Comprehensive Investigation of 225 Patients with Myeloid Malignancies and Erythroid Hyperplasia (≥50%) Demonstrates That Acute Erythroid Leukemia (AEL, according WHO Classification 2008) Differs Significantly From MDS but Overlaps with Other AML Subtypes and Pure AEL Regarding Clinical and Genetic Features.

According to the WHO 2008, acute erythroid leukemia (AEL) is defined by erythroid hyperplasia ≥50% and ≥20% of myeloblasts of non-erythroid cells but <20% of all nucleated cells. “Pure AEL” is defined by ≥80% of erythropoiesis without relevant myeloblasts. Cases with erythroid hyperplasia ≥50% and ≥20% of myeloblasts of all nucleated cells are classified as acute myeloid leukemia (AML-MRC/NOS), while presence of <20% of myeloblasts of non-erythroid cells assigns cases to myelodysplastic syndrome (MDS). As the separation of MDS/AML with erythroid hyperplasia in different categories is still under debate, we studied 225 patients with MDS/AML and ≥50% erythroid cells in bone marrow (BM) for cytomorphology, cyto-/molecular genetics, and prognosis.

The cohort consisted of 225 pts (f/78; m/147; median age, 68.8 yrs; 18.5–88.4 yrs) with BM erythroid hyperplasia (≥50%) and different myeloid subtypes strictly defined according to WHO 2008: MDS: n=107; AML-MRC/NOS: n=32; AEL: n=79, pure AEL: n=7 (the WHO cohort “AML with recurrent genetic abnormalities” was excluded). All pts were investigated by MGG staining of BM and chromosome banding/FISH. In addition, we performed analysis for the NPM1 mutations (n=126 investigated), FLT3-ITD (n=135), MLL-PTD (n=136), and NRAS mutations (n=90).

MDS subtypes were as follows: RA: n=18; RARS: n=18; RCMD: n=21; RCMD-RS (WHO 2001): n=26; RAEB-1: n=22; RAEB-2: n=2. Most AML pts were categorized as “AML with myelodysplasia related changes; AML-MRC” (27/30 cases; 90%); 3 pts were classified as “AML, not otherwise specified; AML-NOS”, 2 pts were not evaluable for this aspect. We first compared the MDS cohort (n=107) with the AML cohort (all 118 pts with AML-MRC/NOS, AEL, and pure AEL): Overall survival (OS) was better in MDS than in the AML cohort (median: not reached vs. 13.9 months; p<0.001). In contrast, OS showed no significant differences across the AML-NOS/MRC, AEL, and pure AEL subgroups (9.3 vs. 13.9 vs. 6.1 months; n.s.). In the total cohort, aberrant karyotypes (KTs) were detected in 105/225 pts (46.7%) and were associated with inferior median OS when compared to normal KTs (aberrant KTs: 12.5 months vs. normal KTs: not reached; p<0.001). Aberrant KTs were more frequent in the AML categories when compared to MDS (69/118; 58.5%; vs. 36/107; 33.6%; p<0.001), but showed no significant differences across the different AML subgroups: AML-MRC/NOS: 20/32; 62.5%; AEL: 44/79; 55.7%; pure AEL: 5/7; 71.4%; n.s.). Performing cytogenetic risk categorization according to revised MRC criteria (Grimwade, 2010) for the whole cohort, unfavorable KTs showed an inferior prognosis compared to intermediate KTs (unfav. KTs: 65/225; 28.9%; median OS: 7.6 months; vs. intermed. KTs: 160/225; 71.1%; not reached; p<0.001). The pts from the AML cohort more frequently had unfav. KTs than those with MDS (AML cohort: 50/118; 42.4% vs. MDS: 15/107; 14.0%; p<0.001). Unfav. KTs were similarly distributed in the AML cohort (AML-MRC/NOS: 14/32; 43.8%; AEL: 32/79; 40.5%; pure AEL: 4/7; 57.1). Regarding the molecular markers, we detected the NPM1mut in 25/126 investigated (19.8%; MDS: 0/43; AML cohort: 22/91; 24.2%), FLT3-ITD in 5/135 (3.7%; MDS: 0/43; AML cohort: 5/92; 5.4%), MLL-PTD in 12/136 (8.8%; MDS: 2/44; 4.5%; AML: 10/92; 10.9%), and NRASmut in 4/90 (4.4%; MDS: 1/42; 2.4%; AML: 3/48; 6.2%). Mutation frequencies did not differ significantly in the MDS vs. AML categories or across the AML-MRC/NOS, AEL, and pure AEL subgroups.

MDS with erythroid hyperplasia (≥50%) was clearly separated from the AML cohort (consisting of AML-MRC/NOS, AEL, and pure AEL, all with ≥50% of erythropoiesis) by less adverse cytogenetics and by improved survival. In contrast, no significant differences were observed across the different acute leukemia subentities regarding prognosis and cyto-/molecular genetic features. These data support the separation of MDS and AML with ≥50% of erythroid precursors according to the WHO classification. However, with respect to different AML subgroups, the separation to AEL, pure AEL, and AML-MRC/-NOS having ≥50% erythropoiesis seems arbitrary: these AML subtypes show no significant differences regarding prognosis or genetic risk profiles. This argues in favor of a combined group of AML with erythroid hyperplasia aiming to facilitate the definition for clinical studies and the development of therapeutic strategies.

Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership, Research Funding. Schnittger:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Kern:MLL Munich Leukemia Laboratory: Employment, Equity Ownership. Haferlach:MLL Munich Leukemia Laboratory: Employment, Equity Ownership.