Chronic Myeloid Leukemia: Reversing the Chronic Phase

Despite considerable scepticism about any possible clinical value of tyrosine kinase inhibitors (TKIs) in the early 1990s, imatinib at an oral dose of 400mg daily has now become standard initial treatment for all CML patients who present in Chronic Phase (CP). After 8 years follow up, the estimated survival for patients treated with imatinib is 85%, which is substantially better than patients treated with interferon alone or interferon plus cytarabine. The adverse effects of imatinib are definitely manageable in most instances.

Despite considerable scepticism about any possible clinical value of tyrosine kinase inhibitors (TKIs) in the early 1990s, imatinib at an oral dose of 400mg daily has now become standard initial treatment for all CML patients who present in Chronic Phase (CP). After 8 years follow up, the estimated survival for patients treated with imatinib is 85%, which is substantially better than patients treated with interferon alone or interferon plus cytarabine. The adverse effects of imatinib are definitely manageable in most instances.

However, the drug is not perfect. Only approximately 60% of patients are still taking imatinib at standard dosage after 6 years, which means that approximately 40% have needed higher doses of imatinib or alternative therapy. There are three newer TKIs, all of which are more active than imatinib in in vitro assays. Dasatinib (Bristol-Myers Squibb), nilotinib, (Novartis) and the third newest agent bosutinib (Wyeth), which is not yet licensed.

There are now well-defined criteria for imatinib failure; patients still in CP who are judged to have experienced treatment failure with imatinib whether as a consequence of intolerance or of resistance, are routinely offered treatment with either dasatinib or nilotinib. Clinical results are similar with the two agents; 40% to 50% of patients resistant to imatinib will be in complete cytogenetic response 2 years after starting their second-generation TKI.

This poses interesting clinical questions: should one or other of the so-called second-generation TKIs now replace imatinib as primary treatment for CML, and if so, which of the two agents should one choose?

In favour is the observation that the cytogenetic and molecular responses do seem to be much more rapid than those achievable with standard dose imatinib; this could be beneficial if the risk of disease progression is related to the quantity of residual disease in a patient’s body and the time that quantity is above a still ill-defined threshold. The adverse effects attributable to both agents seem to be relatively minor or nonexistent in most instances. One could reasonably speculate that the fraction of patients who currently start imatinib but develop resistance while still in CP and then respond to dasatinib or nilotinib would not have developed any TKI resistance if they had started the second generation TKI de novo, and indeed some of those with imatinib resistance who do not respond to second-generation TKIs might never have experienced treatment failure if they started other TKI drugs as initial treatment. These considerations might raise the overall success rate using a second-generation TKI as up-front therapy to 80% or more.

Conversely it could be argued that with 11 years experience of using imatinib, the safety profile is well established and 60% of patients do not need any stronger TKI. Do we really know that more rapid responses translate to superior survival, which must be the ultimate arbiter of whether to start treatment with imatinib or a second generation TKI? Will use of a second-generation increase the proportion of patients in whom we could safely stop therapy? Only time will tell. Finally in some countries considerations of cost also enter the equation.

If you do decide to use a second-generation TKI as primary therapy, which one would you choose? There seems little difference in efficacy. The adverse effect profiles are different; nilotinib is more likely to cause chemical disturbance of liver function and dasatinib more likely to cause pleural or pericardial effusions, but these problems all resolve if the relevant drug dosage is reduced or stopped, in which case one could switch to the other second-generation TKI. Kinase domain mutations may guide choice of therapy in patients resistant to imatinib, but they are exceedingly rare in newly diagnosed patients and thus contribute nothing to choice of initial therapy. Thus there is no good reason for preferring one or other of the newer TKIs.

An interesting compromise strategy would be to use two or three TKIs in sequence. One could, for example, start with imatinib, switch after 6 months to dasatinib (or nilotinib), and then switch again to nilotinib (or dasatinib). At least one such trial is currently in progress. It could prove superior to use of a single second-generation TKI, but this is far from certain. The problem here would be to define realistic endpoints. Studies based on survival would be the gold standard but would also take many years to complete. A more realistic end point might be the incidence of complete molecular response at 1 or 2 years.

The good news for today is the fact that survival for most patients presenting in CP has improved dramatically compared with 15 or 20 years ago. The challenge is to decide how exactly that survival can be improved further or –better still- how to ensure that therapy can safely be discontinued, as seems to be the case now in a small percentage of patients.

Professor John M. Goldman

Imperial College London


This article is a synopsis of the editorial written for the Journal of Clinical Oncology in Jan 2010. The full article including references can be found in the Journal of Clinical Oncology, Jan 2010: 363-365