
Ethnicity-based cutoffs of laboratory parameters resulted in the best predictor model for risk prediction in Indian and American patients with multiple myeloma (MM), a new study has found.
Researchers in the study developed a new AI-enabled risk-staging system that used easily acquired laboratory and clinical parameters to stage patients and found that this new system – Consensus-based Risk Stratification System (CRSS) – better classified patients than the R-ISS.
In the study, the researchers compared baseline clinical and laboratory features of patients from two cohorts: the MM Indian dataset and the Multiple Myeloma Research Foundation dataset. Median values of all parameters other than albumin were significantly different in both groups, substantiating that the two populations are different.
They used seven parameters to design the CRSS: age, albumin, B2M, calcium, eGFR, and hemoglobin along with HRCA. Various models were built using different combinations of parameters using both established and proposed cutoffs for the two datasets. The best staging model was found for both datasets when the proposed cutoffs for the respective cohorts were used.
They then tested the CRSS against the R-ISS and found that the CRSS demonstrated superior performance as compared with the R-ISS in terms of C-index and hazard ratios in both the Indian dataset and the MMRF datasets.
An online calculator has been developed that can predict the risk stage of a patients with multiple myeloma based on the values of parameters and ethnicity.
Ethnicity-based cutoffs of laboratory parameters resulted in the best predictor model for risk prediction in Indian and American patients with multiple myeloma (MM), a new study has found.
Researchers in the study developed a new AI-enabled risk-staging system that used easily acquired laboratory and clinical parameters to stage patients and found that this new system – Consensus-based Risk Stratification System (CRSS) – better classified patients than the R-ISS.
In the study, the researchers compared baseline clinical and laboratory features of patients from two cohorts: the MM Indian dataset and the Multiple Myeloma Research Foundation dataset. Median values of all parameters other than albumin were significantly different in both groups, substantiating that the two populations are different.
They used seven parameters to design the CRSS: age, albumin, B2M, calcium, eGFR, and hemoglobin along with HRCA. Various models were built using different combinations of parameters using both established and proposed cutoffs for the two datasets. The best staging model was found for both datasets when the proposed cutoffs for the respective cohorts were used.
They then tested the CRSS against the R-ISS and found that the CRSS demonstrated superior performance as compared with the R-ISS in terms of C-index and hazard ratios in both the Indian dataset and the MMRF datasets.
An online calculator has been developed that can predict the risk stage of a patients with multiple myeloma based on the values of parameters and ethnicity.