Is Unidentified Cytopenia Truly Unidentified? Or Are We Overlooking Clonality?

Background and

Objectives

Peripheral cytopenia occurs in approximately 2% of the population; however, in up to 0.9%, no cause is identified by conventional tests. Next-Generation Sequencing (NGS) detects somatic variants consistent with clonal hematopoiesis (CH). We aimed to determine the prevalence of CH and pre-myelodysplastic syndrome (pre-MDS) using a 51-gene panel with histopathological assessment. Materials and

Methods

Bone marrow samples from 96 consecutive patients evaluated for cytopenia were retrospectively analyzed for genetic alterations.

Results

In the overall cohort (n = 96), the median follow-up was 8.1 months (range, 1-20). A total of 37 (39%) out of 96 patients were diagnosed with idiopathic cytopenia of undetermined significance (ICUS), 9 (9%) with clonal cytopenia of undetermined significance (CCUS), 9 (9%) with clonal cytopenia and monocytosis of undetermined significance (CCMUS), 34 (36%) with myelodysplastic syndrome (MDS), and 7 (7%) with myelodysplastic/myeloproliferative neoplasm (MDS/MPN). Among 41 patients in whom no cytogenetic abnormalities were detected by fluorescence in situ hybridization (FISH), somatic variants were identified by NGS. In CCUS, 88% of patients had a single variant, most commonly ASXL1 (44%), followed by TET2 (22%). In CCMUS, ASXL1 and DNMT3A (each 25%) were the most frequent variants. The mean variant allele frequency (VAF) was higher in MDS (33.4%) than in CCUS/CCMUS (13.6%). In MDS patients aged 60 years and older, a higher number of variants were found compared to patients aged less than 60 years (p = 0.028). RUNX1 variants (n = 8) were associated with leukopenia (p = 0.012). Patients with SRSF2 variants (n = 4) had significantly poorer progression-free survival (p = 0.001). EZH2 and SETBP1 variants were associated with inferior overall survival (p = 0.04 and p = 0.019, respectively). In MDS patients (n = 34), thrombocytopenia (plt < 100.000) was associated with shorter PFS (p = 0.005).

Conclusions

Given that pre-MDS conditions are considered predictors of hematologic malignancies, conventional diagnostic workup may be insufficient to accurately identify these entities, whereas NGS provides significant additional diagnostic value.