Tool Using Time-Varying Biomarkers Predicts the Risk of Progression from Precursor Disease to MM | ASH Clinical News | American Society of Hematology

For patients with precursor disease to multiple myeloma (MM), the standard for predicting disease progression is the 2/20/20 risk stratification criteria from the International Myeloma Working Group (IMWG). These criteria put patients into one of four risk groups – low, low-intermediate, high-intermediate, or high – based on information that includes levels of monoclonal protein, bone marrow plasma cell percentage, serum free light chain ratio, and cytogenetic results. This tool uses discrete cut-off values even though there is inherent variation in the biomarkers measured throughout disease monitoring.

Researchers from Dana-Farber Cancer Institute in Boston have developed a model that accounts for changes in myeloma-specific biomarkers over time, thereby improving the clinical prediction of disease progression to MM for patients diagnosed with monoclonal gammopathy of undetermined significance (MGUS) or smoldering multiple myeloma (SMM). This model, called PANGEA, was described in Lancet Haematology.

“The PANGEA model is the first precursor progression model that can be used accurately with or without data from bone marrow biopsies,” said study author Annie Cowan, BA, of Dana-Farber Cancer Institute. “[Our] study produced the most accurate model used to predict the risk of disease progression to [MM] in patients with precursor myeloma disease states.”

The PANGEA model incorporates dynamic measurements of monoclonal protein, serum free light chain ratio, creatinine, hemoglobin, age, and bone marrow plasma cell percentage to provide personalized risk-prediction estimates at future timepoints for individual patients.

The researchers used retrospective data from 6,441 patients with either MGUS (n=4,931) or SMM (n=1,510). Fifty-three percent of the participants were female. Biological and clinical variables at baseline and at serial subsequent timepoints were used to model the risk of progression to MM based on a Dana-Farber Cancer Institute patient cohort. Subsequently, the team validated the model using two independent cohorts from the National and Kapodistrian University of Athens together with the University College London and from the Registry of Monoclonal Gammopathies in the Czech Republic.

The authors also compared the PANGEA models, with and without bone marrow data, to the IMWG 2/20/20 risk stratification tool.

The PANGEA model, without incorporation of bone marrow data, improved prediction of progression from SMM to MM compared with the 2/20/20 model, with a C-statistic increase from 0.534 to 0.692 at the first patient visit in the first validation cohort. The PANGEA model also improved prediction of progression from MGUS to MM compared to the IMWG MGUS risk-stratification criteria, with a C-statistic increase from 0.670 to 0.879 for the PANGEA model without bone marrow data.

“Not only is the PANGEA model more accurate in predicting risk of progression than current gold standards, but it can be used repeatedly over time and uniquely provides individualized information for [patients with precursor disease to MM],” Ms. Cowan noted.

She added, “The PANGEA model greatly increases our clinical ability to predict disease progression, but we also now can provide good estimates of risk for patients who have not had a bone marrow biopsy.”

The study authors are developing a prospective cohort of patients with precursor disease to MM to further validate the results of their predictive tool. The team is also incorporating fluorescence in-situ hybridization cytogenetics, genomics, and RNA sequencing as a potential addition to the model that may influence patient progression risk.

Any conflicts of interest declared by the authors can be found in the original article.

Reference

Cowan A, Ferrari F, Freeman SS et al. Personalised progression prediction in patients with monoclonal gammopathy of undetermined significance or smouldering multiple myeloma (PANGEA): a retrospective, multicohort study. Lancet Haematol. 2023;10(3):e203-e212.