Page 37 - 2021_12-Haematologica-web

cTCL1 and CD123 with isolated lymphoid markers (CD22/CD7/CD5) in some cases, suggesting a pre- pDC stage. In all cases, pDC, monocytes and cDC are neoplastic since they harbor the same mutations as CD34+ blasts. RUNX1 is the most commonly mutated gene: detected in all AML with minimal differ- entiation (M0-AML) but not in the other cases. Despite the low number of cases, the systematic associ- ation between M0-AML, RUNX1 mutations and an excess of pDC is puzzling. Further evaluation in a larger cohort is required to confirm RUNX1 mutations in pDC-AML with minimal differentiation and to investigate whether it represents a proliferation of blasts with macrophage and DC progenitor potential.
Frequent RUNX1 mutations in acute leukemia + pDC
Introduction
Plasmacytoid dendritic cells (pDC) are hematopoietic cells mainly developed from a myeloid branch including the macrophage DC progenitor (MDP) with monocyte, conventional DC (cDC) and pDC differentiation potential.1–4 Two types of neoplastic counterparts for pDC have been identified: the first is the well-known blastic pDC neoplasm (BPDCN), initially described as CD4+ CD56+ neoplasm;5-10 and the second is defined as mature pDC proliferation (MPDCP) associated with a myeloid neoplasm, frequently chronic myelomonocytic leukemia (CMML), but also myelodysplastic syndrome (MDS) or acute myeloid leukemia (AML), especially with monocyt- ic differentiation.11-17 MPDCP is not formally referenced in the World Health Organization 2017 classification, but mentioned as a differential diagnosis of BPDCN.7,8 As for BPDCN, MPDCP occur predominantly in male patients (75%) with a median age of 69 years18 and frequent lymph nodes or skin lesions. The mature pDC denomination refers to the morphologically mature and CD56– pheno- type (as with normal pDC), in the absence of the blastic morphology of BPDCN.8 Flow cytometry or immunohis- tochemistry are mandatory for BPDCN diagnosis and rel- atively well defined with CD4+, CD56+, CD123+high, CD303+/-, CD304+/- cells7,19 expressing TCL1 at high levels.20 Conversely, only few cases of MPDCP phenotype have been described.7,14,15 In the same way, the genomic profile of MPDCP is still poorly understood, but a clonal relationship between pDC and the associated neoplasm has been demonstrated with pDC exhibiting leukemic abnormalities such as monosomy 7, trisomy 8, del(5q), CBFB-MYH11 or internal tandem duplication of FLT3 (FLT3-ITD) of blasts in AML16,17,21–23 and the mutational profile of monocytes in CMML.24
We collected AML harboring a heterogeneous pheno- typic presentation with a population of immature blasts associated with an excess of pDC, but also monocytes and sometimes cDC. These cases are hereafter referred to as AML with pDC (pDC-AML). The purpose of this study was to better characterize pDC-AML by analyzing the phenotype of immature blasts and pDC and the mutation- al profiles of each cell population (blasts, pDC, mono- cytes, cDC) after cell sorting, in order to determine whether they share the same profiles.
Methods
Patient samples
Primary cells are routinely referred to our center for BPDCN suspicion with cytologic and phenotypic arguments in peripheral blood (PB) or bone marrow (BM) (collection DC 2016-27 91). Some cases do not meet the criteria for a BPDCN diagnosis, based on the World Health Organization classification 2017.5,8 Especially, 15 cases were collected based on the association of
CD34+ blasts of myeloid origin and pDC (Table 1). All cases were analyzed at diagnosis, except for N35 (day 81 post-induction). The analysis was performed on BM aspiration (n=12) or PB (n=3). BM biopsies were rarely available, preventing from anatomopatholo- gy analysis. Fifteen normal BM aspiration performed for peripher- al thrombopenia/research of metastatic infiltration, ten PB from healthy donors and 21 previously described cases of BPDCN20 were used as controls after written informed consent. This study was approved by the Besançon Ethics Committee (CPP-Est II, Besançon, France).20
Immunophenotype
Flow cytometry was performed using a FACSCanto II cytome- ter (BD Biosciences, San Jose, CA, USA) with DIVA 6.2 software (BD Biosciences) after cell labeling with monoclonal antibodies (Online Supplementary Table S1). The mean fluorescence intensity ratio (MFIR) of cTCL1 was obtained by dividing the mean fluores- cence intensity (MFI) for cTCL1 by that of the isotype control monoclonal antibody (mAb). Cells were considered positive for cTCL1 expression when MFIR was greater than 2. Isotype control was not used for MFIR of CD123 because of its high expression on pDC; thus MFIR was calculated by dividing the MFI of CD123 on pDC by that on lymphocytes.
Cell sorting
One to 10 million cells were sorted using an ARIA III FACS (Becton Dickinson Biosciences) after cell labeling with mAb (Online Supplementary Table S1) in order to select the populations of interest: T-cells (CD45+high, CD3+), immature blasts (CD34+, CD303–), pDC (CD123+high, CD303+), monocytes (CD14+ or CD64+, CD123+low, CD303–) and cDC (CD34-, CD1c+, CD303–).
Molecular biology
Whole genome amplification was carried out using the REPLI-g® Single Cell kit (Qiagen Hilden, Germany), as recom- mended by the manufacturer. Next-generation sequencing (NGS) was performed, from a HaloPlexHS Target Enrichment System (Agilent Technologies Inc., Santa Clara, CA, USA) targeting 70 genes (Online Supplementary Table S2), in paired-end, 2x150 cycles on a MiSeq platform (Illumina Inc., San Diego, CA, USA).
Bioinformatics analysis
Raw data were analyzed using in-house bioinformatics pipelines (Online Supplementary Methods) with annotation of the variants via GenerateReportsTM25 yielding variant call files. Finally, several filters were applied to eliminate intronic regions, synony- mous mutations and polymorphisms.
Statistical analyses
Statistical analyses were performed using Prism 6.0 software (GraphPad, San Diego, CA, USA). The distribution of MFIR was studied using the D’Agostino-Pearson normality test. Quantitative data were compared using ANOVA. Patient groups were com- pared using the Mann-Whitney non-parametric test for quantita- tive variables with a non-Gaussian distribution, the Student t-test
haematologica | 2021; 106(12)
3057