Long-Term Outcomes in Patients with Chronic Myeloid Leukemia in Chronic Phase Receiving Frontline Nilotinib Versus Imatinib: Enestnd 10-Year Analysis

Introduction: In Evaluating Nilotinib Efficacy and Safety in Clinical Trials-Newly Diagnosed Patients (ENESTnd), patients with CML-CP achieved deeper molecular responses and lower rates of progression to accelerated phase/blast crisis (AP/BC) with nilotinib (NIL) vs imatinib (IMA) during the first 5 y of follow-up. Here, we report long-term outcomes with ≥ 10 y of follow-up.

Methods: Patients with newly diagnosed CML-CP were randomized to receive NIL 300 mg twice daily (NIL300; n = 282), NIL 400 mg twice daily (NIL400; n = 281), or IMA 400 mg once daily (n = 283) in the ENESTnd core study. Some patients who discontinued their assigned core treatment due to suboptimal response/treatment failure entered an extension study to receive NIL 400 mg twice daily (for those assigned to NIL300 [n = 26] or IMA [n = 48] in the core study) or IMA 400 mg twice daily (for those assigned to NIL400 [n = 3]). In the current analysis, cumulative efficacy and safety results were analyzed by combining all data from the core and extension studies; extension study results were ascribed to each patient's assigned core study treatment arm. Progressions, deaths, and cardiovascular events (CVEs) were also analyzed according to time intervals (time since randomization for progressions; time since treatment initiation for deaths and CVEs), defined as events occurring up to 5 y and those occurring beyond 5 y. Analyses were based on a data cutoff date of April 1, 2019, when all patients had completed ≥ 10 y of treatment or discontinued.

Results: As previously reported, patient demographics were well balanced between arms. By the data cutoff, 105 (37.2%), 96 (34.2%), and 98 (34.6%) patients assigned to NIL300, NIL400, and IMA, respectively, completed the full study treatment duration; 175 (62.1%), 182 (64.8%), and 184 (65.0%) patients, respectively, discontinued core treatment early. Cumulative 10-y major molecular response (MMR) rates among patients assigned to NIL300, NIL400, and IMA were 82.6%, 80.4%, and 69.6%, respectively. The cumulative MMR rate reached a plateau at ≈ 3.5 y with NIL300 or NIL400 and 4.5 y with IMA. With NIL300, NIL400, and IMA, cumulative 10-y MR⁴ rates were 72.7%, 68.7%, and 55.5%, respectively, and cumulative MR^4.5 rates were 63.8%, 61.6%, and 45.2%, respectively. The difference between MR^4.5 rates achieved with NIL vs IMA by 5 y was similar to that observed by 10 y of study treatment (Figure).

Among patients assigned to NIL300, NIL400, and IMA, a total of 11, 7, and 24 patients, respectively, progressed to AP/BC by the data cutoff (including 1, 1, and 5, respectively, after 5 y) (Table). A total of 32, 23, and 29 deaths on study from any cause were reported in the NIL300, NIL400, and IMA arms, respectively; the most common causes were CML (6, 5, and 15 patients, respectively) and infections/infestations (5, 3, and 8, respectively). Of all deaths on study, 16, 14, and 11 in the NIL300, NIL400, and IMA arms, respectively, occurred after 5 y; almost all (37/41) were not CML related (including 10 due to general disorders, 9 due to infections/infestations, 8 due to neoplasms, and 7 due to cardiac or vascular disorders). Kaplan-Meier-estimated 10-y rates of event-free survival and overall survival (OS) are shown in the Table.

Adverse events (AEs) leading to discontinuation of study treatment occurred in 67 (24.0%), 98 (35.4%), and 56 (20.0%) patients assigned to NIL300, NIL400, and IMA, respectively; serious AEs occurred in 112 (40.1%), 127 (45.8%), and 97 (34.6%) patients, respectively. Safety profiles of NIL and IMA remained consistent with previous analyses (nonhematologic AEs in ≥ 30% of patients: NIL [either dose], rash, headache, nausea; IMA, diarrhea, nausea, muscle spasms). In each arm, CVE rates in the first 5 y of study treatment were similar to the rates beyond 5 y (Table).

Conclusions: NIL showed better long-term efficacy than IMA, with fewer progressions to AP/BC during both the early and later years of the study, and fewer CML-related deaths. MMR and deep molecular response rates were higher with NIL than with IMA throughout the follow-up period. However, CVEs were more frequent with NIL than with IMA in the first 5 y and continued to occur at similar rates beyond 5 y. While the study was not powered for OS, observed 10-year OS rates were similar between NIL and IMA. Overall, these results suggest that the superior efficacy but higher vascular toxicity observed early on with frontline NIL treatment is maintained with long-term treatment.

View largeDownload slide

View largeDownload slide

Close modal