Correlation of Myelodysplasia-Related Flow Cytometric Findings with Mutations in Patients Undergoing Diagnostics for Suspected Myelodysplastic Syndrome

Introduction: Diagnostic procedures in patients with suspected myelodysplastic syndromes (MDS) have been broadened recently by adding flow cytometry and mutation analysis to cytomorphology and cytogenetics. In parallel the concept of pre-MDS conditions has been widely accepted in analogy to MGUS preceding multiple myeloma. Thus, idiopathic cytopenia of undetermined significance (ICUS) has been defined by cytopenia with minimal dysplasia in the absence of mutations and clonal cytopenia of undetermined significance (CCUS) is separated from ICUS by the presence of at least one MDS-related mutation. While this concept implies the outgrowth of a mutated cell clone and the acquisition of additional mutations as well as the gradual manifestation of the dysplastic phenotype as proven by cytomorphology and flow cytometry no comprehensive data is available yet on a multidisciplinary diagnostic approach taking into consideration ICUS and CCUS.

Aim: To describe the correlation of the presence of mutations with the occurrence of flow cytometric aberrations in the context of cytomorphologic and cytogenetic findings in patients with reactive conditions, ICUS, CCUS, MDS and related malignancies.

Patients and methods: In a retrospective analysis patients were identified in whom cytomorphology, cytogenetics, flow cytometry and mutation analysis had been performed in parallel for suspected MDS. Flow cytometric procedures followed ELN criteria. Mutation analysis was performed by next generation sequencing targeting the genes ASXL1, CALR, CBL, CSNK1A1, DNMT3A, ETV6, EZH2, IDH1, IDH2, JAK2, KIT, KRAS, MPL, NPM1, NRAS, RUNX1, SETBP1, SF3B1, SRSF2, TET2, TP53, U2AF1 and ZRSR2 . A total of 1607 patients were identified (median age 72 years, range 18-93; 1007 males, 600 females).

Results: Cytomorphologic evaluation revealed reactive conditions in 522 (33%) patients, ICUS was diagnosed in 248 (15%), MDS in 631 (39%), MDS/MPN in 63 (4%), CMML in 108 (7%) and 35 (2%) were not evaluable. An aberrant karyotype was found in 350 (22%) patients.

The numbers of flow cytometric aberrations were as follows: 0, 90 patients (6%); 1 to 2, 496 (31%); 3 to 4, 456 (28%); 5 to 7, 421 (26%); ≥8, 144 (9%). The myeloid progenitor cell compartment was not affected in 1292 (80%) and showed 1 aberration in 210 (13%), 2 in 86 (5%) and 3 to 4 in 19 (1%). The respective number for the granulocytic compartment were 0, 425 (26%); 1, 438 (27%); 2, 364 (23%); 3, 226 (14%) and 4 to 6, 154 (10%) and for the monocytic compartment 0, 451 (28%); 1, 513 (32%); 2, 307 (19%); 3, 161 (10%) and 4 to 7, 175 (11%). In 847 (53%) the erythroid compartment was not affected and in 760 (47%) 1 aberration was observed.

The numbers of mutations were as follows: 0, 739 patients (46%); 1, 347 (22%); 2, 295 (18%); 3, 157 (10%); 4 to 6, 69 (4%). The genes most frequently mutated were TET2 (342, 21%), ASXL1 (226, 14%), SRSF2 (215, 13%), DNMT3A (129, 8%), SF3B1 (133, 8%), RUNX1 (108, 7%), TP53 (89, 6%), and U2AF1 (83, 5%).

The number of flow cytometric aberrations was strongly correlated to the number of mutations (r=0.494, p<0.0001).

Mean (±SD) numbers of flow cytometric aberrations and mutations, respectively, were highest in CMML (6.2±2.3, 2.6±1.2), MDS/MPN (5.0±2.7, 1.9±1.3) and MDS (4.5±2.4, 1.3±1.2). In contrast, patients with reactive conditions had the lowest numbers (2.4±2.0, 0.3±0.7). Interestingly, numbers for ICUS cases were in between (3.7±2.3, 1.0±1.1). Importantly, re-classifying ICUS cases as CCUS based on the presence of at least one mutation (n=141) not only by definition increased the number of mutations (1.8±1.0) but also increased the number of flow cytometric aberrations (4.4±2.4) to levels observed in MDS while the remaining ICUS cases had levels observed in those with reactive conditions (2.9±1.7).

Mean (±SD) numbers of flow cytometric aberrations and mutations, respectively, in patients with and without cytogenetic aberrations were 4.7±2.5 vs. 3.7±2.4 and 1.4±1.2 vs. 1.0±1.2 (p<0.001 each).

Conclusions: This data confirms the concept of ICUS and CCUS preceding MDS and suggests to broaden this concept by integrating flow cytometric data showing strong correlation to observed mutations. Confirmatory analyses and follow up analyses correlating the findings of the different applied methods with the patients´ outcome should substantiate these results and build the basis for a clinically orientated classification of patients with cytopenia and MDS-related conditions.