Dasatinib is a multi-target kinase inhibitor that, based on in vitro studies, is more than 300 times more potent than imatinib(Drug information on imatinib) in inhibiting the BCR-ABL1 oncoprotein.
As with nilotinib(Drug information on nilotinib), most of the experience to date with dasatinib(Drug information on dasatinib) has been in patients who have failed imatinib. The results are similar to those achievable with nilotinib in this setting; about 40% of patients who fail imatinib may achieve worthwhile responses to dasatinib.[17,21,22] It is worth noting parenthetically that these responses to second-generation TKIs in patients who fail imatinib can be predicted with some reliability soon after starting the second-line treatment—or in some cases, even before this is started.[17,23,24]
Again, most of the information on first-line use comes from a single multi-center industry-sponsored study by the manufacturers, the Dasatinib vs Imatinib Study in Treatment-Naive CML-CP Patients (Dasision trial), in which 519 CML patients were randomly assigned to receive either dasatinib, 100 mg daily, or imatinib, 400 mg daily, as initial therapy. Briefly, dasatinib induced a higher rate of CCyR at 12 months (12 percentage points higher than imatinib). However the difference had disappeared by 24 months; this may have little practical consequence, though, since most physicians will consider that a patient who is not in CCyR by 12 months needs to change therapy. Like nilotinib, dasatinib induces molecular responses in a considerably higher proportion of patients than does imatinib.
Dasatinib is well tolerated; it has a lower incidence of insidious low-grade side effects than imatinib. Pleural effusions are the main complication of dasatinib therapy, but the reported incidence varies among the different series. Most studies report incidences below 20%, although cumulative incidences as high as 54% have been reported.[25-29] Because pleural effusions can occur some time after the start of treatment, the proportion of patients who will eventually be affected is not known. Pleural effusions are easy to manage. Diagnostic thoracocentesis is not usually required, and the effusion almost always resolves on discontinuation of the drug, without the need for any invasive procedure. The practice at the Hammersmith Hospital is to confirm the suspected diagnosis of pleural effusion with a chest x-ray, interrupt the dasatinib, and treat the patient with 0.5 mg/kg of prednisolone(Drug information on prednisolone) for 1 or 2 weeks. Pleural effusions may recur in spite of dose reduction.
Dasatinib is now approved for advanced-phase disease. In chronic-phase CML, a higher proportion of patients achieve CCyR by 1 year with dasatinib than with imatinib. Dasatinib is well tolerated (better than imatinib, if one excludes the pleural effusions). It is given once a day and can be taken with food or fasting (unlike imatinib and nilotinib). Conversely, dasatinib is associated with a relatively high incidence of pleural effusions. The side-effect profile is not yet fully understood, and dasatinib has been associated with pulmonary arterial hypertension. The incidence of this complication, although low, is not clearly established; nonetheless, it seems to be reversible on the discontinuation of the drug. Myelosuppression may be more of a problem than with imatinib or nilotinib, but the anemia may respond to administration of erythropoietin(Drug information on erythropoietin), and the neutropenia may respond to granulocyte colony=stimulating factor. As in the case of nilotinib, in most countries dasatinib is significantly more expensive than imatinib.
Bosutinib is a dual inhibitor of the SRC and ABL tyrosine kinases that is not currently approved for first-line use. A large phase III clinical trial, the Bosutinib Efficacy and safety in chronic myeloid LeukemiA (BELA) trial, has recently been completed, and the drug has very recently been approved by the US Food and Drug Administration (FDA) for second-line use. In the BELA study, 502 patients were randomly assigned to receive 500 mg daily of bosutinib or 400 mg daily of imatinib. At 1 year the cumulative incidence of the CCyR rate was similar for the two drugs, but bosutinib induced a higher proportion of molecular responses.
Bosutinib is well tolerated. Its main side effect is diarrhea, which in our experience can easily be managed symptomatically. However, diarrhea is frequent, even though it is usually self-limited or responds to simple measures. The posology of bosutinib is easy, and the drug is probably less toxic than imatinib.
Patients who start treatment with imatinib may fail at any time as a result of resistance or intolerance. Resistance may be primary, meaning that the patient never achieves any useful level of cytogenetic response, or it may be secondary, in which case the patient may have achieved a CCyR but then lost it despite continuing imatinib at the same dose. A small proportion of patients progress to advanced-phase disease after starting imatinib, but such progression is rare after the first 1 or 2 years of treatment. If one could reliably predict which patients are destined to become resistant to imatinib, one could make a good case for starting such patients on a second-generation TKI and starting all the remaining patients on imatinib at 400 mg daily.
