Improved Prognostic Assessment Of Patients With MDS By IPSS-R and MDS-CI

Myelodysplastic syndrome (MDS) is a broad spectrum of bone marrow failure syndromes characterized by diversity concerning clinical behavior, influence on patient’s outcome and risk for leukemic evolution. For more accurate prognostic assessment, the International Prognostic Scoring System (IPSS) has been recently revised. The purpose of this single-center study was to validate the r-IPSS and to compare this model with the MD Anderson Risk Model Score (MDAS), the IPSS and the (revised) WHO classification-based Prognostic Scoring System (WPSS(-R)). Furthermore, the predictive value of the MDS-Specific Co-morbidity Index (MDS-CI) as several other prognostic factors were analyzed.

Data were retrospectively collected from 222 MDS patients diagnosed at the VUmc between January 2000 and 2013. Selection was based on informative cytogenetics and the availability of peripheral blood and bone marrow counts. The MDAS, (r-)IPSS, (r-)WPSS and MDS-CI were applied and the impact of individual prognostic factors on patients’ outcome was investigated. Kaplan-Meier method was used for estimating survival and the value of prognostic models was determined by Cox’s multivariate regression method.

Of our study population, 25 (11%), 83 (38%), 51 (23%), 37 (17%) and 25 (11%) patients were classified according to the IPSS-R as very low, low, intermediate, high and very high risk with, respectively, median overall survival (OS) of 129 (95%CI 96-161), 89 (95%CI 73-105), 43 (95%CI 30-57-61), 31 (95%CI 20-42) and 19 (95%CI 7-31) months (P < 0.000). Compared to the MDAS, IPSS, WPSS(-R) and MDAS, the IPSS-R had a higher predictive power for OS. The MDS-CI was of significant predictive value additional to IPSS-R (P < 0.003). Low risk MDS-CI patients who underwent stem cell transplantation (SCT) had a significantly higher mean overall survival (OS), respectively 104 versus 43 months (P < 0.038). However, mean OS of intermediate and high risk MDS patients who received SCT was not better.

Our data confirm the predictive value of the IPSS-R and show the additional value of the MDS-CI. Combined application of these models refines the prognostic assessment and identifies co-morbidity among MDS patients, which may assist and influence clinical decision making.

No relevant conflicts of interest to declare.