https://orcid.org/0000-0002-6531-2625
Pure erythroid leukemia (PEL) represents a rare and highly aggressive subtype of acute myeloid leukemia, not otherwise specified (AML, NOS). PEL is presently the only type of acute erythroid leukemia recognized by the World Health Organization (WHO); erythroleukemia (erythroid/myeloid type) being eliminated as a diagnostic entity in 2016, with cases reclassified as either myelodysplastic syndrome with excess blasts (MDS-EB) or as AML with myelodysplasia-related changes (AML-MRC) [Citation1–3]. PEL is defined by the presence of immature erythroid precursors constituting more than 80% of bone marrow nucleated cells, with ≥30% proerythroblasts, and no evidence of a significant myeloblastic component [Citation1,Citation4]. The underlying genetic drivers of PEL remain nonspecific and poorly characterized, though cases will frequently have TP53 mutations and complex karyotypes [Citation3,Citation5,Citation6]. More recent studies of PEL have generally focused on the aspect of gene sequencing aberrancies, without chromosomal microarray (CMA) analysis [Citation3,Citation7,Citation8]. Studies utilizing CMA technology for characterization (array comparative genomic hybridization, array-CGH) are generally only found in pre-2016 cases of acute erythroid leukemia (no longer recognized by the WHO as a discrete entity), and many of these cases would presently be re-categorized as AML-MRC [Citation9,Citation10]. We assessed the utility of single nucleotide polymorphism (SNP)-based CMA in the evaluation of a series of well-defined PEL cases.
Five bone marrow biopsy cases meeting WHO revised 4th edition criteria for PEL were identified at Dartmouth-Hitchcock Medical Center between 2016 and 2020, and diagnoses were all confirmed by two board-certified staff hematopathologists. The cases included three women and two men, with a mean patient age of 68.6 years (range: 46–93 years). The average overall survival was 1.9 months (range: 0.3–4.8 months). Three of five cases were pancytopenic at diagnosis, while the remaining two had bicytopenia consisting of anemia and thrombocytopenia. All cases were markedly hypercellular averaging ∼95% (range: 80–100% cellularity), and two cases were densely myelofibrotic. Compositionally, all the cases demonstrated >80–90% erythroblasts and lacked any significantly expanded myeloblast population.
Genomic DNA was extracted from bone marrow using an automated EZ1 DNA extraction kit (Qiagen; Germantown, MD). Samples were analyzed using a Cytoscan™ HD Array (Thermo Fisher, Santa Clara, CA) according to manufacturer’s protocol, and clinically validated for constitutional testing. Data were analyzed using the Applied Biosystems Chromosome Analysis Suite Software (v4.1.0.90) (Thermo Fisher, Santa Clara, CA). All cases also underwent conventional chromosome analysis (karyotype) and gene sequencing analysis with the TruSight Myeloid Sequencing Panel (v3.0) (Illumina, San Diego, CA). An AML FISH panel was also run in two cases, which included dual-color dual-fusion probes for GATA2-MECOM, DEK-NUP214, RUNX1-RUNX1T1, PML-RARA, and BCR-ABL1 rearrangement; dual-color break-apart probes for KMT2A (11q23.3), RARA (17q21.2), and CBFB (16q22); and loci specific probes against EGR1 (5q31), D5S23 (5p15.2), centromere enumeration probe for chromosome 7 (D7Z1/7p11.1q11.1), D7S486 (7q31), centromere enumeration probe for chromosome 8 (D8Z2/8p11.1q11.1), and D20S108 (20q12) (Abbott Molecular, Des Plaines, IL).
By conventional chromosome analysis, two cases showed highly complex karyotypes, two cases had apparently normal karyotypes, and one case failed to grow in cell culture. NGS studies showed two cases carried two distinct TP53 mutations each, two cases carried one TP53 mutation each, and one case lacked TP53 mutations but had other variants compatible with clonal hematopoiesis. Multiple and some otherwise undetected aberrancies were demonstrated by CMA in all cases, including two cases with a normal karyotype and one with a cell culture failure (). Cases 2 and 5 both had highly complex karyotypes, and CMA demonstrated additional aberrancies not detected by conventional chromosome analysis. Regarding commonalities in microarray findings, four of the five cases showed alterations indicative of chromothripsis/instability of chromosome 19. Three of five cases showed recurrent losses in chromosome 12q. The one case which did not carry a TP53 sequence mutation was found to have copy-neutral loss of heterozygosity (cnLOH) affecting one third of the distal end of chromosome 1p and nearly all of chromosome 11q. cnLOH of 11q was seen in one other case. Losses in chromosomes 4, 5q (proximal), 7p, 7q, 9p21.3 (CDKN2A locus), 11p, 12p, 16p, 17, and X, were also noted to occur twice in our case subset ().
Figure 1. SNP array findings, graphical chromosome overview.
Table 1. PEL case cohort, summary of genetic findings.
Chromothripsis/instability of chromosome 19, evidenced by complex patterns of alternating copy number changes (normal, gain, or loss) along the length of chromosome or chromosome segment, was found in 80% of our cases. Chromothripsis is a catastrophic ‘chromosome shattering' phenomenon, characterized by hundreds of genomic rearrangements in an isolated chromosomal region, and rapidly accrued over a few cell divisions. Although pervasive in many human cancers, this seems to be an extremely rare occurrence in myeloid malignancies such as acute myeloid leukemia, myelodysplasia, and the myeloproliferative neoplasms [Citation11]. One historical study of acute erythroleukemia suggested 19q13.1 as a recurrently affected chromosomal breakpoint [Citation12]. However, recent studies of PEL (post-recognition as a new entity by the WHO in 2016) have not yet specifically appraised aberrancies in chromosome 19 as being potentially recurrent in in this disease [Citation3,Citation6–8].
Our literature search revealed three PEL case series published after 2016 with genetic data made available for their individual cases [Citation3,Citation6,Citation8]. From these reports, an additional 49 PEL cases are identified; 47 with karyotype data, five with TP53 FISH data, 15 with sequencing data, but none had been assessed by CMA (Supplemental Table 1). From the 47 cases with karyotyping data, 32 cases (68%) had reported chromosome 19 abnormalities; including structural aberrancies (n = 18), homogenously staining regions (hsr, n = 1), and/or copy number aberrancies (losses, n = 10; gains, n = 7). Additional chromosomal material of unknown origin was observed within the long arm of chromosome 19 (19q13 region) in eight of these cases, while additional material was also observed within the short arm of chromosome 19 (19p13 region) in 10 cases. The one case with a chromosome 19 hsr also involved the 19p13.1 region [Citation3]. Another study by Li et al. (not included in Supplemental Table 1; genetic results for individual cases was not made available) had 14 PEL cases evolving from chronic myeloid neoplasms, and reported that chromosome 19 was most commonly altered in their cohort (80% of cases, alterations not specified) [Citation5].
Beyond chromosome 19, an additional three of our five PEL cases showed recurrent losses in chromosome 12q. Several other aberrancies were seen twice, but the significance of these abnormalities is uncertain due to the small size of our cohort. Only one of the five cases did not carry a TP53 sequence alteration, and this case was found to have cnLOH affecting one-third of the distal end of chromosome 1p and nearly all of chromosome 11q. cnLOH of 11q was seen in one other case. When compared to the literature-derived cases (Supplemental Table 1), reported aberrancies resulting in loss of material from chromosome 12q or 11q were definitely identified in 19 cases, and potentially found in 16. A total of 43 cases (∼91%) had karyotypes with chromosomal material of unknown origin (marker chromosomes, ring chromosomes, etc.). The WHO recognizes losses of chromosomes 5/5q and 7/7q as being common [Citation1]. These were present in approximately 30% of cases in the literature and 40% of our cases.
While multiple copy number and/or loss-of-heterozygosity aberrancies identified by CMA in all PEL cases, three of five cases failed to demonstrate meaningful aberrancies by conventional chromosome analysis or FISH (). The cases with apparently normal karyotypes are thought to reflect growth of the background non-leukemic hematologic elements in culture, given that FISH studies for one were abnormal and all cases had aberrancies detected by CMA. In the two cases with abnormal karyotypes, the paired CMA showed that some of the apparent structural aberrancies did not result in an overall copy number change per cell, with the apparently lost material present in otherwise unidentifiable chromosomal material (ring and marker chromosomes, material of unknown origin, etc.). The complex abnormal CMA results found in our cases were anticipated, given that complex karyotypes are extremely common in this disease. However, abnormalities were found by CMA that would not be possible to detect and/or definitively characterize by using cytogenetic studies and gene sequencing alone.
The two cases that were also evaluated by an AML FISH panel showed probe signal gains at a variety of loci, with one instance of signal loss at the PML gene locus. None of the FISH signal aberrancies were duplicated between the two cases. A number of copy number aberrancies identified by FISH were not corroborated in the corresponding CMA. These discrepant calls included a FISH probe signal loss at the PML locus (15q24.1) in 11% of cells, and probe signal gains at RPN1 (3q21.3) in 16% of cells, MECOM (3q26.2) in 16% of cells, RUNX1T1 (8q21.3) in 7% of cells, RUNX1 (21q22) in 7% of cells, and PML (15q24.1) loci in 5.5% of cells. Regarding the copy number aberrancies identified by FISH but not corroborated by CMA; all abnormalities were found at a relatively low percentage of the evaluated interphase nuclei. These percentages were all very-near or below the limit of detection for our CMA, which is felt to be the reason for lack of corroboration by microarray.
Multiple groups have recommended CMA as a beneficial tool that complements cytogenetic and molecular studies in the comprehensive genetic profiling of hematologic cancers [Citation13,Citation14]. This seems to be especially true for PEL, where culture failure, chromosomal material of uncertain origin, unusual/cryptic findings, cnLOH, and other aberrancies not-well-resolved by karyotype and sequencing seem to be commonplace. This study and literature review additionally highlight that chromosome 19 aberrancies (specifically chromothripsis of Chr. 19), as a significant and under-recognized recurrent finding in PEL; especially in light of the otherwise extremely rare occurrence of chromothripsis in myeloid malignancies. PEL is an extremely rare entity, but future studies into this entity would likely benefit from comprehensive genomic profiling, including CMA studies.
GLAL-2022-0077-File004.xlsx
Download MS Excel (15.4 KB)Acknowledgements
The authors would like to acknowledge the Laboratory for Clinical Genomics and Advanced Technologies (CGAT) for its support on this project.
Disclosure statement
The authors report there are no competing interests to declare.
Data availability statement
All data generated or analyzed during this study are included in this published article, but may also be available from the corresponding author on reasonable request.
Additional information
Funding
References
- Swerdlow SH, Campo E, Harris NL, et al. World Health Organization classification of tumours of haematopoietic and lymphoid tissues. 4th ed. Lyon, France: International Agency for Research on Cancer; 2017.
- Wang SA, Patel KP, Pozdnyakova O, et al. Acute erythroid leukemia with <20% bone marrow blasts is clinically and biologically similar to myelodysplastic syndrome with excess blasts. Mod Pathol. 2016;29(10):1221–1231.
- Reinig EF, Greipp PT, Chiu A, et al. De novo pure erythroid leukemia: refining the clinicopathologic and cytogenetic characteristics of a rare entity. Mod Pathol. 2018;31(5):705–717.
- Arber DA, Orazi A, Hasserjian R, et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood. 2016;127(20):2391–2405.
- Li H, Hasserjian RP, Kroft SH, et al. Pure erythroid leukemia and erythroblastic sarcoma evolving from chronic myeloid neoplasms. Am J Clin Pathol. 2016;145(4):538–551.
- Park DC, Ozkaya N, Lovitch SB. Acute leukaemia with a pure erythroid phenotype: under-recognized morphological and cytogenetic signatures associated universally with primary refractory disease and a dismal clinical outcome. Histopathology. 2017;71(2):316–321.
- Wang W, Wang SA, Medeiros LJ, et al. Pure erythroid leukemia. Am J Hematol. 2017;92(3):292–296.
- Montalban-Bravo G, Benton CB, Wang SA, et al. More than 1 TP53 abnormality is a dominant characteristic of pure erythroid leukemia. Blood. 2017;129(18):2584–2587.
- Cervera N, Carbuccia N, Garnier S, et al. Molecular characterization of acute erythroid leukemia (M6-AML) using targeted next-generation sequencing. Leukemia. 2016;30(4):966–970.
- Grossmann V, Bacher U, Haferlach C, et al. Acute erythroid leukemia (AEL) can be separated into distinct prognostic subsets based on cytogenetic and molecular genetic characteristics. Leukemia. 2013;27(9):1940–1943.
- Cortés-Ciriano I, Lee JJK, Xi R, et al. Comprehensive analysis of chromothripsis in 2,658 human cancers using whole-genome sequencing. Nat Genet. 2020;52(3):331–341.
- Cigudosa JC, Odero MD, Calasanz MJ, et al. De novo erythroleukemia chromosome features include multiple rearrangements, with special involvement of chromosomes 11 and 19. Genes Chromosomes Cancer. 2003;36(4):406–412.
- Mukherjee S, Sathanoori M, Ma Z, et al. Addition of chromosomal microarray and next generation sequencing to FISH and classical cytogenetics enhances genomic profiling of myeloid malignancies. Cancer Genet. 2017;216–217:128–141.
- Peterson JF, Van Dyke DL, Hoppman NL, et al. The utilization of chromosomal microarray technologies for hematologic neoplasms: an ACLPS critical review. Am J Clin Pathol. 2018;150(5):375–384.