Myelodysplastic subclones in chronic myeloid leukemia: implications for imatinib mesylate therapy

Short-term results with imatinib mesylate (Gleevec; Novartis Pharmaceuticals, Hanover, NJ) for chronic-phase chronic myelogenous leukemia (CML) are extremely encouraging, with greater than 95% hematologic and 83% major cytogenetic responses in previously untreated patients.1 Long-term experience with interferon-α therapy suggests that only those patients achieving major cytogenetic responses experience significant survival benefits.2 Therefore, serial cytogenetic monitoring of imatinib mesylate–treated patients is imperative so that individuals not attaining major cytogenetic responses may be referred for potentially curative allogeneic marrow transplantation or investigative therapies before progressing to accelerated or blastic phases. Although repetitive bone marrow aspirations for cytogenetics are an effective technique to evaluate response, this method is time consuming, costly, and unpleasant for patients. Several investigators have reported that peripheral blood fluorescence in situ hybridization (FISH) for the BCR-ABL translocation, with its ease of procurement, sensitivity, and quantifiable nature may be a more convenient, yet accurate, method of serial monitoring.3,4The specificity of peripheral blood FISH, however, results in its major limitation; an inability to detect acquisition of additional cytogenetic abnormalities. We report a case where FISH could have led to false reassurances of successful imatinib mesylate treatment.

A 55-year-old man presented with leukocytosis and a hypercellular marrow without dysplasia or blast forms. Peripheral blood karyotype analysis revealed 46, XY,-5, del(6)(q23), t(9;22)(q34;q11.2), +mar and FISH for the BCR-ABL translocation was positive. Interferon therapy for one year yielded only a partial hematologic response. Imatinib mesylate resulted in a rapid complete hematologic response. Marrow FISH at 6 and 10 months demonstrated no evidence of the BCR-ABL translocation in 500 examined cells each time, suggesting a complete cytogenetic response. In contrast, marrow karyotyping yielded 46, XY, der(5)ins(5;5)(p15;q22q12)del(5)(q31q35), del(6)(q23) in 2 of 21 metaphases (6 months) and 7 of 20 metaphases (10 months), with the remaining cells normal. No Philadelphia chromosome–positive (Ph⁺) clones were identified.

Braziel et al5 recently reported a 14-month experience with imatinib mesylate therapy among interferon refractory/intolerant CML patients. Eighteen of 19 patients achieved a complete hematologic remission and 6 patients achieved a complete cytogenetic response. Two patients showed evidence of clonal evolution in clones possessing the Ph⁺, and 2 developed trisomy 8 independent of the Ph⁺ clone. The authors commented that they had observed trisomy 8 as the sole abnormality, independent of the Ph⁺clone, in 3 other patients enrolled on phase 1 trials with imatinib mesylate. Additional cases of cytogenetic abnormalities distinct from the Ph⁺ clone have rarely been reported following hydroxyurea6 and interferon7,8 therapies. The acquisition of these cytogenetic abnormalities may represent the changing natural history of CML, treatment-related effects (although doubtful since imatinib mesylate would not be expected to be leukemogenic), or a manifestation of an underlying stem-cell defect which produces both Ph⁺and Ph⁻ clones.

These observations of clonal cytogenetic evolution and cytogenetic polyclonality (ie, a separate non-Ph⁺ clone) have important implications for monitoring responses to imatinib mesylate therapy. It is possible that myelodysplastic subclones may be masked by a more dominant Ph⁺ clone and that selective inhibition of the Ph⁺ clone by imatinib mesylate may permit polyclonal emergence. Because peripheral blood FISH analysis is specific for theBCR-ABL translocation, these cases would go undetected if FISH alone were used to monitor long-term response. Thus, false reassurances of successful imatinib mesylate treatment might be drawn and patients not referred for alternate therapies. These cases support a role for monitoring algorithms that include periodic bone-marrow karyotyping of CML patients responding to imatinib mesylate therapy to exclude unrecognized coexisting myelodysplastic clones and/or clonal evolution in the absence of Ph⁺ cells.