The Multiple Myeloma (MM) Risk Calculator provides individualized risk predictions for newly diagnosed MM patients using machine learning models. It offers three model options: one incorporating cytogenetics, another excluding it, and a dynamic model that adjusts risk based on best response to 1st line therapy.
The output includes risk percentiles for Overall Survival (OS) and Progression-Free Survival (PFS), allowing comparison within a large NDMM cohort. Lower percentiles indicate lower risk. The calculator also stratifies results by transplant eligibility and treatment type—IMID-based, PI-based, or combined PI-IMID therapy—to aid in personalized clinical decision-making.
This Multiple Myeloma (MM) Risk Calculator utilizes machine learning (ML) models to provide individualized risk predictions based on clinical and cytogenetic data, enabling a more personalized approach to prognosis in newly diagnosed MM (NDMM) patients. The calculator offers risk assessments for both Overall Survival (OS) and Progression-Free Survival (PFS), depending on the available data input.
- Model with Cytogenetics: Uses clinical parameters (e.g., age, hemoglobin, albumin) alongside cytogenetic markers (1q gain and 17p deletion).
- Model without Cytogenetics: Incorporates core clinical variables without requiring cytogenetic data, useful when this information is unavailable.
- Treatment-related Model: Incorporates the initial treatment regimen (PI-based, IMID-based, or a combination of both) to enhance the accuracy of predictions.
- Dynamic Model: Adds the patient's best response to 1st line therapy to refine predictions.
- OS Percentile: Indicates the relative risk of mortality; a lower percentile suggests a lower risk compared to the overall population, while a higher percentile signals an elevated risk.
- PFS Percentile: Reflects the relative risk of disease progression or death; a lower percentile here suggests a lower risk, while a higher percentile implies a higher likelihood.
- Patient Stratification Based on Transplant Eligibility: Patient scores are compared against cohorts classified as either eligible or ineligible for autologous hematopoietic stem cell transplantation.
Mosquera Orgueira A, Gonzalez Perez MS, D'Agostino M, Cairns DA, Larocca A, Lahuerta Palacios JJ, et al. Development and Validation of Novel Machine Learning-Based Risk Scores for Multiple Myeloma: Insights from the Harmony Alliance Big Data Platform. ASH Annual Meeting Abstracts. 2024;803:2226. On behalf of the Harmony Alliance Consortium. Available at: https://ash.confex.com/ash/2024/webprogram/Paper201448.html