Incidentally‐detected t(9;22)(q34;q11)/BCR‐ABL1‐ positive clone developing into chronic phase chronic myeloid leukaemia after four years of dormancy

The World Health Organization (WHO) Classification (4th Edition) assigns chronic myeloid leukaemia (CML) to the myeloproliferative neoplasms, which are defined as clonal disorders with proliferation of one or more of the myeloid lineages. CML is characterized by t(9;22)(q34;q11) translocation and resultant BCR-ABL1 fusion gene found in all myeloid lineages as well as lymphoid cells and endothelial cells. We herein describe a case of incidentally-detected t(9;22) (q34;q11)/BCR-ABL1-positive clones developing into CML after a dormancy period. Our case highlights the significance of distinction between subclinical and definite CML. A 64-year-old woman was referred to us for systemic evaluation of refractory primary intraocular lymphoma (PIOL). She had received topical therapy with intravitreal methotrexate injections for the preceding 4 months. Imaging studies did not find any intracranial or distant lesions. The bone marrow (BM) was unremarkable without lymphoma infiltration (Fig 1A). Of note, routine G-banded karyotyping identified t(9;22)(q34;q11) in five of 20 metaphase cells analysed (Fig 1B). We considered the possibility of CML; however, the BM was normocellular without increase in neutrophils or neutrophil precursors. The myeloid:erythroid ratio was within a normal range and no fibrosis was detected. Thus, the BM lacked morphological features of CML. The patient did not complain of fatigue or weight loss and there was no evidence of splenomegaly. Laboratory findings were unremarkable, including white blood cell (WBC) count of 5 6 9 10/l with normal differential, haemoglobin level 126 g/l, and platelet count 274 9 10/l. Reverse transcription-nested polymerase chain reaction (RT-nested PCR) assay of peripheral blood (PB) detected e14/a2 (b3/a2) BCR-ABL1 mRNA. Real-time PCR using the LightCycler system (Roche Diagnostics, Mannheim, Germany) revealed BCR-ABL1 level (normalized to RNA18S1) of 4 00 9 10 . Regular fluorescence in situ hybridization (FISH) analysis for t(9;22)(q34; q11)/BCR-ABL1 using LSI-BCR/ABL dual colour ES probe (Vysis Inc., Downers Grove, IL, USA) did not detect BCRABL1 fusion signals either in BM mononuclear cells or in PB neutrophils. We then performed metaphase FISH analysis on BM, which revealed fusion signals in 13 (33%) of 40 metaphases scored. Thus, the discrepancy between the findings of G-banding and interphase FISH analysis was explained by the growth difference between the t(9;22)(q34;q11)/BCRABL1-positive cells and normal haematopoietic precursors in the ex vivo culture system. Despite positive cytogenetic and molecular findings, lack of myeloid lineage proliferation hampered the diagnosis of CML. Thus, systemic treatment for PIOL, comprising chemoimmunotherapy and whole brain radiotherapy was prioritized according to the protocol at our institution (Taoka et al, 2012). BCR-ABL1 mRNA was negative upon completion of the treatment (Fig 2). We employed a watchful waiting strategy for the t(9;22)(q34; q11)/BCR-ABL1-positive clones by monitoring BCR-ABL1 level using real-time PCR as previously described (Emig et al, 1999). RT-nested PCR for BCR-ABL1 mRNA became positive in six months and the BCR-ABL1 level gradually increased during the following three years. FISH analyses for t(9;22) (q34;q11) in PB neutrophils were repeatedly negative. BCRABL1 level eventually reached 2 78 9 10 , which was almost comparable to the mean level (5 87 9 10 ) of 30 newly diagnosed CML patients at our institution (Nannya et al, 2008). Laboratory results showed a slightly increased WBC count of 9 8 9 10/l with mild basophilia. The patient remained asymptomatic and without splenomegaly. A followup BM analysis showed a hypercellular marrow with proliferation of myeloid lineage cells (Fig 1C). G-banded karyotyping identified t(9;22)(q34;q11) in 17 of 20 metaphases, and FISH analysis of BM aspirate detected BCR-ABL1 fusion signals in 82% of interphase cells scored. The fusion signals were also present in 94% of PB neutrophils and 20% of PB mononuclear cells (Fig 1D). These findings confirmed the diagnosis of chronic phase CML, and treatment was initiated with dasatinib. After three months, G-banding and FISH analyses of BM indicated a normal karyotype, and real-time PCR analysis showed a 3 5-log decrease in PB BCR-ABL1 level. Incidentally-detected t(9;22)/BCR-ABL1-positive clones failing to meet the criteria for CML may be considered as subclinical CML, which can progress to definitive CML after a dormant period. Subclinical CML can be morphologically distinguished from definitive CML in that it lacks proliferation of myeloid lineages. Subclinical CML is therefore not classified as CML or even as a myeloproliferative neoplasm according to the WHO classification. In our case, previous intravitreal methotrexate injection therapy was probably irrelevant to the pathogenesis of subclinical CML, considering its correspondence