Evaluation of the Simplified Score to Predict Early Relapse in Multiple Myeloma (S-ERMM) in the MMRF CoMMpass study

Background: Zaccaria and colleagues recently proposed a new risk score to identify patients at high risk for relapse within 18 months of diagnosis (ER18), the Score for Early Relapse in Multiple Myeloma (S-ERMM). We performed external validation of the S-ERMM using data from the CoMMpass study. Patients and methods: Clinical data was obtained from the CoMMpass study. Patients were assigned S-ERMM risk scores and risk categories by the three iterations of the International Staging System (ISS): ISS, R-ISS and R2-ISS. Patients with missing data or early mortality in remission were excluded. Our primary endpoint was the relative predictive ability of the S-ERMM versus other risk scores for ER18 as assessed by area-under-the-curve (AUC). Results: 476 patients had adequate data to assign all four risk scores. 65%, 25% and 10% were low, intermediate and high risk by S-ERMM. 17% experienced ER18. All four risk scores stratified patients by risk for ER18. S-ERMM (AUC: 0.59 [95% CI 0.53–0.65]) was similar to R-ISS (0.63 [95% CI 0.58–0.69]) and statistically inferior to ISS (0.68 [95% CI 0.62–0.75]) and R2-ISS (0.66 [95% CI 0.61–0.72]) for prediction of ER18. Sensitivity analyses were performed and did not significantly impact results. Conclusion: The S-ERMM risk score is not superior to existing risk stratification systems for predicting early relapse in NDMM and further studies are needed to identify the optimal approach.