Background: Myelodysplastic syndromes (MDS) comprise a heterogeneous group of diseases diagnosed and classified based on cytomorphology and cytogenetics according to the WHO classification. Flow cytometry and mutation analysis may provide additional diagnostic potential.

Aim: To correlate the diagnostic results derived from flow cytometry and mutation analysis with those of cytomorphology and cytogenetics in patients with suspected MDS. To estimate the impact of these findings on the cytomorphologic reevaluation during follow up.

Methods: Between February 2008 and July 2016 bone marrow samples from a total of 1681 patients with cytopenias and suspected MDS were prospectively analyzed by a combined diagnostic approach. This included in all cases cytomophology and cytochemistry, cytogenetics based on chromosome banding analysis supplemented by FISH analysis, flow cytometric assessment according to ELN criteria (Westers et al., Leukemia 2012) and mutation analysis for ASXL1, EZH2, RUNX1 and TP53which represent the prognostically most important molecular markers both in the pivotal study on molecular genetics in MDS (Bejar et al. NEJM 2011) and in a large multicenter study (Bejar et al., ASH 2015). Patients diagnosed with non-MDS hematologic malignancies were excluded. Patients´ age ranged from 17 to 95 years (median 72) and male:female ratio was 1.27.

Results: 816/1681 (49%) patients were diagnosed with MDS based on cytomorphology. An aberrant karyotype was found in 319/1681 (19%) patients. Flow cytometry was in agreement with MDS in 889/1681 (54%) patients. The number of patients with mutations in the respective genes were 193/1681 (12%) for ASXL1, 37 (2%) for EZH2, 84 (5%) for RUNX1 and 69 (4%) for TP53. At least one of these mutations was present in 318/1681 (19%) patients and one, two and three genes were mutated in 261 (16%), 49 (3%) and 8 (1%) patients, respectively.

Comparison between cytomorphology and flow cytometry revealed concordant results in 1300 (77%) patients (both positive for MDS in 667 (40%) and both negative for MDS in 633 (38%) patients). Cytomorphology diagnosed MDS while flow cytometry was negative (C+F-) in 149 (9%) cases and flow cytometry was in agreement with MDS while cytomorphology was negative (F+C-) in 232 (14%) cases.

Analyzing genetic results in these discordant cases revealed an aberrant karyotype in 34/149 (23%) of C+F- cases and in 30/232 (13%) of F+C- cases, respectively. At least one of the four analyzed genes was found mutated in 19/149 (13%) of C+F- cases and in 37/232 (15%) of F+C- cases, respectively. Combining these findings, an aberrant karyotype or at least one mutated gene were found in 45/149 (30%) of C+F- cases and in 55/232 (24%) of F+C- cases, respectively. In contrast, in cases rated MDS by both cytomorphology and flow cytometry (C+F+) an aberrant karyotype or at least one mutated gene were found in 354/667 (53%) cases while this was true for 61/633 (10%) C-F- cases only (p<0.001).

Follow-up analyses of bone marrow samples by cytomorphology were available for 116 cases initially not diagnosed with MDS by cytomorphology. 40 of them were initially rated in agreement with MDS by flow cytometry. Median follow-up time was 1.0 year. In 29 patients MDS was diagnosed by cytomorphology at follow-up. In the total of 116 patients with follow-up analyses the Kaplan-Meier estimate of probability of MDS was 40% at 2 years. Probability of MDS at 2 years was non-significantly higher in cases initially rated in agreement with MDS by flow cytometry as compared to others (48% vs. 35%). The respective impact of the presence of an aberrant karyotype or at least one mutated gene was even higher (2 year probability of MDS 71% vs. 23%, p<0.001). Combining flow cytometric and genetic results revealed the highest probability of MDS in case of positivity for both (F+G+, 81% at 2 years), followed by G+F- (65%), F+G- (29%) and F-G- (20%, p=0.002).

Conclusion: In patients with cytopenia not diagnosed with MDS by cytomorphology the presence of cytogenetic aberrations and molecular mutations typically associated with MDS reveals a high probability of development of MDS, particularly if in parallel flow cytometric evaluation is in agreement with MDS. Further study is warranted aiming at a respective extension of diagnostic criteria.

Disclosures

Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Meggendorfer:MLL Munich Leukemia Laboratory: Employment. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract

Sign in via your Institution