biomarkers for improved risk stratification and patient outcome prediction.
Area of Science:
- Hematology
- Oncology
- Medical Diagnostics
Background:
- Systemic mastocytosis (SM) lacks directly compared risk stratification models.
- Existing models do not fully leverage all available prognostic factors for patient outcome prediction.
Purpose of the Study:
- To design and validate a novel risk stratification model for SM.
- To compare the predictive accuracy of the new model against existing risk scores for progression-free survival (PFS) and overall survival (OS).
Main Methods:
- Retrospective prognostic modeling study using discovery (n=422) and validation (n=853) cohorts.
- Multivariable analysis identified independent prognostic factors for PFS and OS.
- Developed the Global Prognostic Score for Mastocytosis (GPSM) and compared its predictive capacity (C-index) with pre-existing models in a total of 1275 patients.
Main Results:
- The GPSM-PFS and GPSM-OS models demonstrated clear discrimination between risk groups for PFS and OS.
- GPSM-PFS showed high accuracy for PFS prediction (C-index 0.90), outperforming existing models.
- GPSM-OS accurately predicted OS in the entire cohort (C-index 0.92) and showed capacity in advanced SM.
Conclusions:
- The GPSM models offer robust prognostication for systemic mastocytosis.
- The study highlights the potential for prospective risk stratification using accessible biomarkers.
- GPSM-PFS and GPSM-OS emerged as highly accurate predictive models for survival outcomes in SM.
