Select patients with CML may safely discontinue dasatinib

The resounding success of tyrosine kinase inhibitors in the treatment of chronic myeloid leukemia has been somewhat tempered by the need for indefinite therapy. Indeed, relapse has typically been the rule when these agents are discontinued, and they are not generally considered to be curative. Although these drugs are typically quite tolerable — particularly when compared to historical alternatives, such as interferon or allogeneic hematopoietic stem cell transportation — they are immensely expensive and their side effects can be considerable on occasion. Hence, there is significant interest in identifying patients in whom therapy can be safely stopped.

Recent data has shown that imatinib (Gleevec, Novartis) can be discontinued in a subset of patients with sustained deep molecular responses. Imagawa and colleagues now report that dasatinib (Sprycel; Bristol-Myers Squibb, Otsuka Pharmaceutical) also can be stopped in carefully selected patients who had achieved deep molecular responses for more than 1 year on the drug. 

Nearly half of these patients maintained their deep molecular responses off of dasatinib, with a median follow-up 20 months. All relapses occurred within 7 months of cessation; and, fortunately, all of those patients regained deep molecular responses with resumption of a TKI.

Prior resistance to imatinib was highly correlated with relapse after discontinuation. Conversely, high natural killer cell counts, low gamma delta T cells, and low CD4-positive regulatory T cells all were associated with maintained remission off therapy. These data, while not demonstrating a causal link, do suggest the emergence of an antileukemic cellular immune response while on dasatinib, corroborating earlier findings. Whether such a response is sufficient to eradicate or at least provide long-term control of CML without the need for ongoing therapy remains to be seen. In fact, this immune effect seems to wane once off of dasatinib, and it is possible that late relapses could occur as a consequence.

Although these results are promising, they should be interpreted with caution. Just over half of patients with long-term deep molecular responses still relapsed after discontinuation of dasatinib. Although all of these patients again achieved deep molecular responses back on TKI therapy, this has not been the case for some patients who stopped imatinib. Longer follow-up and larger series of patients are needed before firm conclusions can be drawn.

Moreover, the majority of patients with CML still do not achieve sustained deep molecular responses, even with second-generation TKIs, and are thus not eligible for discontinuation of therapy at present. Perhaps newer, immune-based therapies, such as the checkpoint inhibitors, may facilitate durable treatment-free remission in more patients.

For now, discontinuation of TKI therapy is best restricted to patients in clinical trials, when possible.

Treatment-free remissions (TFR) in chronic myeloid leukemia has gained considerable attention as a means of alleviating the physical and financial side effects of long-term tyrosine kinase inhibitor therapy. Trials on imatinib (Gleevec, Novartis)-treated patients, mostly from Europe and Australia, consistently show 40% to 60% success, with differences between studies explicable by more or less stringent criteria for defining molecular relapse. It is important to remember that an increase of BCR-ABL1 in this setting is different from relapse on continued TKI therapy, and should probably better be referred to as recurrence

Srinivas Tantravahi, MD

Michael Deininger, MD, PhD

Whether TFR is safe and feasible in patients on dasatinib (Sprycel; Bristol-Myers Squibb, Otsuka) after imatinib failure had not been evaluated thus far. The DADI study led by Imagawa and colleagues addressed this question in patients with deep molecular response (DMR) for at least 12 months on second- or subsequent-line dasatinib.

DMR was defined as BCR-ABL1 less than 0.0069% IS (International Scale), intermediate between previously reported discontinuation studies. The researchers used stringent criteria to define molecular recurrence and the estimated TFR rate (49%) may have been higher if a loss of major molecular response had been used, as previously reported in the A-STIM study.

The rationale for the aggressive approach toward reinitiation of therapy was the perceived high risk of the study population. Indeed patients with resistance to prior imatinib therapy had a significantly lower rate of TFR compared with patients with imatinib intolerance, indicating that the inherent high-risk disease biology was unaltered by switching to the more potent second-generation TKI dasatinib. It is reassuring that all patients with recurrence recaptured DMR after restarting dasatinib, and no progression to advanced phase disease was observed.

Finally, similar to prior studies, high numbers of natural killer (NK) cell numbers (and low numbers of gamma delta-positive and regulatory T-cell cells) were predictive of TFR, suggesting that NK cell mediated tumor immunity may be required for continued suppression of dormant residual leukemia cells. Proving a causal relationship will require additional work, for example demonstrating that enhancing NK cell immunity would overcome the adverse prognostic impact in imatinib-resistant patients.

Molecular monitoring in a single central laboratory enabled careful and consistent patient selection and monitoring. The same high standards need to be applied if dasatinib discontinuation is considered outside of a clinical trial and dense molecular monitoring is important to identify recurrence as early as possible.  The high recurrence rate in patients with prior resistance to imatinib suggests that either 1 year of sustained DMR is insufficient, or that such patients may never be good TFR candidates. Both will be clarified in future clinical studies.

Reference:

Rousselot P, et al. J Clin Oncol. 2013;doi:10.1200/JCO.2012.48.5797.