ALBERT

Description: Juvenile myelomonocytic leukemia (JMML) is a rare mixed myelodysplastic/myeloproliferative neoplasm seen in early childhood. JMML in general is a poor prognosis disease, including risk of transformation to acute leukemic blast crisis in approximately 15% of patients. Like adult chronic myeloid leukemia (CML) the blast phase of JMML is typically myeloid, however B cell and T cell transformations have been reported. Interestingly, in CML, a minority of patients will have a small aberrant B-lymphoblast population that does not inevitably herald progression to B-lymphoblastic blast crisis, however this has not been reported in JMML. Here, we report a case of a child with JMML found to have an aberrant population of precursor B lymphoid blasts. An 11-month-old boy presented to our clinic with splenomegaly and complete blood count (CBC) findings of a low platelet count of 69 x103/UL, low absolute neutrophil count of 0.61 X103/UL and elevated monocytes (36.3%), with an absolute monocyte count of 3.2 x 103/UL. Bone marrow aspirate and biopsy were consistent with JMML with no abnormal blast population reported at the time. Karyotyping confirmed the presence of monosomy 7, with concurrent fluorescent in-situ hybridization (FISH) evaluation showing monosomy 7 in 74% of cells examined. Next generation sequencing on the aspirate revealed a KRAS p.G12A activating mutation at an allele frequency of 35%, further validating the diagnosis of JMML. The patient did not have a matched related or unrelated donor available, thus he was followed closely with serial physical exams, blood counts and bone marrow evaluations. His splenomegaly resolved, blood counts remained stable to improved without intervention, but repeat bone marrow 6 months after the diagnosis showed similar frequencies of monosomy 7 and mutant KRAS. With still no suitable donor available, we opted to pursue a trial of hypomethylating agent, 5-azacitidine that has shown promising results in recent reports. A bone marrow evaluation was done prior to starting azacitidine to establish a pre-therapy baseline. Surprisingly, we were halted by detecting a B lymphoid blast population (6%), with characteristic flow findings, including positive CD45 (dim), CD19, CD10, CD20 (partial), CD22 (dim), CD34, HLA-DR, CD52, CD99, CD58 (heterogeneous) and CD38. Additionally, a new partial CDKN2A deletion was detected by FISH in 10% of cells. These findings were concerning for an emerging precursor B-acute lymphoblastic leukemia, possibly driven by the new CDKN2A deletion. However, we flow sorted the aberrant B cell population from the marrow aspirate, which showed the partial CDKN2A deletion in only 1.25% of these cells, unlike monosomy 7, which was seen in 100% of the abnormal B cell population. A retrospective review of the flow data from the marrow 4 months prior revealed the same aberrant B cell population in a similar percentage (5%) that was originally attributed to hematogones with low CD45 expression. The stability of this population over the course of months was inconsistent with an evolving lymphoid blast crisis. We observed the patient closely, and his blood counts and clinical exam remained stable over the following weeks. A repeat bone marrow showed that the abnormal B-cell population had decreased slightly to 2.6% and the partial CDKN2A deletion was now seen in only 0.4% of the examined cells. Thereafter, treatment with azacitidine (5 daily doses-100mg/m2/dose) was initiated with an excellent response after one cycle, with monosomy 7 detected in only 12% of cells and KRAS allelic burden down to 4.3%. Additionally, the aberrant B cell population declined to 0.15% and partial CDKN2A deletion was not detected by FISH. While longer follow up and additional patients will be necessary to reach definitive conclusions, our case suggests that the immunophenotypic detection of a small, stable abnormal B-lymphoblast population in patients with JMML does not necessarily herald impending acute leukemia. Disclosures No relevant conflicts of interest to declare.