Home > Haematology > ASH 2021 > Acute Myeloid Leukaemia > Improved risk stratification in MDS via gene-based scoring system

Improved risk stratification in MDS via gene-based scoring system

Presented by
Dr Elsa Bernard, Memorial Sloan Kettering Cancer Center, NY, USA
ASH 2021
The newely developed molecular international prognosis scoring system for myelodysplastic syndromes (IPSS-M) delivers personalised, reproducible, and interpretable risk scores to stratify patients with MDS across 6 risk categories. The IPSS-M showed improved risk discrimination compared with the IPSS-R model and clear prognostic separation across risk categories [1].

“The current risk stratification methods for patients with MDS does not include gene mutations,” explained Dr Elsa Bernard (Memorial Sloan Kettering Cancer Center, NY, USA). “The IPSS-M was developed to improve risk stratification in MDS by considering clinical, cytogenetic, and genetic parameters.”

The research team collected samples of 3,675 participants with blast percentages <20% and white blood cell count <13x109/L, representative of all IPSS-R risk categories. Subsequently, the associations between genetic mutations and leukaemia-free survival, overall survival, and acute myeloid leukaemia (AML) transformation were assessed, demonstrating that 14, 16, and 15 gene mutations were linked to worse outcomes on the 3 endpoints, respectively. TP53 multi-hit, MLL partial tandem duplication, and FLT3 mutations showed the strongest associations with adverse outcomes. SF3B1 mutations were related to favourable outcomes, excluding SF3B1 mutations with specific concomitant other mutations.

The model used the weighted sum of 16 main effect genes and a genetic variable from 15 residual genes, the IPSS-R cytogenetic categories, and continuous clinical parameters (i.e. marrow blasts, platelets, haemoglobin) to calculate an individual risk score on a continuous scale. Furthermore, 6 risk categories, demonstrating prognostic separation across the relevant endpoints, were created.

The IPSS-M score outperformed the IPSS-R score regarding risk assessment: a 5-point increase in concordance index from IPSS-R to IPSS-M was reported for each endpoint. In total, 46% of the patients were re-stratified, and 7% of the patients were re-stratified by at least 2 strata. Furthermore, the new model was applicable across various settings, including low-blast-count AML and therapy-related MDS.

In conclusion, the IPSS-M model improved risk stratification in patients with MDS by assessing 31 gene mutations next to conventional parameters, delivering a highly personalised risk score.

  1. Bernard E, et al. Molecular International Prognosis Scoring System for Myelodysplastic Syndromes. Abstract 61, ASH 2021 Annual Meeting, 11–14 December.


Copyright ©2022 Medicom Medical Publishers

Posted on