Genetic and Clinical Features of Chronic Myelmonocytic Leukemia with Fibrosis

Introduction

Chronic myelomonocytic leukemia (CMML) is an overlap syndrome with both myeloproliferative and myelodysplastic features. Clinical outcomes can be variable and risk stratification models such as GFM and Molecular Mayo Model (MMM) are useful. These models integrate age, WBC, anemia, thrombocytopenia, mutation status, monocyte count, and blast/promonocyte count, to segregate patients¹. The presence of fibrosis in MDS (MF 2-3) is often associated with high grade disease, poor cytogenetics, and worse prognosis ². The role of moderate to severe fibrosis in CMML (CMML-F) is not well studied. We investigated mutational landscape of CMML-F and whether CMML-F is associated with more aggressive disease thus warranting incorporation into risk models.

Methods

Total Cancer Care (TCC) and PathNet databases at Moffitt Cancer Center were queried for patients diagnosed with CMML between 2014 and 2017 with available Next Generation Sequencing (NGS) profiling (Genoptix 5-gene panel, Genoptix 21-gene panel, FoundationOne, Custom TrueSeq Myeloid). The cases were individually reviewed by a board-certified hematopathologist to confirm the diagnosis. The degree of reticulin fibrosis was manually collated from the pathology report and graded according to the World Health Organization (WHO) grading of bone marrow fibrosis (grade 0-3). Grade 1-2 or 2-3 fibrosis in the report were designated 1.5 and 2.5, respectively. CMML-F was defined as grade 2.5 or higher or collagen fibrosis. t-test and two tailed Fisher exact tests were performed for statistical analysis.

Results

Of 108 CMML patients (median age of 69.7 years), bone marrow fibrosis data was available for 91 individuals. The degree of fibrosis was as follows: Grade 0= 33, Grade 1= 34, Grade 1.5= 2, Grade 2= 15, Grade 2.5= 2, Grade 3=5 (of which 2 had collagen fibrosis). The CMML patients without fibrosis (MF<2.5) showed a longer median overall survival when compared to CMML-F (24.79 months versus 20.43 months; p=0.67), but it was not statistically significant. One of the 8 CMML-F patients had AML transformation (12.5%), similar to the 9 out of 82 CMML patients without fibrosis (11%). One of 2 patients (50%) with collagen fibrosis showed leukemic transformation, higher than the transformation rate in non-collagen fibrosis patients (11%; p=0.21). The most common mutations in CMML-F were: ASXL1 (25%), SRSF2 (25%), JAK2 (16.7%), and TET2 (16.7%). The most common mutations in non-CMML-F were: TET2 (60%), ASXL1 (45%), SRSF2 (38%), and RUNX1 (19%). Of note, TET2 mutations was less likely to occur in CMML-F (p=0.008). The average marrow blast percentage in CMML-F was 4.2% while in non-CMML-f was 8.6% (p=0.25).

Conclusions:

In this study we demonstrate that CMML-F is less likely to harbor TET2 mutations than CMML without fibrosis. However, unlike MDS, the presence of moderate-to-severe fibrosis does not correlate with worse prognosis in CMML. Large cohorts warranted to identify novel prognostic markers that could be incorporated into risk stratification schemas.