Overcoming imatinib resistance

The remarkable success of imatinib in the treatment of chronic myelogenous leukemia (CML) is perhaps the most convincing example of effectiveness of a molecularly targeted therapy in a human malignancy. However, resistance to imatinib can be a problem. Although infrequent in patients treated in chronic phase, resistance eventually develops in the majority of patients treated in the advanced phase of the disease. Overexpression of the BCR/ABL gene and acquisition of BCR/ABL kinase domain mutations that interfere with imatinib binding are now recognized to be the most common mechanisms of resistance.

The observation of resistance to imatinib has led to an interest in therapies that may be effective in this situation. La Rosée and colleagues (page 208) now report that cells resistant to imatinib because of BCR/ABL overexpression or kinase domain mutations remained sensitive to treatment with arsenic trioxide and decitabine. Interestingly, the combination of these agents with imatinib resulted in enhanced inhibition but only where imatinib resistance was related to BCR/ABL overexpression or mutations that retained residual imatinib sensitivity. Synergistic growth inhibition required administration of imatinib at doses sufficient to achieve a threshold level of kinase inhibition resulting in apoptosis in these cells. Lower, less effective doses of imatinib could have an antagonistic effect. These results indicate that the use of imatinib-containing combinations may be a rational approach to treat resistance but applies only to mechanisms that are responsive to imatinib dose escalation and that escalation of imatinib dose to achieve effective kinase inhibition is essential.

Other ways to tackle imatinib resistance are also being explored. These include developing new BCR/ABL kinase inhibitors that are not only more potent but also retain activity against several imatinib-resistant kinase domain mutants; targeting mechanisms downstream of the BCR/ABL kinase; and several different approaches for reducing BCR/ABL levels, including the use of RNAi, and enhancing protein degradation by inhibition of heat shock protein 90 or administration of arsenic trioxide. La Rosée and colleagues make the additional observation that arsenic trioxide retained the ability to inhibit BCR/ABL expression in imatinib-resistant cells, and their studies suggest that inhibition of BCR/ABL expression in combination with escalated doses of imatinib may sufficiently suppress BCR/ABL activity to achieve growth inhibition.

This study represents a rational approach to tackling the problem of resistance to imatinib. It is likely that the clinical situation will be more complex since several different mechanisms of resistance may be active in polyclonal cell populations from CML patients. In addition to the mechanisms discussed in the preceding paragraph, other factors such as activation of signaling pathways independent of BCR/ABL activity, increased drug efflux, and altered drug binding may also contribute to resistance. Therefore, additional assays to directly assess inhibition of kinase activity and cell growth in an accurate and reproducible manner may be required to guide therapy. This notwithstanding, the work of La Rosée and colleagues emphasizes the potential importance of resistance testing in the design of clinical trials. The results of ongoing trials testing the effectiveness of such combinations in the treatment of imatinib-resistant CML will be eagerly awaited.