WHO-Defined Chronic Myelomonocytic Leukemia-2 (CMML-2) Patients Rapidly Progress to AML Suggesting This Entity Represents a Transitory Clinical State

Introduction:

CMML is a heterogenous myeloid neoplasm with an estimated median overall survival (OS) of 31 months with 20% of patients transforming to acute myeloid leukemia (AML). It is subclassified by the WHO into CMML-0, CMML-1 and CMML-2 based upon bone marrow and peripheral blood blast percentage, with this distinction carrying variable prognostic relevance across published reports. CMML-2 is characterized by 10-19% bone marrow myeloblasts and/or 5-19% peripheral blood myeloblasts. However, whether this represents a distinct clinical entity or a transitional state inevitably leading to AML has not been explored.

Methods:

CMML patients with baseline disease classification and detailed treatment information were identified from the Moffitt Cancer Center CMML database. Baseline demographic, clinical and molecular information was evaluated. The Kaplan-Meier method was used to determine OS, progression free survival (PFS) and AML transit time, defined as the time from diagnosis to AML transformation. Treatment naïve patients with a baseline diagnosis of myelodysplastic syndrome with excess blasts-2 (MDS-EB2) were identified from an existing MDS clinical database as a comparator. Statistical analysis was performed using Graphpad Prism and SPSS.

Results:

Between 1/1/2000 and 7/1/2019, 623 patients with a pathologic diagnosis of CMML were identified. The median age was 71 and 68.4% were male. At the time of diagnosis, 356 patients were classified as CMML-0, 156 as CMML-1 and 111 as CMML-2. Median OS was calculated after stratifying patients by WHO category identifying 3 distinct prognostic groups with a median OS of 40 months in CMML-0, 22.4 months in CMML-1 and 14.0 months in CMML-2 (p<0.0001). The rate of transformation to AML was 18.5% in CMML-0, 26.9% in CMML-1 and 42.3% in CMML-2 (p<0.0001) with a median PFS of 37.2, 16.8 and 8.2 months, respectively (p<0.0001). CMML-2 patients transforming to AML had a lower baseline WBC (16.0 vs 27.0, p=0.03), lower baseline monocyte count (2.86 vs. 4.8, p=0.02) and increased transfusion dependency (63.8% vs. 45%, p=0.04). Comprehensive molecular data was available in a subset of patients with no mutations statistically different, although a trend toward a higher frequency of EZH2 (21.1% vs. 4.4%, p=0.15) and SETBP1 (11% vs. 0%, p=0.19) mutations was observed in patients that transformed. No differences in NPM1 mutation were observed in CMML-2 transforming cases (4 vs 3 cases). In patients with CMML-2 who progressed to AML the median time to transformation was 7.5 months (vs. 23.7 months in CMML-0 and 9.5 months in CMML-1, p<0.0001). However, because a large fraction of CMML-2 patients received hypomethylating agent (HMA) therapy, CMML-2 cases were further stratified into those who received HMA (n=61) vs. those who received no treatment (n=30) prior to AML transformation. While no OS advantage was identified with HMA treatment (mOS of 15.4 vs. 13 months, p=0.58), untreated patients displayed an extremely high rate of AML transformation (66.3% vs. 36.1%, p=0.02), a shorter PFS (2.8 vs 12.2 months, p<0.0001) and transit time to AML (3.4 vs. 11.6 months, p=0.002). Untreated patients who did not transform were older (median age of 72 vs. 66, p=0.06) with higher ECOG performance status (36% ECOG >2 vs 5.2%, p=0.08) and demonstrated a dismal median OS of 1.9 months. Untreated patients with CMML-2 were then compared to 85 patients with MDS-EB2 who did not receive therapy. While no significant difference in OS was identified, CMML-2 patients had a significantly higher rate of transformation to AML (66.3% vs. 22.3%, p<0.0001) and inferior PFS (2.8 vs. 5.1 months, p=0.0008).

Conclusion:

Subclassification of CMML by the WHO system based on bone marrow and peripheral blood blast content effectively stratifies patients into three groups with a distinct risk of AML transformation, PFS, and OS. Untreated CMML-2 patients display a rapid and extremely high rate of transformation supporting a transitional phase to AML. Further, the majority of untreated CMML-2 patients that did not pathologically transform had a particularly poor prognosis clinically consistent with aggressive AML. This is distinct from patients with MDS-EB2 in whom the risk of AML transformation and PFS were far lower. HMA effectively reduced the risk of AML transformation, PFS, and significantly delayed transit time to AML.