Introduction

The term ICUS has been used to describe patients with persistent unexplained cytopenia(s) who do not meet the minimal diagnostic criteria for MDS. While studies on ICUS have been few and small to date, it is clear that a subset of patients do progress to overt MDS or acute myeloid leukemia (AML), supporting the concept of an early or pre-phase of MDS (Wimazal, Leuk Res 2007; Hanson, Leuk Res 2009; Schroeder, Ann Oncol 2010). Through the use of NGS-based profiling, the majority of patients with MDS are now known to harbor one or more somatic mutations in driver genes, with RNA splicing and epigenetic pathways most commonly altered (Haferlach, Leukemia 2014). The aim of this study is to elucidate the genomic landscape of ICUS and to evaluate its potential relationship to lower-risk MDS.

Methods

DNA exon sequences in SF3B1, SRSF2, U2AF1, ZRSR2, TET2, IDH1, IDH2, DNMT3A, EZH2, ASXL1, SETBP1, TP53, PHF6, RUNX1, ETV6, CBL1, NRAS, KIT, JAK2, MPL, and NPM1 were determined from the bone marrow aspirates of 250 patients with ICUS and 90 patients with lower-risk MDS (7 MDS with isolated del(5q), 7 MDS-unclassified, 12 refractory cytopenia with unilineage dysplasia, 27 refractory cytopenia with multilineage dysplasia, and 37 refractory anemia with ring sideroblasts) using a clinically validated and sensitive NGS assay on the Illumina MiSeq platform. The lower limit of detection of the assay was 5% with a minimum depth of coverage of 500x. Mutations were compared for concordance between duplicate samples and annotated using software that queried databases and published literature containing somatic mutations and germline variants.

Results

One or more somatic mutations were detected in 33% (82/250) of patients with ICUS. Among this mutated subgroup (ICUS-MUT), the mean number of mutations was 1.8 per patient. TET2 mutation, postulated to be an early genetic event in the pathogenesis of myeloid neoplasms, was the most common mutation detected (38%). DNMT3A (20%), ASXL1 (18%), SRSF2 (15%), and ZRSR2 (11%) mutations were the next most common, with the remaining mutations each detected in <10% of patients. The two most common pathways involved were epigenetic (65%) and RNA splicing (18%). The mean allele frequency for mutations was 33% (range 6-99%), with 14% of mutations having an allele frequency of <10%. The most common type of mutation was missense (60%), followed by frameshift (17%), nonsense (16%), splicing (4%), and other (3%). While no significant difference was observed in the median hemoglobin, absolute neutrophil count, and platelet count between the ICUS-MUT and wild-type (ICUS-WT) subgroups, the ICUS-MUT subgroup was significantly older (78 vs. 69 years, P < 0.0001) and more commonly male (male:female ratio of 1.3:1 vs. 0.8:1, P= 0.0139).

In lower-risk MDS, one or more somatic mutations were detected in 83% (75/90) of patients. The mean number of mutations was 1.8 per patient, identical to that observed in ICUS-MUT. SF3B1 mutation, which is highly associated with ring sideroblasts, was the most common mutation detected (47%). ASXL1 (16%), TET2 (14%), and DNMT3A (10%) mutations were the next most common, with the remaining mutations each detected in <10% of patients. The two most common pathways involved were RNA splicing (58%) and epigenetic (49%). The mean allele frequency for mutations was 32% (range 5-89%), with 6% of mutations having an allele frequency of <10%. The most common type of mutation was missense (66%), followed by frameshift (14%), nonsense (12%), splicing (7%), and other (1%). While no significant difference was observed in the median absolute neutrophil count, platelet count, age, and male:female ratio between lower-risk MDS and ICUS-MUT, lower-risk MDS patients were significantly more anemic (hemoglobin 9.7 vs. 10.5 g/dL, P= 0.0058).

Conclusions

Through the use of NGS-based profiling, one or more somatic mutations were detected in 33% of patients with ICUS. The ICUS-MUT subgroup displayed clinical parameters that were more similar to lower-risk MDS than the ICUS-WT subgroup. The ICUS-MUT subgroup was also similar to lower-risk MDS at the genomic level, with similar pathways involved, mean number of mutations, mean allele frequency, and type of mutations observed. Prior studies have shown that a subset of patients with ICUS do progress to overt MDS or AML. NGS-based profiling may be helpful in identifying those patients with potential early or pre-phase of MDS.

Disclosures

Kwok:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Reddy:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Lin:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Flamholz:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Yung:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Dabbas:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. McGinniss:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Nahas:Genoptix, Inc., a Novartis company: Employment, Equity Ownership. Kines:Genoptix, Inc., a Novartis company: Employment. Xu:Genoptix, Inc., a Novartis company: Employment, Equity Ownership.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution