Pure erythroid leukemia evolving from a therapy-related myelodysplastic syndrome secondary to treatment for chronic lymphocytic leukemia


  • Conflicts of interest: Nothing to report.

Correspondence to: Amir T. Fathi, Department of Hematology/Oncology, Massachusetts General Hospital, Harvard Medical School, Boston, MA. E-mail: afathi@partners.org

A 75-year-old male carried a diagnosis of chronic lymphocytic leukemia (CLL) for many years for which he received several courses of chemotherapy, in succession cladribine, cyclophosphamide, vincristine, and prednisone, and fludarabine and cyclophosphamide. Approximately 14 years after his CLL diagnosis, he developed progressive and refractory thrombocytopenia which did not respond to anti-CLL therapy. A bone marrow biopsy was performed and was markedly hypercellular and left shifted. There were several lymphoid aggregates consistent with CLL, but comprising only about 20% of the marrow cellularity. The hypercellular areas of hematopoiesis contained increased small, dysplastic megakaryocytes, and increased blasts (Image 1A), confirmed by CD34 immunohistochemistry (Image 1B). Cytogenetics revealed a t(7;14)(p11.2;q32) abnormality in five of 20 metaphases. Overall, these pathologic features were consistent with a therapy-related myelodysplastic syndrome. He was treated with 5-azacitidine, and a repeat bone marrow biopsy after two cycles of therapy revealed a decrease in the number of CD34+ blasts, indicating a modest response. Five months after his diagnosis of therapy-related myeloid neoplasm, due to worsening pancytopenia and persistent fevers, he underwent a bone marrow biopsy, which revealed evolution to pure erythroid leukemia (PEL) (Image 1C,D), with primitive, glycophorin positive erythroid blasts present in cohesive sheets (Image 1E) that were CD34-negative and comprised 80% of the cellularity. Also, present were aggregates of persistent PAX5-positive CLL cells (Image 1F). The bone marrow aspirate contained a preponderance of primitive erythroblasts, including binucleate and bizarre forms, in a background of small CLL cells (Image 1G). The cytogenetics obtained from this bone marrow aspirate revealed a highly complex, near-triploid karyotype (Image 1H). After rapid clinical deterioration, his goals of care were limited to palliation, and he expired 2 days later.

Image 1.

(A) The bone marrow is markedly hypercellular, with lymphoid aggregates consistent with CLL (lower center) and left-shifted erythroid and myeloid cells (hematoxylin and eosin). (B) CD34 immunostain reveals increased blasts, comprising 10–15% of the cellularity. (C) The bone marrow contains sheets of large blasts, with persistent CLL aggregates (upper right) (hematoxylin and eosin). (D) The blasts contain vesicular chromatin, multiple nucleoli, and smooth, amphophilic chromatin, consistent with erythroblasts; clusters of small, dysplastic megas (left center) are also present (hematoxylin and eosin). (E) By immunohistochemistry, the blasts are glycophorin positive, confirming erythroid lineage. (F) The background small lymphocytes are PAX5 positive, consistent with CLL. (G) The bone marrow aspirate contains numerous large erythroid blasts with deeply basophilic cytoplasm, frequent cytoplasmic vacuoles, and often irregular and bizarre nuclei (Wright-Giemsa). (H) Bone marrow cytogenetics revealed a complex abnormal: 57–64, XYY<3n>, +2, +3, −4, +6, −9, −10, −11, −12, −13, −17, −18, ass(18)(q23), +21, −22, +3–4mar[cp15]/46,XY[3].

PEL is a rare type of acute myeloid leukemia, which often arises from preceding myelodysplastic syndromes and may be therapy related in some cases [1-4]. Here, we present a highly unique and, to our knowledge, the first reported case of a PEL evolving out of a myeloid-neoplasm thought to be secondary to treatment for CLL. The large cells in the bone marrow in a background of CLL may have led to an erroneous diagnosis of Richter's transformation, but the cells had the immunophenotype of erythroblasts rather than large B cells.