A number of efforts have been made to predict response to imatinib. It is interesting to note that the Sokal score developed in the 1980s and based on patients treated with busulfan(Drug information on busulfan) or hydroxyurea still has some predictive value in the era of TKIs. Because imatinib is transported into cells by human organic cation transporter 1 (hOCT1), some investigators have studied hOCT1 expression by leukemia cells and concluded that low endogenous hOCT1 levels predict poorer responses to imatinib.[32-34] The notion that the speed of response to imatinib, measured either as the speed with which a patient achieves a given level of cytogenetic response or a given level of reduction of BCR-ABL1 transcripts in the peripheral blood, seems now to be emerging as the best single prognostic marker.
In 2003, the Liverpool group proposed that patients whose transcript number fails to fall to below 10% by 3 months have overall prognoses poorer than those with lower transcript numbers. We at the Hammersmith have recently reported the use of receiver operating characteristic analysis to study 282 patients who started treatment with imatinib. We showed that using a cut-off of 9.8% discriminated very significantly between patients with relatively good and those with relatively poor subsequent survival. The same technique—but with different cut-offs at 3 months—could be used to predict progression-free survival and indeed the achievement of a complete molecular response (Table 2). Others have reported similar findings using a 10% cut-off at 3 months. Thus it now seems that the speed in achieving transcript levels below 9.8% or 10% could be the single prognostic criterion necessary to identify the good imatinib responder.
Cure of CML
Investigators in France have recently reported the results of stopping imatinib in select patients. They collected data on 100 CML patients from throughout France who had started treatment with imatinib and had achieved complete molecular responses that had been maintained on treatment for a further 2 or more years. These patients then stopped the imatinib and were monitored closely at the molecular level. Relapse was defined as recurrence of detectable transcripts in the blood. The actuarial risk of relapse at 18 months was 61%, and almost all the patients who did relapse did so within the first 6 months of stopping therapy. The authors concluded that at least some of the patients who had not relapsed might prove to have been “cured” of their leukemia. This of course raises the question of how best to define cure, but for practical purposes one could say that a patient who survives for an appreciable number of years without detectable evidence of leukemia in his or her body has at least achieved an “operational” cure. Similar data have been published in a smaller number of patients from Australia. If some patients treated with imatinib can achieve an operational cure, the number treated with one or another of the second generation of TKIs might be greater, and such a finding could in due course become an important factor in determining which TKI is best for long-term treatment.
For the present, it seems impossible to make a firm recommendation that would apply to all new CML-CP patients scheduled to start therapy with a TKI. Instead there are two contrasting options.
1. The clinician could argue that he or she has considerable experience with imatinib and less experience with the newer agents. If 50% or more of patients do well with imatinib, would this not be an argument for starting all on imatinib and only changing to another agent if necessary? This argument gains traction if one can indeed identify early on those patients destined to fare badly. It gains further support from the expectation that the cost of imatinib may fall precipitously within the next few years. Also, one could perhaps increase the short-term efficacy of imatinib by increasing the starting dosage to 600 or even 800 mg daily or by combining it with interferon alfa.
2. Conversely, the clinician could argue that the second-generation TKIs undoubtedly act more rapidly to reduce the leukemia cell burden and may thus protect patients from progression in the first year or two after the start of treatment. They are also useful in managing some imatinib-refractory patients, and it might therefore be better to start all new patients on a second-generation TKI in the expectation that the overall incidence of resistance would be lower than with imatinib. It also seems likely that the cost of these agents will fall in the foreseeable future as the various TKIs become more readily available. Between nilotinib and dasatinib there is little reason to choose one over the other on the basis of differential efficacy. For an individual patient, the choice may depend on how he or she perceives the risks or harm associated with the side effects of the two agents.
It is quite likely that the choice between initial treatment with imatinib or with a second-generation TKI will become much clearer in the next few years.
Financial Disclosure: Dr. Goldman has received honoraria from Novartis, Bristol-Myers Squibb, Ariad, and Pfizer. Dr. Marin has no significant financial interest or other relationship with the manufacturers of any products or providers of any service mentioned in this article.