Inter-modality variance of blast quantification in patients with myelodysplastic neoplasms (MDS) and its impact on risk stratification and overall survival

Abstract

In the revised International Prognostic Scoring System (IPSS-R), blast enumeration by bone marrow cytology (BM-c) is crucial for risk stratification in myelodysplastic neoplasms (MDS), in the IPSS-Molecular (IPSS-M) however, incorporation of molecular data gained more prognostic impact. Blast count differences in the commonly used methods BM-c, flow cytometry (BM-f) and histology (BM-h) could possibly impact categorisation and overall survival (OS). In n = 145 investigated cases treated since 2012 in our institution, discordance in IPSS-R blast categories was found in 62/145 (43 %) of cases with evaluable blast counts in ≥ 2 methods. Discordant cases scored by BM-c showed either no change (24 %), a downgrade (48 %) or an upgrade (28 %) of their IPSS-R category when applying BM-f or BM-h. Discordant LR-MDS patients had significantly worse OS than concordant (72 vs 35 months, p = 0.031). In contrast, stratification by IPSS-M revealed no OS difference for discordant LR-MDS patients (p = 0.46). We could demonstrate that discordance occurred in almost half of patients, leading to re-stratification in a substantial amount of cases. Furthermore, OS was worse for discordant patients, but differences were ameliorated by inclusion of molecular data. This highlights the growing importance of molecular data over blast quantification in MDS.