Novel Imatinib/Nilotinib Strategy Could Optimize CML Treatment

News
Article

A single-arm, open-label trial in Australia found that selective early switching from imatinib to nilotinib is feasible and effective in patients with CML.

Ball-and-stick models of imatinib and nilotinib

Ball-and-stick models of imatinib (top) and nilotinib (bottom)

A single-arm, open-label trial in Australia found that selective early switching from imatinib to nilotinib is feasible and effective in patients with chronic myeloid leukemia (CML).

“The TIDEL-II study aimed to optimize treatment outcomes by maximizing the number of patients reaching [European Leukemia Net] treatment milestones,” wrote study authors led by Timothy P. Hughes, of SA Pathology in Adelaide, Australia. The study built upon TIDEL-I, in which imatinib treatment was intensified based upon early targets.

In the new TIDEL-II study, two sequential cohorts totaling 210 patients with CML were enrolled. All patients in both cohorts began treatment with imatinib 600 mg/day. At 22 days imatinib could be intensified to 800 mg/day if plasma trough levels were below 1,000 ng/ml.

BCR-ABL1 levels (targets included ≤ 10%, ≤ 1%, and ≤ 0.1%) at 3, 6, and 12 months were then used to determine treatment plan. In cohort 1, if patients failed any target they were escalated to imatinib 800 mg/day, and then switched to nilotinib 400 mg twice daily if the same target was failed 3 months later. In cohort 2, failing to hit a target resulted in immediate switching to nilotinib; intolerance or loss of response in either cohort also resulted in switching to nilotinib.

At 12 months, 10% of cohort 1 and 12% of cohort 2 had a confirmed complete molecular response. At 24 months, these rates were 22% and 28%.

A total of 78 patients failed to reach TIDEL-II targets, the authors reported. Fourteen of them remained on imatinib therapy (13 on 800 mg/day), and 12 achieved major molecular response (MMR) at 24 months; 10 withdrew from the study with no further intervention. A total of 54 patients switched to nilotinib after a median of 7 months; 39% of those (21 patients) were in MMR at 24 months. Another 19 patients switched from imatinib to nilotinib after imatinib toxicity (but without failing any targets), and another 5 patients switched after a loss of response to imatinib between 8 and 23 months.

At the 2-year mark, 55% of all the patients remained on imatinib, and 30% remained on nilotinib. Only 12% of the full cohort were > 10% BCR-ABL1 at 3 months. The overall survival rate in the trial was 96%, and transformation-free survival was 95% at 3 years.

“TIDEL-II represents a novel and effective treatment option for the management of treatment-naive [chronic phase]-CML patients,” the authors concluded. It balances the fact that while some patients may need a second-generation tyrosine kinase inhibitor (TKI) such as nilotinib, using those drugs as first-line therapy can lead to long-term toxicity. “The strategy allows selection of patients who are less sensitive to or are intolerant to imatinib, and switching them to nilotinib in a time-dependent manner to minimize treatment failure.” They added that as generic imatinib becomes available, this may become an even more attractive approach given the high cost of some second-generation TKIs.

Recent Videos
CAR T-cell therapy initially developed for mantle cell lymphoma was subsequently assessed in marginal zone lymphoma.
The efficacy of the BOVen regimen in chronic lymphocytic leukemia facilitated its evaluation in patients with mantle cell lymphoma.
Increasing the use of patient-reported outcomes may ensure that practitioners can fully ascertain the impact of treatment for rare lymphomas.
Retrospective and real-world registry studies may be necessary to guide clinical decision-making for rarer lymphomas with insufficient prospective data.
Ongoing studies seek to evaluate immunotherapy in earlier lines of therapy for patients with early-stage Hodgkin lymphoma.
A paucity of prospective, well-vetted data to guide therapy in patients with rare lymphomas may result in a reliance on expert consensus guidelines.
5 experts in this video
5 experts in this video
Related Content