Comment on Willis et al, page 2128
Leukemic clones with kinase domain mutation do not invariably expand when detected by cDNA sequencing and allele specific oligonucleotide (ASO)-PCR assay in imatinib-naive CML patients.
Imatinib, a selective inhibitor of the break-point cluster region–Abelson (BCR-ABL) tyrosine kinase, has dramatically changed the therapeutic approaches of all phases of chronic myelogenous leukemia (CML). Although impressive rates of cytogenetic responses were reported,1 a substantial number of patients in advanced phases of the disease and some in early chronic phase will develop resistance to the treatment. Point mutations within the ABL kinase domain have been reported as a common mechanism of resistance.2 Structural data indicate that some mutations, such as T31I or E255K, indicate poor prognosis, since they interfere with imatinib binding to the ABL kinase domain. Several investigators have found that mutation detection is significantly associated with disease progression and poor outcome.3,4 However, the prognostic significance of ABL mutation detection is less clear since new technologies with high sensitivity are now developed.
In this issue of Blood, Willis and colleagues explore the incidence and significance of kinase domain mutations using cDNA sequencing and allele-specific oligonucleotide–polymerase chain reaction (ASO-PCR) assay in imatinib-naive CML patients. Of 36 patients with different phases of the disease, 15 were positive for mutations detected in pretherapeutic samples. Not surprisingly, the frequency of the detection was high in accelerated and blastic phases of the disease. The surprise was the detection of mutation in sustained complete cytogenetic responders and that the mutant clones failed to expand. Expansion of leukemic cells is a complex mechanism that involves the leukemic cell itself as well as the immune system, the bone marrow microenvironment, and cell reactivity against various exogenous cytokines. The presence of BCR-ABL rearrangement in bone marrow cells and BCR-ABL kinase domain mutation may not be sufficient for the leukemic expansion. The critical issues are the hierarchic level of stem cells at which the mutation is detected and the cell-cycle status of these stem cells. Detection of kinase mutation in purified stem cells has recently been described5,6 ; in contrast to the observations of Willis et al, these patients' outcome was poor. Stem cell dormancy and the number of residual stem cells with mutation should also to be considered. Quiescent leukemic cells are insensitive to high-dose imatinib and could explain positive molecular disease.7 Such residual disease could persist for a long period of time without relapse, suggesting an important role for the immune system. In sum, CML seems to be a more complicated disorder with a variety of different leukemic stem cells.
There are 2 other important messages. Clonal cytogenetic evolution and prior exposure to 6-thioguanine were significantly associated with mutation detection, so design of new trials testing a combination of cytotoxic drug and imatinib should carefully take into account the increased risk of mutagenesis. A second bone marrow cytogenetic evaluation of patients treated with imatinib is still a useful biologic exam for the follow-up of patients. Finally, the detection of cell mutation should not be considered as a failure of treatment, since the technology used is highly sensitive. ▪
