Improving the long-term outcome of patients with CML remains an important objective. Despite the excellent results achieved with imatinib, at least 25% to 30% of patients do not achieve an optimal outcome.1
In a recently reported single-institution analysis on 204 consecutive patients treated with standard-dose imatinib, the 5-year probability of EFS was 81%. However, considering intolerance to imatinib or lack of achievement of the desirable end points (eg, major cytogenetic response) as events, the 5-year EFS was 63%.16
To improve these results, we investigated dasatinib as initial therapy for CML. Dasatinib is more potent in vitro than imatinib,11
is effective in at least 50% of patients in whom imatinib has failed,9
and may be less prone to development of mutations than imatinib.17
In this study, dasatinib induced a high rate of CCyR (98%), with most responses occurring early. For example, by the sixth month of therapy, more than 90% of patients had already achieved a CCyR. These results compare favorably with the historical experience at our institution in similar patients treated with standard- or high-dose imatinib. Responses with dasatinib occurred considerably faster and ultimately at a higher rate compared with responses reported with standard-dose imatinib (45% at 6 months) and with high-dose imatinib (57% at 6 months).18
Still, considering that ultimately, the CCyR rate with standard-dose imatinib reaches 82%, the main advantage achieved with front-line dasatinib therapy might be the faster achievement of response. It is unclear what the long-term implications of these earlier responses may be. A subanalysis of the IRIS study suggested that the time of achievement of CCyR does not impact the probability of EFS.19
For this particular subanalysis, only patients who achieved a CCyR were considered. Because achieving CCyR is known to favorably affect long-term outcome, such analysis is affected by a selection bias of patients who are already known to have a favorable outcome. However, it is important to consider that a patient who has not achieved CCyR at any given time may either improve and eventually achieve CCyR or may not improve and eventually experience progression. A recent report has suggested that as more time evolves without achieving CCyR, the probability of eventually achieving CCyR decreases and the risk of progression increases.20
MMR was also achieved earlier, with 71% of patients achieving MMR at 12 months with dasatinib, compared with the reported rates of 46% with standard-dose imatinib and 54% with high-dose imatinib.18
These extrapolations should be considered with caution because the comparisons refer to historical controls. The inclusion criteria for all of these studies (standard- and high-dose imatinib and dasatinib) were identical, and there were no obvious differences in the patient characteristics. Still, the long-term impact of dasatinib or other second-generation tyrosine kinase inhibitors as initial therapies for CML will require prospective randomized trials. The results reported here are comparable to those achieved with nilotinib in a parallel study reported separately conducted in parallel to this study, with patients allocated to each study in alternating sequence.
Overall, treatment with dasatinib is well tolerated. The toxicity profile compares favorably with what has been reported with imatinib. For example, the rates of grade 3 or 4 anemia, neutropenia, and thrombocytopenia (6%, 21%, and 10%, respectively) are similar to what has been reported with standard-dose imatinib (4%, 25%, and 17%, respectively)21
and high-dose imatinib (5% to 10%, 14% to 36%, and 18% to 25%, respectively).5,22
Nonhematologic adverse events are generally mild and manageable. The most common nonhematologic adverse events (musculoskeletal pain and fatigue, 6% each) occurred at a similar frequency to what has been reported with high-dose imatinib (musculoskeletal pain, 3% to 7%; fatigue, 1% to 6%), but they are probably more frequent compared with standard-dose imatinib (fatigue, 1%; pain, 3%). One adverse event not previously noted is peripheral neuropathy manifested as numbness and tingling in hands (most frequently) or feet. These were usually grade 1 to 2, but three patients had grade 3 peripheral neuropathy requiring treatment interruptions and dose reductions, after which symptoms improved.
There was no significant difference in toxicity between the two schedules, except for a trend for less pleural effusions with the once-daily schedule. Results from a randomized trial suggested a decreased rate of adverse events, particularly myelosuppression and pleural effusion, with 100 mg once daily compared with 50 mg twice daily or higher doses.23
The overall rate of adverse events is lower in the front-line setting than after imatinib failure.9
This is not unexpected and is similar to what was reported with imatinib front-line therapy or after interferon failure.24,25
This lower rate of events would make it difficult to demonstrate a significant difference with the once-daily versus twice-daily schedules, particularly in a relatively small sample size. Because of the mature results of the previously mentioned large-scale randomized trial and the trends in favor of the once-daily schedule in this trial, we have discontinued random assignment to the twice-daily schedule. The study continues with the once-daily schedule only.
We conclude that dasatinib is effective as initial therapy for CML in chronic phase, with a high rate of response and rapid achievement of CCyR in nearly all patients by 6 months from the start of therapy. These results are achieved with a favorable toxicity profile. The possible superiority of this approach compared with standard therapy will need to be confirmed in randomized studies to evaluate a possible impact on EFS and, ultimately, overall survival. Results from such a study are expected shortly.